AP3456 Mathematics & Physics PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 319

DO NOT DISTRIBUTE

AP 3456

Volume 8
MATHEMATICS AND PHYSICS

Table of Contents
PAR SEC CHA
TITLE
T
T
P
1
MATHEMATICS
1
Simple Arithmetic
1 Fractions and Decimals
2 Percentages and Proportions
3 Averages
4 Basic Vector Processes
2
Aids to Calculation
1 Indices and Logarithms
2 The Electronic Calculator
3 Graphs
4 Unit Conversions
3
Algebra
1 Principles and Rules
2 Equations
4
Geometry and Trigonometry
1 Plane Trigonometry
2 The Circle

Page 1 : Wed Jan 02 02:41:25 2002

Volume 8

8- 0- 0- 0

DO NOT DISTRIBUTE

AP 3456

2
1

Spherical Trigonometry
Introduction to Calculus
1 Functions and Limits
2 Differentiation
3 Integration
Introduction to Statistics
1 The Scope of Statistical
Method
2 Descriptive Statistics
3 Elementary Theory of
Probability
PHYSICS
Heat
1 The Nature of Heat
2 Temperature and Expansion
Electro Optics
1 The Nature of Light
2 Mirrors and Lenses
3 Infra-red Radiation
4 Lasers
Acoustics
1 The Nature of Sound
2 Acoustic Measurement
3 Oceanography
4 Sound in the Sea
Mechanics
1 Statics
2 Kinematics
3 Dynamics
4 Hydraulics
5 Introduction to Gryoscopes
Weapon Applications
1 Conventional Damage
Mechanisms
2 Guidance
3 Ballistics
4 Sighting
4A Annex to Chapter 4 Gyro-Gunsight

MATHEMATICS
Simple Arithmetic
Chapter 1 - Fractions and Decimals

Page 2 : Wed Jan 02 02:41:25 2002

Simple Arithmetic

8- 1- 1- 1

DO NOT DISTRIBUTE

AP 3456

Fractions
1. Description. Vulgar fractions are numerical quantities which are not whole numbers, expressed in terms of a numerator
divided by a denominator. There are two types: proper fractions which are less than 1 and improper fractions which are greater
than 1. Mixed numbers include whole numbers and vulgar fractions.
2. Comparing Two or More Fractions. To compare two or more fractions they must first be given the same denominator.
Example:
5 11

Arrange 7 , 14 and 4 in order of size.


The lowest common multiple of 7, 14 and 4 is 28.
5 4 20
=
7 4 28

11 2 22
=
14 2 28

3 7 21
=
4 7 28
In order of size
5 3 11
< <
7 4 14
3. Reducing a Fraction to the Lowest Terms. Reducing a fraction to the lowest terms, or simplest form, means finding the
equivalent fraction with the smallest possible numerator and denominator.
Examples:
12 2
=
18 3
after dividing numerator and denominator by 6
20 4
=
35 7
after dividing numerator and denominator by 5.
4. Addition and Subtraction of Fractions. If the denominators of the fractions are the same the numerators may simply be
added or subtracted.
Example:
3 7 1 3 + 7 + 1 11
1
+ + =
=
=2
5 5 5
5
5
5
If the denominators are different it is necessary to find the lowest common multiple so that the fractions may be rewritten with

Page 3 : Wed Jan 02 02:41:25 2002

Simple Arithmetic

8- 1- 1- 1

DO NOT DISTRIBUTE

AP 3456

the same denominator.


Example: 23 + 45 + 46. The lowest common multiple is 30 and the fractions may be expressed as
20 24 20 64
2
+ +
=
=2
30 30 30 30
15
5. Multiplication of Fractions. Fractions may be multiplied by first multiplying the numerators together and then multiplying
the denominators.
Example:
10 9 10 9
90
6
=
=
=
15 7 15 7 105 7
6. Division of Fractions. To divide one fraction by another, the divisor should be inverted and the fractions then multiplied.
Example:
3 5 3 7 21
1
= =
=1
4 7 4 5 20
20

Decimals
7. Description. Decimals are fractions in which the denominators are powers of 10. Decimals are written using a decimal
point instead of in the fraction form.
8. Changing Fractions to Decimals. A fraction may be converted to a decimal by dividing the numerator by the denominator.
Example:
7
= 7 8 = 0.875
8
It is also possible to convert a fraction to a decimal by expressing the denominator as a power of 10.
Example: By multiplying numerator and denominator by 4
52
13
=
25 100
and so
52
= 0.52
100
9. Changing Decimals to Fractions in their Lowest Terms. To change a decimal to a fraction the decimal should be written as
a numerator with a denominator of a suitable power of 10.
Example:
Express 0.68 as a fraction.

Page 4 : Wed Jan 02 02:41:26 2002

Simple Arithmetic

8- 1- 1- 1

DO NOT DISTRIBUTE

AP 3456

0:68 =

68
17
=
100 25

10. Addition and Subtraction of Decimals. When adding or subtracting decimals it is essential to ensure that the decimal points
are in line.
Example :
212.2 + 14.9 + 6.3 + 0.36

212.20
+14.90
+ 6.30
+ 0.36
233.76
11. Multiplying Decimals By Powers of 10. When multiplying by powers of 10 the decimal point is moved one place to the
right for each power of 10.
Example:
0.7 10 = 7.0
0.7 100 = 70.0
0.7 10000 = 7000.0
12. General Multiplication of Decimals. To multiply decimals the numbers should be multiplied together and then the number
of decimal places should be counted and the point set accordingly.
Examples:
0.9 0.007 = 0.0063
(decimal places 1 + 3 = 4)

2.652 0.04 = 0.10608


(decimal places 3 + 2 = 5)

410 0.12 = 49.20


(decimal places 0 + 2 = 2)
13. Division of Decimals by Powers of 10. When dividing decimals by powers of 10 the point should be moved one place to
the left for each multiple of 10.
Example:
0.7 10 = 0.07

Page 5 : Wed Jan 02 02:41:26 2002

Simple Arithmetic

8- 1- 1- 1

DO NOT DISTRIBUTE

AP 3456

0.7 100 = 0.007


0.7 10,000 = 0.00007
14. General Division of Decimals. To divide decimals both numbers should be multiplied by whatever power of 10 is required
to convert the denominator into a whole number, then division may be carried out in the normal fashion.
Examples:
4.55 0.5, may be written as
4.55 10 45.5

=
= 9:1
0.5 10
5

42.6 0.03, may be written as


42.6 100 4260

=
= 1420
0.03 100
3

0.0272 0.4, may be written as


0.0272 10 0.272

=
= 0.068
0.4
10
4
15. Significant Figures. If a number is given as an approximation it may be rounded to a multiple of 10. Thus 76,282 may be
given as 76,000 which is accurate to the nearest thousand or to 2 significant figures, the 7 and the 6. To 3 significant figures it
would be 72,300 because 76282 is nearer to 76300 than to 76200. The general rule is to consider the next digit to the right of
the one to which 'significant figure' accuracy is required. If it is greater than 5, then the previous figure should be increased by
one, and the appropriate number of noughts appended. If it is less than 5, then the previous figure should stand, again with the
appropriate number of noughts added.
16. Decimal Places. Numbers are often rounded off or given correct to a certain number of decimal places depending on the
degree of accuracy required. A calculator may give pi as 3.141592654 which for most purposes will be given to 3 decimal
places and written as 3.142.
17. For some fractions the division never ends, but numbers - or a series of numbers - are repeated:
1
= 0.3333 ....
3

4
= 0.571428571428 ....etc
7
Such decimals are called recurring decimals. The repeating pattern can be shown by placing a dot over the first and last digits in
the recurring group:
1
= 0.3_
3

Page 6 : Wed Jan 02 02:41:26 2002

Simple Arithmetic

8- 1- 1- 1

DO NOT DISTRIBUTE

AP 3456

4
= 0:5_ 71428_
7

Simple Arithmetic
Chapter 2 - Percentages and Proportions
Definition of Percentage
1. Percent means per hundred. A percentage is a fraction with a denominator of 100. Thus 13 percent means 13 divided by
13
100 or 100 , and is written as 13% or 13pc.

Percentages as Fractions or Decimals


2.

To convert a percentage into a fraction or a decimal it should be divided by 100.

Examples:

42% expressed as a fraction =

42
21
=
or as a decimal 42%=0.42
100 50

261
1
79
or as a decimal =0.263
26 % expressed as a fraction = 3 =
100 300
3

9.8% expressed as a fraction =

98
49
9.8
=
=
or as a decimal 0.098
100 1000 500

Fractions of Decimals as Percentages


3.

To convert a fraction or decimal to a percentage it should be multiplied by 100.

Examples:
3
3
as a percentage = 100% = 60%
5
5

7
7
as a percentage = 100% = 87.5%
8
8

0.62 as a percentage = 0.62 100% = 62%

Finding a Percentage
4. To find a percentage of a given quantity the quantity should first be multiplied by the required percentage and then divided
by 100.
Example:
Find 36% of 180

Page 7 : Wed Jan 02 02:41:27 2002

Simple Arithmetic

8- 1- 1- 2

DO NOT DISTRIBUTE

AP 3456

180

36
= 64.8
100

Expressing one Quantity as a Percentage of Another


5.

To express one quantity as a percentage of another first express one as a fraction of the other and then multiply by 100.

Example:
Express 49 miles as a percentage of 392 miles.
49
100 = 12.5%
392

Percentage Increase or Decrease


6.

To increase or decrease an amount by a given percentage the amount should be multiplied by the new percentage.

Example:
Increase 650 by 6%

650

106
= 689
100

(or 650 1.06 = 689)


Decrease 650 by 6%

650

94
= 611
100

(or 650 0.94 = 611)


7. To find an original quantity given a quantity which has been increased or decreased by a percentage it is necessary to first
divide the quantity by the new percentage and then multiply by 100.
Example:
After an increase of 8% a quantity is 178, what was the original quantity?
The increased quantity is 108% of the original.
108% of original quantity is 178

1% of the original is

Page 8 : Wed Jan 02 02:41:27 2002

178
108

Simple Arithmetic

8- 1- 1- 2

DO NOT DISTRIBUTE

AP 3456

So 100% of original is

178
100 = 164.81
108

Ratios
8. A ratio enables the comparison of two or more quantities of the same kind and is calculated by dividing one quantity by the
other.
Example:

The ratio of 375 to 500 =

375 3
=
500 4

and is written as 3:4


To nd the ratio of 5 km to 700 m:

Ratio =

5000
= 50:7
700

9. Division in a Given Ratio. To divide a quantity according to a ratio 3:4:5, the quantity is first divided by 3+4+5, then 3
parts, 4 parts and 5 parts are allocated.
Example:
Divide 2400 in the ratio 3:4:5
2400
2400
=
= 200.
3+4+5
12

Sums are 600, 800 and 1000


10. Increasing and Decreasing in a Given Ratio
Example:
If fuel consumption of 60 kgs per minute is increased in a ratio of 5:4 what is the new consumption?

Consumption =

5
60 = 75 kgs per min.
4

Scales
11. If a map has a scale of 1:50,000 it means that 1 cm on the map represents 50,000 cm on the ground. In the same way 1 km
on the ground is represented by
100,000
cm or 2 cm.
50,000
The scale 1:50,000 could also be given as 2 cm to 1 km.

Page 9 : Wed Jan 02 02:41:27 2002

Simple Arithmetic

8- 1- 1- 2

DO NOT DISTRIBUTE

AP 3456

Proportion
12. If two quantities are in direct proportion then an increase in one quantity causes a predictable increase in the other. An
inversely proportional relationship means that an increase in one quantity causes a predictable decrease in the other.
Examples:
a. Direct Proportion. If 400 cards cost 28 find the cost of 650 cards.
Cost of 650 cards =

650
28 = 45.50 pounds
400

b. Inverse Proportion. If it takes 6 men 12 days to paint a hangar how long will it take 9 men?

Time for 9 men =

6
12 = 8 days
9

The 1 in 60 Rule
13. The 1 in 60 rule is used as a method of assessing track error and closing angle, and has long been favoured as a pilot
navigation technique because of its flexibility, ease of use and relative accuracy (ie up to about 20) - see Fig 1. The 1 in 60 rule
postulates that an arc of one unit at a radius of 60 units subtends an angle of one degree.

8-1-1-2 Fig 1 The 1 in 60 Rule

14. In practical use, the distance along track is compared with the distance off track and the ratio of one to the other is reduced
to an angle (see Fig 2).

Track error (degrees) =

Distance o Track 60
Distance along Track

Thus an aircraft passing over a feature 2 miles port of the required track after flying 30 miles has a track error
2
30 60 = 4 degrees.

8-1-1-2 Fig 2 Calculation of Track Error

Page 10 : Wed Jan 02 02:41:27 2002

Simple Arithmetic

8- 1- 1- 2

DO NOT DISTRIBUTE

AP 3456

Simple Arithmetic
Chapter 3 - Averages
Introduction
1. Averages are discussed in some detail in Section 6, Statistics, where it can be seen that an 'average' might mean any one of
three quite different values. Of these the most useful and most commonly used is more accurately described as the arithmetic
mean.

Arithmetic Mean
2.

The arithmetic mean of a set of values is defined as:


The sum of all the values
The number of values

Examples:
a. A rugby scrum has players of weights 92kg, 89kg, 86kg, 94kg, 97.5kg, 97kg, 96kg, 95.5kg. The arithmetic mean (or
average) weight of the players may be calculated as follows:
Average weight

92+89+86+94+97.5+97+96+95.5
8
= 93.375kg

b. The times taken to travel to work from Monday to Friday are 1 hr 12min, 1hr 18min, 1hr 14min, 1hr 21min, 1hr 22min.
The average time taken in travelling to work can be calculated as follows:
Average time

Page 11 : Wed Jan 02 02:41:28 2002

Simple Arithmetic

8- 1- 1- 3

DO NOT DISTRIBUTE

AP 3456

72+78+74+81+82
mins
5
= 1 hr 17.4 mins

3. The arithmetic mean is useful for presenting large amounts of data in a simplified form, and is most accurate when used in
calculations involving data which do not include extreme values. This form of average may also yield data which are capable of
further statistical analysis or mathematical treatment. It uses every value in a distribution, and is the most readily understood and
commonly accepted representation of the term 'average'.

Limitations in the Use of Arithmetic Mean


4.

The arithmetic mean may produce distortions because of extreme values in a distribution.

Example: The values of stamps to be auctioned are estimated at 15, 17, 23, 24, 20, 500. The average (arithmetic mean)
of their values is given by:
15+17+23+24+20+500
= 99.83 pounds
6
However, it would clearly be misleading to describe the stamps as being of average value of approximately 100. A more
accurate and fair description would be that with one exception the average value of the stamps is approximately 20.
5. The arithmetic mean can also produce impossible quantities where data is necessarily in discrete values, (eg 1.825 children
in an average family).

Weighted Averages
6. When calculating an average from more than one set of data, the figures cannot be combined without giving due regard to
the relative sizes of the samples.
Example: A class of 40 students score an average of 60 marks and a class of 20 students score an average 68 marks. The
average might be calculated as:
60+68
= 64
2
but this is clearly incorrect. The marks should be weighted according to the number of students in each group thus:
60(40)+68(20)
40+20

2400+1360
60

= 62.66
This is termed the weighted average, and it gives a more accurate measure in this type of situation. Weighted averages may also
be used when it is desired to give certain quantities greater importance than others within a distribution.

Page 12 : Wed Jan 02 02:41:28 2002

Simple Arithmetic

8- 1- 1- 3

DO NOT DISTRIBUTE

AP 3456

Simple Arithmetic
Chapter 4 - Basic Vector Processes
Introduction
1. Many physical quantities like mass, volume, density, temperature, work and heat, are completely specified by their
magnitudes. Such quantities are known as scalar quantities or scalars. Other physical quantities possess directional properties as
well as magnitudes, so that each magnitude must be associated with a definite direction in space before the physical quantity can
be completely described. It is found that some, though not all, of these directed quantities possess a further common property in
that they obey the same triangle (or parallelogram) law of addition. Directed quantities which obey the triangular law of
addition are known as vectors.
2. Definition of a Vector. Any quantity which possesses both magnitude and direction, and which obeys the triangle law of
addition is a vector.

Graphical Representation of Vectors


3.

A Vector quantity may be represented by a line having:


a. Direction.
b. Magnitude.
c. Sense.

Fig 1 shows a vector having a direction defined by the angle , a magnitude equal to the length OP and a sense indicated by the
arrow. The vector OP may be represented by a symbol which may be either in bold type or underlined, eg OP may be
represented by a. When a vector is shown graphically a scale should be given.

8-1-1-4 Fig 1 Graphical Representation of a Vector

8-1-1-4 Fig 2 Resolution of a Vector into Components

Page 13 : Wed Jan 02 02:41:28 2002

Simple Arithmetic

8- 1- 1- 4

DO NOT DISTRIBUTE

AP 3456

The Resultant Vector


4.

The resultant of a system of vectors is that single vector which would have the same effect as the system of vectors.

The Resolution of Vectors into Components


5. It may be convenient to resolve a vector into two components acting at right angles to each other. In Fig 2 vector OP is at an
angle of 30 to Ox. The vector may be divided into two components OA and OB at right angles to each other. Resolved
graphically the component OA may be measured as 6.9 units and OB as 4 units. To calculate the magnitudes of OA and OB
mathematically use is made of the trigonometrical ratio of OA to OP, thus:
OA
= cos 30
OP

. _ . OA = OP cos 30 = 6.93 units


and
OB AP
=
= sin 30
OP OP

. _ . OB = OP sin 30 = 4 units

Addition of Vectors
6. Co-linear Vectors. The simplest case of the addition of vectors occurs when the vectors are parallel. There are two cases to
consider:
a. Parallel Vectors Acting in the Same Direction. Consider two forces acting on a body, one of magnitude 4 units and the
other of magnitude 3 units. The two forces act in the same direction. Fig 3 shows the two vectors, a of four units and b of
three units. The sum of the vectors is a + b = 4 + 3 = 7 units.

Page 14 : Wed Jan 02 02:41:28 2002

Simple Arithmetic

8- 1- 1- 4

DO NOT DISTRIBUTE

AP 3456

b. Parallel Vectors Acting in Opposite Directions. When two forces acting on a body are parallel and in opposite
directions the vector representation is as Fig 4. The forces are of magnitude seven units and three units. The sum of the
vectors is a b = 7 3 = 4 units.

8-1-1-4 Fig 3 Addition of Parallel Vectors Acting in the Same Direction

8-1-1-4 Fig 4 Addition of Parallel Vectors Acting in Opposite Directions

7. Non Co-linear Vectors. Any two vectors may be added together. Fig 5 shows a triangle of vectors. A displacement from O
to A is represented by vector a, and a further displacement from A to B is represented by vector b. The sum of the
displacements is equivalent to a displacement from O to B, or a + b.

8-1-1-4 Fig 5 Triangle of Vectors

Page 15 : Wed Jan 02 02:41:28 2002

Simple Arithmetic

8- 1- 1- 4

DO NOT DISTRIBUTE

AP 3456

8. Vector Difference. The difference of two vectors may be represented as a + (b), the vector b being b rotated through
180 degrees as shown in Fig 6.

8-1-1-4 Fig 6 Vector Difference

The Polygon of Vectors


9. When more than two vectors are to be resolved they can be added or subtracted two at a time as shown in Fig 7. To resolve
a + b + c + d PQ is drawn to represent a, then from the terminal point of PQ, QR is drawn to represent b, and so on until all
the vectors are represented.
PR = a + b and in the triangle PRS
PS = PR + RS = a + b + c
In the triangle PST, PT = PS + ST = a + b + c + d. Any number of vectors can be summed in this way.

8-1-1-4 Fig 7 The Polygon of Vectors

Page 16 : Wed Jan 02 02:41:29 2002

Simple Arithmetic

8- 1- 1- 4

DO NOT DISTRIBUTE

AP 3456

Aids to Calculation
Chapter 1 - Indices and Logarithms
INDICES
Introduction
1.

When a number is successively multiplied by itself it is said to be raised to a power. Thus:


5 5 5 is ve raised to the power of 3 (or 5 cubed)
4 4 is four raised to the power of 2 (or 4 squared)
6 6 6 6 6 is six raised to the power of 5

This is usually written as the number that is to be multiplied, known as the base, together with the number of times it is to be
multiplied as a superscript, known as the index. Thus:
4 4 = 42 Index
"
Base
and
6 6 6 6 6 = 65
The notation is not confined to actual numbers; algebraic symbols and expressions may be similarly expressed. Thus:
m m m m = m4
3
and (a + 2) (a + 2) (a + 2) = (a + 2)

Page 17 : Wed Jan 02 02:41:29 2002

Aids to Calculation

8- 1- 2- 1

DO NOT DISTRIBUTE

AP 3456

Multiplication and Division Rules


5
3
2. Multiplication. Suppose it is necessary to multiply 2 by 2 .
Now 25 = 2 2 2 2 2
and 23 = 2 2 2

25 23 = (2 2 2 2 2) (2 2 2) = 28
ie the result is obtained by adding the indices, eg
216 28 = 224
5
3
3. Division. If it is necessary to divide, say, 2 by 2 then this may be written as:
22222
222
2
Cancelling the terms yields the result 2 ie the result is obtained by subtracting the indices,
eg
m10 m6 = m4
4
4
4
4
0
Consider m m . Clearly any number or expression divided by itself = 1. By the subtraction rule m m = m .
0
m = 1. Indeed by the same reasoning any number or expression raised to the power zero = 1.

Negative Indices
4.

5
6
Consider 2 2 . This is equivalent to
1
22222
=
222222 2

5
6
1
By the division rule 2 2 = 2
So 21 = 12 ie the negative index indicates a reciprocal.
Similarly for example m3 =

1
m3

Fractional Indices
5. Consider the problem "What number, when multiplied by itself = 2?" Expressing this in index form and using the
multiplication rule:

Page 18 : Wed Jan 02 02:41:29 2002

Aids to Calculation

8- 1- 2- 1

DO NOT DISTRIBUTE

AP 3456

2a 2a = 2 1

a + a = 1 ie 2a = 1

a=

1
2

Thus the fractional power 2 has the meaning of square root.


1

Similarly 3 = cube root, 4 = fourth root and so on.

Power of a Power
6.

23
Consider the expression (2 ) . This is equivalent to:
(2 2) (2 2) (2 2) = 2

The result is obtained by multiplying the indices. In general terms:


m n

(a ) = amn

Standard and Engineering Forms


7. In science and engineering a very wide range of numerical values are frequently encountered. For example the velocity of
light is approximately 300,000,000 metres per second whilst the wavelengths of light are in the approximate range of
0.0000000008 metres to 0.0000000004 metres. Such very large and small numbers are clearly cumbersome in use and often
difficult to comprehend quickly. In order to overcome this difficulty it is common practice to express numbers in a standard
form or in an 'engineering' form making use of index notation.
8. The Standard Form. The standard form of a number consists of only one digit in front of the decimal point which is then
multiplied by the appropriate power of 10, ie in the form:
A 10n
where A is between 1.0000 ... and 9.9999 ... and the index, n, is the required power of 10. Thus for example:
67.9 in standard form = 6.79 10 1
679 in standard form = 6.79 10 2
300,000,000 in standard form = 3 10 8
0.00679 in standard form = 6.79 10 3
0.0000000008 in standard form = 8.0 10 10
9. Engineering Form. Engineering notation is also commonly available on calculators. It differs from the standard form in
that the power of 10 is always a multiple of 3. For example:

Page 19 : Wed Jan 02 02:41:29 2002

Aids to Calculation

8- 1- 2- 1

DO NOT DISTRIBUTE

AP 3456

300,000,000 in engineering form = 300 10 6


0.679 in engineering form = 679 10 3

Summary
10. In summary the rules for the handling of numbers or algebraic expressions in index form are as follows:
a0 = 1

am an = am+n
a m a n = a m n
m n

(a ) = amn

a m =

am =

1
am

p
m

LOGARITHMS
Introduction
11. The concept of logarithms is closely associated with the notion of indices. If a positive number, y, is expressed in index
form with a base a, ie
y = ax
then the index, x, is known as the logarithm of y to the base a. Thus:
If y = ax , then x = log a y
For example :
y = 32 = 2 5

log2 32 = 5

If log10 y = 3

Page 20 : Wed Jan 02 02:41:30 2002

Aids to Calculation

8- 1- 2- 1

DO NOT DISTRIBUTE

AP 3456

then y = 10 3

Common Logarithms
12. The most commonly used form of logarithms is to the base 10. The abbreviation 'log' is used and unless a base is explicitly
stated or otherwise implied then 10 may be assumed. Using index notation any positive integer, N, may be written as:
N = 10x

then log10 N = x
Values of the common log of any number may be found either from tables or from an electronic calculator. Prior to the
widespread use of electronic calculators, logs were used as an aid to calculation. As logs are no more than indices they obey the
same rules as indices. Thus if it is necessary to multiply two numbers this can be achieved by finding the logs of the numbers,
adding these and then finding the number corresponding to this log. Similarly, division may be accomplished by subtracting
logs, the power of a number can be found by multiplying its log by the power, and the root of a number by dividing its log by the
root index.

Naperian, Natural or Hyperbolic Logarithms


13. In many natural processes the rate of growth or decay of a substance is proportional to the amount of substance present at a
given time. It has been found that this relationship can be expressed in terms of a universal constant known as the exponential
constant, e. The number, e, is irrational and is the sum of the infinite series:

1+1+

1
1
1
1
1
+ + + + + :::::
2! 3! 4! 5! 6!

where the symbol ! means factorial, ie that number multiplied by all of the positive integers less than itself, eg 6! = 6 5 4 3
2 1.
14. By taking an appropriate number of terms, e can be calculated to any desired level of accuracy. Note that as the terms have
factorials of ever increasing numbers as their denominators then each successive term becomes smaller and the reduction in
significance is rapid. As a comparison, the third term is 0.5, the seventh term is 0.00139 and the tenth term is 0.00000276. A
value to 4 significant figures can be calculated from the first seven terms as 2.718.
15. Logarithms with e as the base are known as natural or Naperian (occasionally hyperbolic) logarithms. They are frequently
encountered in scientific texts and are the only logarithms used in calculus. The abbreviation Ln is generally used. Whereas
natural logarithms follow the same rules as common logarithms and can be used for the same purposes they are rather more
difficult to extract from tables. In any case the use of logarithms to carry out arithmetic has been superseded by the electronic
calculator.

Decibels
16. An application of logarithms is encountered in the field of amplification or gain, which is often expressed in units of bels or
more normally decibels. If P(I) is the input power into an amplifier and P(O) is the output power, the gain is given by:

G = log

or = 10 log

Page 21 : Wed Jan 02 02:41:30 2002

P(O)
bels
P(I)

P(O)
decibels
P(I)

Aids to Calculation

8- 1- 2- 1

DO NOT DISTRIBUTE

AP 3456

P(O)

The ratio of the powers P(I) can be expressed in terms of output and input voltages as:
2

P(O) V(O)
=
2 =
P(I)
V(I)

V(O)
V(I)

G = 10 log

= 20 log

V(O)
V(I)

V(O)
decibels
V(I)

Similarly in terms of output and input currents:

G = 20 log

I(O)
decibels
I(I)

17. Example. If an amplifier has a gain of 30 decibels, calculate the input voltage required to produce an output of 50 volts.

Using G = 20 log

30 = 20 log

1.5 = log

V(O)
V(I)

50
V(I)

50
V(I)

Taking antilogs:

31.62 =

V(I) =

50
V(I)

50
31.62

= 1.581 V

Aids to Calculation

Page 22 : Wed Jan 02 02:41:30 2002

Aids to Calculation

8- 1- 2- 2

DO NOT DISTRIBUTE

AP 3456

Chapter 2 - The Electronic Calculator


Introduction
1. The pocket calculator provides considerable computational power for the scientist, engineer, statistician, mathematician and
student. The range of calculators now available suits the needs of most specialists and even the most basic calculator, used
correctly, can evaluate quite complex functions. The range of machines starts at the four function calculator and moves in steps
up to the calculator with hundreds of functions, graphics display and printer. This chapter briefly reviews the initial
development towards multi-function machines, and highlights some of the useful features now widely available to users.

The Four Function Calculator


2. The four function calculator has keys for instructing the calculator to add, subtract, multiply and divide. Used correctly, and
with a note pad, a four function machine is capable of an extraordinary range of work. Unfortunately, complex functions usually
require repetitive multiple keying operations which are not only time consuming but also lead to mistakes. The development of
calculators has had as one underlying aim the reduction of the number of keying actions required and consequently the reduction
of errors and the saving of time. The only constraint upon progress has been the rate at which technology has advanced to
provide the miniaturized electronic components and circuits.
3. The easiest additional function to add to the four function calculator was the square function, because all that was required
was the double entry of a number. The reciprocal and square root functions followed swiftly, and the seven function calculator
became the next step in the remarkably rapid increase in calculator power. The range of functions grew, and by grouping
functions into particular calculators, specialized machines then became available.

Scientific and Engineering Calculators


4. The introduction of Trigonometrical functions and their inverses, Hyperbolic function and inverses, Logarithms, Squares,
Roots, Factorials, Degrees-Radians-Gradians, Polynomials and many other functions necessary for scientific and engineering use
led to the general purpose scientific calculator so commonly seen today. There have been problems to overcome beyond the
availability of functions; one such problem is associated with the display of data.
5. The Standard Display. A normal display has a capacity of ten digits and any extra digits which may be keyed are simply
dropped. This is a severe limitation to the scientist who may frequently require to carry out calculations involving figures of
more than ten digits.
6. Scientific Notation. Many multi-function calculators are able to display data in a form which takes account of very large
figures by the simple expedient of using a mantissa of up to seven digits and sign, and an exponent of 10 and sign, ie a number
11
500,000,000,000 is displayed as 5 11 which means 5 x 10 . This facility is now usually built in but it may only be available
by selection of a scientific notation key.
3
6
3
7. Engineering Notation. Engineers and technicians frequently work with exponents in multiples of three eg 10 , 10 , 10 ,
6
10
which are then associated with the terms kilo, mega, milli, micro. When engineering notation is available the results of
calculations will always be presented in the appropriate form so that the mantissa has one, two or three digits to the left of the
3
3
decimal point ie 5,000 is shown as 5 x 10 , 50,000 is shown as 50 x 10 etc.

Algebraic Hierarchy
8. Most calculators now use the algebraic operating system and the standard rules of algebraic hierarchy have been
programmed into them. These algebraic rules assign priorities to the various mathematical operations. Without a fixed list of
priorities expressions such as 5 x 4 + 3 x 2 could have several meanings:

or
9.

5 (4 + 3) 2
(5 4) + (3 2)
((5 4) + 3) 2
5 (4 + (3 2))

=
=
=
=

70,
26,
46,
50.

The rules of algebraic hierarchy state that multiplication is to be performed before addition. So algebraically, the correct

Page 23 : Wed Jan 02 02:41:30 2002

Aids to Calculation

8- 1- 2- 2

DO NOT DISTRIBUTE

AP 3456

answer is (5 x 4) + (3 x 2) = 26. The complete list of priorities for interpreting expressions is:
a. Immediate Math functions (trigonometric, logarithmic, square root etc)
x
1/y
b. Exponentiation (y ) and Roots (x )
c. Multiplication and Division
d. Addition and Subtraction
e. Equals
The algebraic hierarchy depends upon there being a memory facility in the calculator because data needs to be stored until the
completion of all entries.

Memory
10. The memory facilities generally available are user data memory and program memory although different manufacturers
may have different names for them. The amount of user data memory versus program memory is variable and may even be
adjustable in some calculators when user data memory facilities can be shared for program use. The user data memory is a
location in which numbers can be stored for later use. The memory is frequently addressable so that data may be entered into
particular registers. It may also be possible to operate on the numbers stored in memory without affecting pending operations or
the value of the display register.

Statistical Calculators
11. Statistical evaluation by calculator is heavily dependent upon good memory facilities. Usually the selection of the
statistical mode clears the memories of other data to leave them free to receive statistical data. Statistics is a second function
facility on most calculators. Large sets of data points describing some parameter of a large number of items may be reduced to a
few representative numbers; the mean, variance and standard deviation. Statisticians have been relieved of laborious repetitive
calculations by the modern calculator.

Programming
12. A program is a series of instructions that you may wish to use numerous times. The methods of entering programs, if
available, vary between calculators and are not covered here.
13. Subroutines. A subroutine is a sequence of key strokes having a certain purpose which, when needed repeatedly in a
calculation, can be called into use. The calculator is instructed to go temporarily to the sequence of steps, run that sequence and
then to return to the point where the subroutine was called. The calculator remembers the next execution step as the place to
return to. It is good practice to write programs as subroutines so that they can be used by other programs without modification.

A Practical Precaution
14. Whether a calculator has four functions or in excess of a hundred, plus graphics and printer, it should always be
remembered that careless keying can produce serious errors. It is therefore essential that the operator should be able to estimate
the result of any calculations embarked upon so that at least the right order of answer is known beforehand. The worst errors can
be avoided by the expenditure of just a few minutes spent on a rough calculation.

Aids to Calculation
Chapter 3 - Graphs
Introduction
1. The term 'graph' is usually applied to a pictorial representation of how one variable changes in response to changes in
another. This chapter will deal with the form of simple graphs, together with the extraction of data from them, and from the
'families of graphs' and carpet graphs that are frequently encountered in aeronautical publications. Although not strictly a

Page 24 : Wed Jan 02 02:41:31 2002

Aids to Calculation

8- 1- 2- 3

DO NOT DISTRIBUTE

AP 3456

'graph', the Nomogram will also be covered. The pictorial representation of data in such forms as histograms, frequency
polygons, and frequency curves will be treated in Section 6, Statistics.

Coordinate Systems
2. Graphs are often constructed from a table of, say, experimental data which gives the value of one variable, x, and the
experimentally found value of the corresponding variable, y. In order to construct a graph from this data it is necessary to
establish a framework or coordinate system on which to plot the information. Two such coordinate systems are commonly used:
Cartesian coordinates and Polar coordinates. Both systems will be described below, but the remainder of this chapter will be
concerned only with the Cartesian system.
3. Cartesian Coordinates. Cartesian coordinates are the most frequently used system. Two axes are constructed at right
angles, their intersection being known as the origin. Conventionally the horizontal 'x' axis represents the independent variable;
the vertical 'y' axis represents the dependent variable, ie the value that is determined for a given value of x. Any point on the
diagram can now be represented uniquely by a pair of coordinate values written as (x,y) provided that the axes are suitably
scaled. It is not necessary for the axes to have the same scale. Thus in Fig 1 the point P has the coordinates (3,4), ie it is located
by moving 3 units along the x axis and then vertically by 4 'y' units. It is sometimes inconvenient to show the origin (0,0) on the
diagram when the values of either x or y cover a range which does not include 0. Fig 2 shows such an arrangement where the x
axis is scaled from 0 but the corresponding values of y do not include 0. The intersection of the axes is the point (0,200). It
should be noted from Fig 1 that negative values of x or y can be shown to the left and below the origin respectively.
4. Polar Coordinates. Polar coordinates specify a point as a distance and direction from an origin. Polar coordinates are
commonly encountered in aircraft position reporting where the position is given as a range and bearing from a ground beacon;
they are also used in certain areas of mathematics and physics. As with Cartesian systems it is necessary to define an origin, but
only one axis or reference line is required. Any point is then uniquely described by its distance from the origin and by the angle
that the line joining the origin to the point makes with the reference line. The coordinates are written in the form (r,), with in
either degree or radian measure. Conventionally, angles are measured anti-clockwise from the reference line as positive and
clockwise as negative. Fig 3 illustrates the system. Point Q has the coordinates (3,30) or (3,330) in degree measure; (3,6 )

or (3,11
6 ) in radian measure.

8-1-2-3 Fig 1 Cartesian Coordinates

8-1-2-3 Fig 2 Cartesian Coordinates - Displaced Origin

Page 25 : Wed Jan 02 02:41:31 2002

Aids to Calculation

8- 1- 2- 3

DO NOT DISTRIBUTE

AP 3456

8-1-2-3 Fig 3 Polar Coordinates

The Straight Line Graph


5. Table 1 shows a series of values of x and the corresponding values of y. Fig 4 shows these points plotted on a graph.

Table 1 - Values of x and y


x

8-1-2-3 Fig 4 Graph of y = 2x

Page 26 : Wed Jan 02 02:41:31 2002

Aids to Calculation

8- 1- 2- 3

DO NOT DISTRIBUTE

AP 3456

It will be seen that all the points lie on a straight line which passes through the origin. It is clear from the table of values that if
the value of x is, say, doubled then the corresponding value of y is also doubled. Such a relationship is known as direct
proportion and the graphical representation of direct proportion is always a straight line passing through the origin. In general
the value of y corresponding to a value of x may be derived by multiplying x by some constant factor, m, ie:
y = mx
In the example m has the value 2 ie y = 2x. Because such a relationship produces a straight line graph it is known as a linear
relationship and y = mx is known as a linear equation. Such relationships are not uncommon. For example the relationship
between distance travelled, (d), speed, (s), and time, (t) is given by d = st. This would be a straight line graph with d plotted on
the y axis and t on the x axis.
6. It is of course possible for a straight line through the origin to slope down to the right rather than up to the right as in the
previous example. In this case positive values of y are generated by negative values of x and the equation becomes:
y = mx
7.

Consider now the values of x and y in Table 2, and the associated graph, Fig 5.

Table 2 - Values of x and y


x

1
0

8-1-2-3 Fig 5 Graph of y = 2x + 2

Page 27 : Wed Jan 02 02:41:32 2002

Aids to Calculation

8- 1- 2- 3

DO NOT DISTRIBUTE

AP 3456

Clearly the graph is closely related to the previous example of y = 2x. In essence the line has been raised up the y axis parallel to
the y = 2x line. Investigation of the table of values will reveal that the relationship between x and y is governed by the
equation:
y = 2x + 2
and in general a graph of this type has the equation:
y = mx + c where c is a constant.
It will be apparent that the equation y = mx is identical to the equation y = mx + c if a value of 0 is attributed to the constant c.
Thus y = mx + c is the general equation for a straight line, m and c being constants which can be positive, negative or zero. A
zero value of m generates a line parallel to the x axis. The value of c is given by the point at which the line crosses the y axis
and is known as the intercept.
8. Gradient. Consider Fig 6 which shows two straight line graphs: y = 2x and y = 4x. Both lines pass through the origin and
the essential difference between them is their relative steepness. The line y = 4x shows y changing faster for any given change
in x than is the case for y = 2x. The line y = 4x is said to have a steeper gradient than the line y = 2x. The gradient is defined as
y
the change in y divided by the corresponding change in x, ie x . Taking the equation y = mx and rearranging, yx = m, ie the
constant m is the gradient of the straight line. As the line y = mx + c has been shown to be parallel to y = mx, this clearly has the
same gradient, given by the value of m. In the equation:
distance = speed x time

Page 28 : Wed Jan 02 02:41:32 2002

Aids to Calculation

8- 1- 2- 3

DO NOT DISTRIBUTE

AP 3456

'speed' is equivalent to 'm' in the general equation, and it is apparent that the gradient, speed, represents a rate of change - in this
case the rate of change of distance with time. This concept of the gradient representing a rate of change will become important
when dealing with calculus in Section 5.

8-1-2-3 Fig 6 Graphs of y = 2x and y = 4x

Non-Linear Graphs
9. Not all relationships result in straight line graphs, indeed they are a minority. A body falling to earth under the influence of
gravity alone falls a distance y feet in time t seconds governed by the equation:
y = 16t2
Table 3 shows a range of values of t with the corresponding values of y, and Fig 7 the associated graph.

Table 3 - Values of t and y


t
y

0
0

1
16

2
64

3
144

4
256

5
400

Although not relevant in this example, notice that negative values of t produce identical positive values of y to their positive
counterparts. The graph is therefore symmetrical about the y axis and the shape is known as a parabola. The constant in front of
2
the t term determines the steepness of the graph.

8-1-2-3 Fig 7 Graph of y = 16t

Page 29 : Wed Jan 02 02:41:32 2002

Aids to Calculation

8- 1- 2- 3

DO NOT DISTRIBUTE

AP 3456

10. Consider now the problem "How long will it take to travel 120 kms at various speeds?" This can be expressed as the
equation:

t=

120
s

where t = time in hours and s = speed in kms/hr.


This is an example of inverse proportion, ie an increase in s results in a proportional decrease in t. If values are calculated for s
and t and a graph is plotted it will have the form illustrated in Fig 8 known as a hyperbola.

8-1-2-3 Fig 8 Graph of t = 120/s - Inverse Proportion

11. Graphs of y = sin x and y = cos x will be encountered frequently. The shape of the graphs are shown below (Fig 9a) and
(Fig 9b).

8-1-2-3 Fig 9a y = sin x

Page 30 : Wed Jan 02 02:41:32 2002

Aids to Calculation

8- 1- 2- 3

DO NOT DISTRIBUTE

AP 3456

8-1-2-3 Fig 9b y = cos x

The sine graph is shown with the x axis scaled in degrees while the cosine graph has the x axis scaled in radians. Either is
correct; the radian form is frequently encountered in scientific texts. Sketches of these graphs are useful when trying to
determine the value and sign of trigonometric functions of angles outside of the normal 0 to 90 range. Notice that both graphs
repeat themselves after 360 2 radians).
12. Finally it is worth considering the graph that describes the relationship:
y = eax
where a is a positive or negative constant.
This form of equation is very common in science and mathematics and variants of it can be found in the description of
radioactive decay, in compound interest problems, and in the behaviour of capacitors. The irrational number 'e' equates to 2.718
x
x
to 4 significant figures. The graph of y = e is shown in Fig 10a and that of y = e
in Fig 10b.

8-1-2-3 Fig 10a y = e to the Power of x

Page 31 : Wed Jan 02 02:41:32 2002

Aids to Calculation

8- 1- 2- 3

DO NOT DISTRIBUTE

AP 3456

8-1-2-3 Fig 10b y = e to the Power of -x

The significant point about these graphs, which are known as exponential graphs, is that the rate of increase (or decrease) of y
increases (or decreases) depending upon the value of y. A large value of y exhibits a high rate of change. It is also worth noting
that there can be found a fixed interval of x over which the value of y doubles (or halves) its original value no matter what initial
0.693
value of y is chosen. This is the basis of the concept of radioactive decay half-life. The interval is equivalent to a
where a
is the constant in the equation y = eax .
13. Logarithmic Scales. Clearly plotting and interpreting from exponential graphs can be difficult. The problem can be eased
by plotting on a graph where the x axis is scaled linearly while the y axis has a logarithmic scale. This log-linear graph paper
x
reduces the exponential curve to a straight line. A comparison between the linear and log-linear plots of y = e is shown in
Figs 11a and 11b.

Page 32 : Wed Jan 02 02:41:33 2002

Aids to Calculation

8- 1- 2- 3

DO NOT DISTRIBUTE

AP 3456

8-1-2-3 Fig 11a y = e to the Power of x, Linear Scales

8-1-2-3 Fig 11b y = e to the Power of x, Log-Linear Scales

The Presentation and Extraction of Data


14. So far this chapter has been concerned with the mathematical background to simple graphs. More commonly graphs will be
encountered and used as a source of data, especially in the field of flight planning and aircraft performance. Whilst occasionally
these graphs will be either the simple forms already described or variations on these forms, more often rather complex graphs are
used as being the only practical way of displaying complex relations. Two such types of complex graph will be described here
in order to establish the method of data extraction. Finally the nomogram will be discussed.

Page 33 : Wed Jan 02 02:41:33 2002

Aids to Calculation

8- 1- 2- 3

DO NOT DISTRIBUTE

AP 3456

8-1-2-3 Fig 12 Carpet Graph

15. Carpet Graphs. An example of a carpet graph is shown in Fig 12. The aim of the graph is to indicate the lag in the
altimeter experienced in a dive. Unlike the graphs already discussed where one input, 'x', produced one output, 'y', the carpet
graph has two inputs for one output. The output is on the conventional 'y' axis but there is no conventional 'x' axis, rather there
are two input axes. On the right hand edge of the 'carpet' diagram are figures for dive angle whilst on the bottom edge are
figures for indicated air speed. To use the graph it is necessary to enter with one parameter, say dive angle, and follow the
relevant dive angle line into the diagram until it intersects the appropriate IAS line. Intermediate dive angle and IAS values need
to be interpolated, thus in the example values of 17 and 375 knots have been entered. From the point of intersection a
horizontal line is constructed which will give the required lag correction figure where it intersects the 'y' axis, 118 feet in the
example.
16. Families of Graphs. It is often necessary to consider a number of independent factors before coming to an end result. In
this situation a family of graphs is frequently used to present the required information. Fig 13 shows such a family designed for
the calculation of the aircraft's take-off ground run. Apart from the aircraft configuration which is indicated in the graph title,
there are five input parameters. There will very often be a series of related graphs with variations in the title, for example in this
case there will be another family of graphs for an aircraft with wing stores. It is clearly important that the correct set is selected.
The method of using the graph will be described with reference to the example.
17. At the left end is a small carpet graph. Starting with the value of outside air temperature (21) proceed vertically to
intersect the altitude line (2000 feet). Alternatively enter the 'carpet' at the intersection of the altitude and temperature relative to
ISA. From this intersection proceed horizontally into the next graph to intersect the vertical reference line, marked RL. From
this point parallel the curves until reaching the point representing the value of runway slope as indicated on the bottom scale (1%
uphill). From here construct a horizontal line to the next graph reference line. Repeat the procedure of paralleling the curves for
aircraft mass (4.8 tonnes) and then proceed horizontally into the last graph for head/tail wind (20 knots head) which is used in
the same manner. Finally the horizontal line is produced to the right hand scale where the figure for ground run can be read
(1900 feet).
18. The Nomogram. The nomogram is not strictly a graph but a diagrammatic way of solving rather complex equations. There
are usually two input parameters for which one or two resultant outputs may be derived. Fig 14 shows a nomogram for the
determination of aircraft turning performance. The equations involved are:

Page 34 : Wed Jan 02 02:41:33 2002

Aids to Calculation

8- 1- 2- 3

DO NOT DISTRIBUTE

AP 3456

Rate of turn TAS =

TAS
Radius of turn

= Acceleration tan Bank Angle


This nomogram consists of four parallel scaled lines. Two known values are joined by a straight line, and the intersection of this
line with the other scales gives the unknown values. In the example illustrated, an input TAS of 180 kts with a Rate 1 turn gives
a resultant of 1.1 g, and a radius of turn of 1nm.

8-1-2-3 Fig 13 Family of Graphs

8-1-2-3 Fig 14 A Nomogram

Page 35 : Wed Jan 02 02:41:34 2002

Aids to Calculation

8- 1- 2- 3

DO NOT DISTRIBUTE

AP 3456

Aids to Calculation
Chapter 4 - Unit Conversions
S I Units
1.

The Systeme International d'Unites is a metric system based upon seven fundamental units which are:

Length
Mass

- metre (m)
- kilogram (kg)

Page 36 : Wed Jan 02 02:41:34 2002

Aids to Calculation

8- 1- 2- 4

DO NOT DISTRIBUTE

Time
Electric Current
Luminous Intensity
Temperature
Amount of
substance

AP 3456

second (s)
ampere (A)
the candela (cd)
kelvin (k)
mole (mol)

Other SI units in common usage are:

Frequency
Energy
Force
Power
Electric Charge
Potential difference
Capacitance
Inductance
Magnetic Field

hertz (Hz)
joule (J)
newton (N)
watt (W)
coulomb (C)
volt (V)
farad (F)
henry (H)
tesla (T)

Magnitudes
2.

Prefixes used with SI Units to indicate magnitudes:

Factor

Name of Prefix

Symbol

1018

atto-

1015

femto-

1012

pico-

109

nano-

106

micro-

103

milli-

102

centi-

101
10
102

deci-

decahecto-

da
h

103
106

kilo-

mega-

109
1012

giga-

tera-

1015
1018

peta-

eta-

Conversion Factors
3.

The following conversion factors have been selected as those most likely to be of general use.

Page 37 : Wed Jan 02 02:41:34 2002

Aids to Calculation

8- 1- 2- 4

DO NOT DISTRIBUTE

AP 3456

Length
4.

To Convert
0.0394
3.2808
1.0936
5.399 x 104
0.6214
Multiply By

To

Multiply By

inch (in)
feet (ft)
yard (yd)
nautical mile (nm)

millimetre (mm)
metre (m)
metre (m)
metre (m)

25.40
0.3048
0.9144
1852.0

mile

kilometre (km)

1.6093

To

To Convert

Area
5.

To Convert
0.1550

To

Multiply By

10.7639

sq in (in2)
sq ft (ft2)

sq centimetre (cm2)
sq metre (m2)

6.4516
0.0929

1.1960

sq yd (yd 2)

sq metre (m2)

0.8361

Multiply By

To

To Convert

Volume
6.

To Convert
0.2200
0.2643
0.0353
35.3147
1.3080

gal (UK)
gal (US)
cu feet (ft3)
cu feet (ft3)

To

Multiply By

litre (l)
litre (l)
litre (l)

4.5460
3.785
28.3161

cu metre (m3)
cu metre (m3)

0.0283

16.3871

0.0610

cu yard (yd 3)
cu inch (in3)

1 x 103

cu metre (m3)

cu centimeter (cm3)
litre (l)

Multiply By

To

To Convert

0.7646

1000.0

Mass
7.

To Convert
0.0353
2.2046

ounce (oz)
pound (lb)

Page 38 : Wed Jan 02 02:41:34 2002

To

gram (g)
kilogram (kg)

Aids to Calculation

Multiply By
28.3495
0.4536

8- 1- 2- 4

DO NOT DISTRIBUTE

AP 3456

0.0685

slug

kilogram (kg)

Multiply By

To

To Convert

14.5939

Velocity
8.

To Convert

To

3.2808
1.9685
0.6214
2.2369
0.5400
0.5921
1.9426

feet/sec (ft/s)
feet/min (ft/min)
miles/hour (mph)
miles/hour (mph)
knot (kt)
knot (kt)
knot (kt)

metres sec (m/s)


centimetres/sec (cm/s)
kilometres/hour (km/h)
metres/sec (m/s)
kilometres/hour (km/h)
feet/sec (fps)
metres/sec (m/s)

Multiply By

To

To Convert

Multiply By
0.3048
0.5080
1.6093
0.4470
1.8520
1.6889
0.5148

Acceleration
9.

To Convert
3.2808
0.1020
Multiply By

feet/sec2 (ft/s2)
gravitational acc (g)
To

To

metre/sec2 (m/s2)
metre/sec2 (m/s2)

Multiply By
0.3048
9.8067

To Convert

Force
10.

To Convert

0.2248
2.2046
7.2330
0.1020
32.174

pound-force (lbf)
pound-force (lbf)
poundal (pdl)
kilograms-force (kgf)
poundal (pdl)

Multiply By

To

To

Newton (N)
kilogram-force (kgf)
Newton (N)
Newton (N)
pound-force (lbf)

Multiply By
4.4482
0.4536
0.1383
9.8067
0.0311

To Convert

Torque
11.

To Convert
0.7376
8.8507
0.1020

To

pound-force foot (lbf ft)


Newton metre (Nm)
pound inch (lb in)
Newton metre (Nm)
kilogram force metre (kgf m) Newton metre (Nm)

Page 39 : Wed Jan 02 02:41:34 2002

Aids to Calculation

Multiply By
1.3558
0.1130
9.8067

8- 1- 2- 4

DO NOT DISTRIBUTE

Multiply By

To

AP 3456

To Convert

Pressure
12.

To Convert
9.869 x 103
0.0680
0.1450

To

Multiply By

atmosphere (atm)

kilopascal (kPa)

101.30

atmosphere (atm)

pound-force in2 (psi)


kilopascal (kPa)

14.6960

kilopascal (kPa)
kilopascal (kPa)
inches mercury (in Hg)
pascal (Pa)

100.0
0.1000
0.0295
1.000

inches mercury (in Hg)


kilo pascal (kPa)

0.0394
0.1334

0.0100
10.00
33.86
1.000

pounds-force in2 (psi)


bar
millibar (mbar)
millibar (mbar)
Newton/metre2 (Nm2)

25.4
7.493

mm mercury (mm Hg)


mm mercury (mm Hg)

Multiply By

To

6.8948

To Convert

Density
13.

To Convert
0.0624
103
0.0100

To

Multiply By

pound/foot3 (lbft3)
gram/centimetre3 (g/cm3)

kilogram/metre3 (kg/m3)
kilogram/metre3 (kg/m)

16.0185

pound/gal

kilogram/metre3 (kg/m3)
kilogram/litre (kg/l)

99.776

10.0221

pound/gal

Multiply By

To

1000.0

0.0998

To Convert

Power
14.

To Convert

1.3410
1.8182

horsepower (hp)
horsepower (hp)

0.7376

foot pounds-force/sec (ft


lbf/s)

Multiply By

To

To

kilowatt (kW)
foot pounds-force/sec (ft
lbf/s)
kilowatt (kW)

Multiply By
0.7457
550.0
1.3558 x 103

To Convert

Energy, Work, Heat


15.

Page 40 : Wed Jan 02 02:41:36 2002

Aids to Calculation

8- 1- 2- 4

DO NOT DISTRIBUTE

To Convert
0.7376
0.2388
9.478 x 104
3412.1
0.3725
1.3410
9.478 x 103
Multiply By

AP 3456

To

Multiply By

foot pound-force (ft lbf)


calorie (cal)
British thermal unit (Btu)

joule (j)
joule (j)
joule (j)

1.3558
4.1868
1055.1

British thermal unit (Btu)

kilowatt hour (kWh)

horsepower hour (hph)


horsepower hour (hph)
therm

megajoule (Mj)
kilowatt hour (kWh)
megajoule (Mj)

2.931 x 104
2.6845
0.7457
105.51

To

To Convert

Algebra
Chapter 1 - Principles and Rules
Introduction and Notation
1. Algebra is that branch of mathematics dealing with the properties of, and relations between, quantities expressed in terms of
symbols rather than numbers. The use of symbols allows general mathematical statements to be written down rather than just
specific ones. For example the relationship between C and F can be expressed as:

F=

9
C + 32
5

5
or C = (F 32)
9
Thus given a value of temperature in either scale, the corresponding value in the other scale can be calculated. This is a
considerably more concise method of relating the two scales than having a table showing the equivalent values, which in
practice would have to be limited to a specified range of temperatures and with a specified level of precision. The algebraic
relationship is in general more accurate than any representation by graph or nomogram.
2. Normally when an algebraic expression is written down the conventional multiplication sign is omitted, both for brevity and
to avoid confusion with the often used symbol, x. Sometimes a full-stop is used instead. The division sign is usually replaced by
the solidus, /, or by separating the expression to be divided and the divisor by a horizontal line, thus for example:
(3x 6) (7x + 3)

may be written as (3x 6) / (7x + 3)

or more commonly as

(3x 6)
(7x + 3)

The Laws of Algebra


3.

There are several laws of algebra which govern how algebraic expressions may be manipulated:

Page 41 : Wed Jan 02 02:41:36 2002

Algebra

8- 1- 3- 1

DO NOT DISTRIBUTE

AP 3456

a. Commutative Law. This law states that additions and subtractions within an expression may be performed in any order.
So may divisions and multiplications, eg
x+y=y+x

x+yz=xz+y

xy = yx

xyz = zxy = yzx


b. Associative Law. This law states that terms in an algebraic expression may be grouped in any order, eg
x + y + z = (x + y) + z

= x + (y + z)

xyz = x(yz) = (xy)z


c. Distributive Law. This law states the product of a compound expression and a single term is the algebraic sum of the
products of the single term with all the terms in the expression, eg
x(y + z) = xy + xz

4x(2y 4z) = 8xy 16xz


d. Laws of Precedence. These laws dictate the order in which algebraic operations should be effected.
First deal with terms in brackets
Then work out multiplications and divisions
Finally work out additions and subtractions.
Operations within brackets are dealt with using the same precedence.
4. Addition and Subtraction. Within an algebraic expression, like terms, eg all x terms, all y terms, all xy terms, all z terms
etc, may be collected together and combined into a single term; unlike terms cannot be so combined, eg
3x2 + 6x 4y + 2y + 5xy x 2 + 9xy
= 2x2 + 6x 2y + 14xy

Page 42 : Wed Jan 02 02:41:37 2002

Algebra

8- 1- 3- 1

DO NOT DISTRIBUTE

AP 3456

2
whereas: 3x + 6x 4y + 5xy cannot be simplified any further by addition or subtraction of terms.
5. Multiplication and Division. If two expressions which are to be multiplied together, or one divided by the other, have the
same sign the result is positive while if their signs are different the result is negative. The rules of indices (Sect 2 Ch 1) similarly
apply to algebraic expressions. Thus for example:
4xy2 x 12x3 y4 = 48x2 y6

25a4 b6 5a2 b = 5a2 b5


6.

When multiplying an expression within brackets then all of the terms within the bracket must be multiplied, eg
2

5(3x 4y + 5) = 15x 20y + 25


7. When two bracketed expressions are to be multiplied together then all of the terms within one set of brackets must be
multiplied by all the terms within the other set, eg
2

(3x + 6)(2x 4) = 6x 12x2 + 12x 24

Factorization
8. A factor is a term by which an expression may be divided without leaving a remainder; a common factor is a term which is
common to all of the terms of the expression. Thus for example in the expression bx + by, b is a common factor and the
expression may be rewritten as b(x + y). Similarly in the expression:
24a3 + 6a2 12a
6a is common to all the terms and thus it may be rewritten as:
2

6a(4a + a 2)
9.

Often an expression can be arranged into groups of terms where each group has its own factor, eg
ax + bx + ay + by

can be regarded as two pairs of terms, thus:


(ax + bx) + (ay + by)
then each pair has its own common factor, x and y respectively, and so can be rewritten as:
x(a + b) + y(a + b)
(a + b) now appears as a common factor and so the expression can be further factorized as:

Page 43 : Wed Jan 02 02:41:37 2002

Algebra

8- 1- 3- 1

DO NOT DISTRIBUTE

AP 3456

(a + b)(x + y)

Algebra
Chapter 2 - Equations
Introduction
1. An equation is a mathematical statement expressing an equality, ie it equates one algebraic expression with another.
Equations may range in complexity from simple linear equations which contain only one unknown quantity and whose graphical
representation is a straight line, to complex equations containing elements of calculus. This chapter will deal with simple linear,
simultaneous and quadratic equations.

Transposition
2. Perhaps the most common use of an equation is in the determination of the value of one parameter given the values of other
terms. Thus for example given the equation for temperature conversion:

F=

9
C + 32
5

if values for C are given then F can be determined by substituting the value in the equation, eg for 20 C:

F=

9 20
+ 32
5

= 36 + 32 = 68 F
3. However, suppose that it is necessary to find the Celsius equivalent of a Fahrenheit temperature. Clearly the equation needs
to be rearranged so that C becomes the subject. In order to achieve this it is important to remember that operations may be
carried out on the equation provided that the same process is applied to both sides of the ' = ' sign. The only forbidden operation
is division by zero; multiplication by zero is not forbidden but of course gives the trivial result 0 = 0.
4.

In the example the first step is to subtract 32 from both sides of the equation:
9
F 32 = C + 32 32
5

Next both sides of the equation can be multiplied by 5/9, remembering that all terms must be multiplied:
5 9
5
(F 32) = C
9
9 5
Thus

C=

5
(F 32)
9

5. Frequently equations will be encountered which contain powers and/or roots. These can be dealt with in an analogous
fashion provided again that the same operations are carried out on both sides of the equation.

Page 44 : Wed Jan 02 02:41:37 2002

Algebra

8- 1- 3- 2

DO NOT DISTRIBUTE

AP 3456

6. As an example, the periodic time of a simple pendulum, of length L, is given by the equation:

T = 2

L
seconds
g

Suppose it is required to make 'L' the subject of the equation. The first operation is to square both sides to remove the square
root:
T2 = (2)

L
L
= 4 2
g
g

2
Then divide both sides by 4 :
L
T2
=
g
4 2
Finally multiply both sides by g (and conventionally the subject term is taken to the left side).

Thus L =

gT2
4 2

7. Sometimes the way to proceed is not immediately obvious; the relationship between the distance of an object from a lens,
(u), the distance of its image, (v), and the focal length of the lens, (f), is given by:
1 1 1
+ =
u v
f
Suppose it is necessary to make f the subject of this equation. One technique is to initially multiply by the product of the
denominators of the left-hand side, ie by uv.
uv uv uv
+ =
u
v
f

ie v + u =

uv
f

Next, taking the 'f' term to the left and inverting both sides.
1
f
=
uv u + v
Finally multiply again by uv:
f=

8.

uv
v+u

However there are often alternative methods, for example taking the optical equation again:

Page 45 : Wed Jan 02 02:41:38 2002

Algebra

8- 1- 3- 2

DO NOT DISTRIBUTE

AP 3456

1 1 1
+ =
u v
f
The left-hand side can be combined into one term by using a common denominator, uv, thus:
v+u 1
=
uv
f
then inverting both sides
f=

uv
v+u

Simple Linear Equations


9. A simple linear equation is one which has only one unknown quantity and in which the unknown quantity has no power
other than 1, eg 4x + 6 = 30. The solution, ie finding the value of x which satisfies the conditions of the equation, is
accomplished using the transposition techniques discussed above. Thus in the example:
4x + 6 = 30
Subtract 6 from both sides:
4x = 30 6 = 24
Divide both sides by 4
x = 24 / 4 = 6
Notice that addition and subtraction is equivalent to transferring the term to the other side of the equation accompanied by a
change of sign, eg
Taking
4x + 9 = 3x 6
The '3x' term can be transferred to the left -hand side if its sign is changed to '', and similarly the '9' may be transferred to the
right -hand with a sign change, thus:
4x 3x = 6 9
ie
x = 15

Linear Simultaneous Equations

Page 46 : Wed Jan 02 02:41:38 2002

Algebra

8- 1- 3- 2

DO NOT DISTRIBUTE

AP 3456

10. Linear simultaneous equation are independent equations, with no powers other than 1, relating to more than one unknown.
All of the equations must be true at the same time. For example:
x + 3y = 20

(1)

9x y = 12

(2)

In general, to find values of all the unknowns which satisfy the equations then it is necessary to have as many independent
equations as there are unknowns, for example if there are 5 unknowns then 5 independent equations would be required. For a
pair of simultaneous equations with two unknowns there are two methods of solution; elimination and substitution.
11. Solution by Elimination. In this method one or both of the equations are manipulated so that the coefficient of one of the
unknowns is identical in each equation. One equation is then subtracted from the other to eliminate one unknown resulting in a
simple equation in the other unknown which can be solved readily. This value is then substituted back into one of the original
equations to generate another readily soluble simple equation. Taking the examples from paragraph 10:
Multiply equation (1) by 9:

Subtract equation (2)

9x + 27y = 180
9x y = 12
28y = 168

Divide by 28

y=6

Substitute this value for y in equation (2)


9x 6 = 1

9x = 18

x =2
12. Solution by Substitution. In this method one equation is rearranged to express one unknown in terms of the other. This
'value' is then substituted into the other equation, which reduces to a simple equation in one unknown. After solution of this
simple equation the value is substituted into either equation to find the other unknown value. Taking the same example
equations:
x + 3y = 20

(1)

9x y = 12

(2)

Rearrange equation (1) to make 'x' the subject:


x = 20 3y

Page 47 : Wed Jan 02 02:41:38 2002

Algebra

8- 1- 3- 2

DO NOT DISTRIBUTE

AP 3456

Substitute this into equation (2) and solve for y:


9(20 3y) y = 12

180 27y y = 12

28y = 168

y=6
Substitute this value into equation (1):
x + 18 = 20

x=2

Quadratic Equations
2
13. A quadratic equation contains the square of the unknown quantity but no higher power. The simplest type has the form
px
= n where n is a positive number. The solution is a simple matter of finding the square root of the positive number ie x = n,
remembering that the result can have a negative or positive value. More commonly a quadratic equation has the form:
ax2 + bx + c = 0
where a and b and c are numbers.
14. It is instructive to examine the graphs representative of quadratic functions; their shape is parabolic. The solutions, or
roots, of a quadratic equation are where y = 0 on the graph, ie where the graph crosses the x axis. Four examples are illustrated
in Fig 1.

8-1-3-2 Fig 1 Graphs of Quadratic Equations

Page 48 : Wed Jan 02 02:41:38 2002

Algebra

8- 1- 3- 2

DO NOT DISTRIBUTE

AP 3456

2
15. Fig 1a shows the graph of y = x 4x + 3. It will be seen that there are two positive roots; where x = 1 and where x = 3, ie
2
where the graph crosses the x axis. If the function had been x + 4x + 3 then the graph would have crossed the x axis to the left
of the origin, ie giving two negative roots, 1 and 3.
2
16. Fig 1b shows the graph of y = x x 2. Here the graph crosses the axis at two points, one to the left and one to the right
of the origin, thus there is one positive root and one negative root.
2
17. Fig 1c shows the graph of y = x 4x + 4. Here the graph does not cross the x axis, rather the x axis is a tangent to the
curve at the value x = 2. Here the two roots are said to be coincident or equal.
2
18. Fig 1d shows the graph of y = x 2x + 2. In this case the graph does not cross the x axis at all therefore there are no real
roots.

Page 49 : Wed Jan 02 02:41:39 2002

Algebra

8- 1- 3- 2

DO NOT DISTRIBUTE

AP 3456

Solving Quadratic Equations


19. Clearly a quadratic equation can be solved as illustrated above by drawing the graph of the function and establishing where
the curve crosses the x axis. However this method is tedious and often inaccurate. There are two other methods of solution in
common use: factorization and formula
20. Factorization. Some quadratic equations, but not all, are readily solved by the factorization method. The equation must
first be arranged so that all the terms are on the left -hand side with just a zero to the right of the '=' sign. The problem then is to
find factors of the expression on the left -hand side, remembering that it will not always be possible. Consider the equation:
x2 4x + 3 = 0
The left -hand side can be factorized as:
(x 1)(x 3) = 0
As the left -hand side is now comprised of the product of two factors equal to zero, then one of the factors at least must equal
zero, ie either:
x 1 = 0
._. x = 1
or
x3=0
._. x = 3
21. Formula. The formula method can be used to solve all quadratic equations that have a solution and, unless the factors are
readily apparent, is normally the preferred method of solution. The equation must first be arranged into the form:
ax2 + bx + c = 0
The formula to be used is then:

x=

b2 4ac

2a

As an example take the equation:


2x2 + 3x 2 = 0
In terms of the standard form, a = 2, b = 3, c = 2
Thus putting these values into the formula:

Page 50 : Wed Jan 02 02:41:39 2002

Algebra

8- 1- 3- 2

DO NOT DISTRIBUTE

AP 3456

x=

x=

9 ( 16)
4

3 + 5
4

._. x =

1
2

or

x=

3 5
4

. _ . x = 2
2
22. The part of the formula 'b 4ac' is known as the discriminant and gives information about the roots of the equation. There
are three possible cases:
2
a. b > 4ac. This generates a positive term and so will have two real square roots. Thus there will be two real roots to the
equation. This is the situation shown by Figs 1a and 1b.
2
2
b. b = 4ac. This is the case illustrated by Fig 1c. b 4ac = 0 and the roots are coincident and equal to b/2a.
2
2
c. b < 4ac. This makes b 4ac negative. There are no real square roots to a negative number and therefore the equation
has no real roots. This is the case illustrated in Fig 1d where the graph does not cross the x axis. Although there are no real
solutions in this case, this form of equation has many applications in, for example, control systems and aerodynamics.

8-1-3-2 Fig 1a

Page 51 : Wed Jan 02 02:41:39 2002

Algebra

8- 1- 3- 2

DO NOT DISTRIBUTE

AP 3456

8-1-3-2 Fig 1b

Page 52 : Wed Jan 02 02:41:39 2002

Algebra

8- 1- 3- 2

DO NOT DISTRIBUTE

AP 3456

8-1-3-2 Fig 1c

Page 53 : Wed Jan 02 02:41:39 2002

Algebra

8- 1- 3- 2

DO NOT DISTRIBUTE

AP 3456

8-1-3-2 Fig 1d

Page 54 : Wed Jan 02 02:41:39 2002

Algebra

8- 1- 3- 2

DO NOT DISTRIBUTE

AP 3456

Geometry and Trigonometry


Chapter 1 - Plane Trigonometry
Definitions and Axioms
1. Angles. An angle is formed by the intersection of two lines. In Fig 1 AOB is an angle which is formed by a line which
starts from the position OA and rotates about O in an anti-clockwise direction to the position OB. In this case O is the "vertex"
of the angle while OB and OA are the "arms" of the angle. With anti-clockwise rotation the angle is regarded as positive; if
clockwise the angle is regarded as negative. Unless otherwise stated it is assumed that rotation will be anti-clockwise.

8-1-4-1 Fig 1 An Angle

2. Measurement of Angles. One complete revolution is divided into 360 degrees (). The degree is sub divided into 60

Page 55 : Wed Jan 02 02:41:40 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

minutes ('), and the minute is sub-divided into 60 seconds ("). Ninety degrees constitutes one right angle.
3. Acute, Obtuse and Reflex Angles. An acute angle is one which is less than 90 degrees, an obtuse angle is greater than 90
degrees but less than 180 degrees and a reflex angle is greater than 180 degrees (see Fig 2a, Fig 2b and Fig 2c).

8-1-4-1 Fig 2 Acute, Obtuse and Reflex Angles

4. Complementary and Supplementary Angles. If two angles together make up 90 degrees they are said to be complementary
angles and each is the complement of the other. If two angles together make up 180 degrees they are said to be supplementary
and each is the supplement of the other.
5. Slope and Gradient. The slope of the line CA in Fig 3 is the angle ACB. The gradient of the line CA is the ratio AB/CB =
1/5.

8-1-4-1 Fig 3 Slope and Gradient

Page 56 : Wed Jan 02 02:41:40 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

6. Angles Formed by Two Intersecting Straight Lines. When two straight lines intersect as in Fig 4 the sum of the adjacent
angles is 180 degrees, and the vertically opposite angles are equal.
A+B = B+A 1 = A1 +B1 = B1 +A = 180
A = A1
B = B1

8-1-4-1 Fig 4 Two Intersecting Straight Lines

7. Parallel Lines Cut by a Transversal. When a transversal intersects two parallel straight lines as in Fig 5:

8-1-4-1 Fig 5 Parallel Straight Lines and Transversal

Page 57 : Wed Jan 02 02:41:40 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

a. The corresponding angles are equal (The corresponding angles are equal (B =D, F =H, A =C, G =E).
b. The alternate angles are equal (F =B, C =G).
c. The sum of the interior angles on the same side of the transversal are equal to 180 (C + B = 180, F + G =
180).
8. Angle Properties of a Triangle. Angle Properties of a Triangle. The sum of the angles of a triangle is 180 . When one side
of a triangle is produced as in Fig 6 the exterior angle thus formed is equal to the sum of the two interior opposite angles.
6A

+ 6B + 6C = 180
+ 6BAC = 6ACO

6ABC

8-1-4-1 Fig 6 Angle Properties of a Triangle

9. Congruency of Triangles. Two triangles are congruent if one can be superimposed on the other, so that they exactly
coincide with regard to their vertices and their sides. Their areas must consequently be equal. Thus the three sides of one

Page 58 : Wed Jan 02 02:41:40 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

triangle must have the same lengths as the three sides of the other and the angles opposite to the equal sides must be equal.
Triangles can be proved congruent when:
a. The three sides of one are equal to the corresponding sides of the other.
b. They have two sides and the included angle of one, equal to the corresponding sides and included angle of the other.
c. They have two angles and a corresponding side equal.
10. The Theorem of Pythagoras. The conventional notation used for the solution of triangles is to denote the three angles by
the capital letters A,B,C and the sides opposite these angles by the small letters a,b,c. Pythagoras' theorem states that in any right
angled triangle the square on the hypotenuse is equal to the sum of the squares on the other two sides, ie in Fig 7 where the angle
2
2
2
at C = 90, c = b + a . This theorem is of considerable importance and can be used to find one side of a right angled triangle
when the other two are known. For example if a 12 metre ladder rests against a house so that its foot is 4 metres from the wall, it
is possible to calculate how far up the side of the house the ladder will reach (see Fig 8).

8-1-4-1 Fig 7 Pythagoras' Theorem

8-1-4-1 Fig 8 A Practical Application of Pythagoras' Theorem

Page 59 : Wed Jan 02 02:41:40 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

From Pythagoras it is known that


a2 = b2 +c2

. _ . b2 = a2 c2
ie b =

144 16, thus the ladder reaches 11.3 metres up the wall.

11. Similar Triangles. If two triangles have three angles of one equal to the three angles of the other they are not necessarily
congruent.

8-1-4-1 Fig 9 Similar Triangles

Consider Fig 9 in which angle A is common to the three triangles AFG, ADE and ACB.
6AFG

6AGF

= 6ADE = 6ACB (corresponding angles)


= 6AED = 6ABC (corresponding angles)

Such triangles are said to be similar. When triangles are equiangular the ratios of corresponding sides are also equal.
Thus
AF AD AC
=
=
AG AE AB
and it follows that
AG FG
AF AG
=
and
=
AD AE
AE DE
NOTE:
A similar relation holds good for two polygons which are equiangular.

Page 60 : Wed Jan 02 02:41:41 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

12. The Relationship Between Sides and Areas of Similar Triangles. Triangles A1 B1 C1 and A2 B2 C2 are similar triangles
with heights h1 and h2 respectively (see Figs 10a and 10b).

8-1-4-1 Fig 10 Areas of Similar Triangles

Then
h 1 h2
=
a1 a2

. _ . h1 =

a1 h2
a2

Also
Area of A 1 B1 C1 12 a1 h1
=
Area of A 2 B2 C2 12 a2 h2
Substituting for h1
a h

1
1 2
Area of A 1 B1 C1
a 2
2 a1 a 2
= 1
= 12
Area of A 2 B2 C2
a2
2 a2 h 2

Similarly the areas can be proved proportional to the squares of b1, b2 and c1, c2. Hence the areas of similar triangles are
proportional to the squares of the corresponding sides.

Trigonometrical Ratios
13. In any right angled triangle the side opposite to the right angle is called the hypotenuse and the other two sides are called

Page 61 : Wed Jan 02 02:41:41 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

the opposite and adjacent according to their position relative to the angle under consideration. Fig 11 shows a right-angled
triangle ABC in which, relative to angle A, the side BC is opposite and the side AC is adjacent. The reverse is true relative to
angle B. There are six trigonometrical ratios:

8-1-4-1 Fig 11 Right-Angled Triangle

The sine of an angle


(sin) =

opposite
hypotenuse

(cos) =

adjacent
hypotenuse

The cosine of an angle

The tangent of an angle


(tan) =

opposite
adjacent

The cosecant of an angle


(cosec) =

1
sin A

The secant of an angle


(sec) =

1
cos A

(cot) =

1
tan A

The cotangent of an angle

14. The trigonometric ratios in terms of the triangle in Fig 11 are:

Page 62 : Wed Jan 02 02:41:41 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

sin A =

a
c

sin B =

b
c

cos A =

b
c

cos B =

a
c

tan A =

a
b

tan B =

b
a

cosec A =

c
a

cosec B =

sec A =

c
b

sec B =

c
a

cot A =

b
a

cot B =

a
b

sin A
cos A

cot A =

cos A
sin A

c
b

and it may be deduced that

tan A =

sin A = cos(90 A)

cos A = sin(90 A)

tan A = cot(90 A)

cot A = tan(90 A)

15. The Trigonometric Ratios for Angles of any Magnitude. So far, only acute angles have been considered but it is also
necessary to be able to find the trigonometrical ratios of obtuse, reflex and sometimes negative angles. Consider a set of
rectangular axes OX, OY (see Fig 12). To determine any trigonometrical ratio of any angle, the angle is set up on this system of
axes as follows. A radius vector, OP, initially along OX, is considered to turn about O in a counter-clockwise sense through the
required angle, A. For a negative angle it turns in the clockwise sense. From P drop a perpendicular, PN, on to the x-axis. Any
trigonometrical ratio of A is then referred to the right-angled triangle OPN and the acute angle which OP makes with the x-axis.
OP is always taken to be +ve, but ON and PN take the signs which would be attached to them when regarded as the coordinates
of the point P. Thus, for the angles A and A in the figure:

8-1-4-1 Fig 12 The Four Quadrants

Page 63 : Wed Jan 02 02:41:41 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

PN
and is +ve
OP

sin A =

sin (A) =

cos A =

ON
and is ve
OP

cos (A) =

tan A =

P'N
and is ve
OP'

ON
and is ve
OP'

PN
and is ve
ON

tan (A) =

P'N
and is +ve
ON

The reciprocal ratios, cosec, sec and cot have the same sign respectively as sin, cos and tan. These more general definitions of
the trigonometrical ratios, which apply to all angles of any magnitude and sign, are consistent with the former definitions which
applied only to acute angles, since, for an acute angle, the radius vector, OP would lie in the first quadrant and ON and OP
would then both be +ve. It has been shown that for angles in the second quadrant, sin is +ve while cos and tan are -ve. Similarly
it can be shown that for angles in the third quadrant, tan is +ve, while sin and cos are -ve, and for angles in the fourth quadrant,
cos is +ve while sin and tan are -ve. Hence the "all, sin, tan, cos" rule for determining the sign of a trigonometrical ratio. Fig 13
indicates which functions are positive in each of the quadrants.

Page 64 : Wed Jan 02 02:41:42 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

8-1-4-1 Fig 13 Signs of Trigonometrical Functions by Quadrant

The Sine Rule


16. The three sides and three angles of a triangle are sometimes called its "elements". When given a sufficient number of these
elements it is possible to find the remainder. Thus for example if two sides and one angle or one side and two angles are known,
and in each case an angle and the side opposite to it are included, the Sine Rule can be used to evaluate the unknown sides and
angles.
17. In the triangle ABC at Fig 14 draw AD perpendicular to BC and let AD = p.

8-1-4-1 Fig 14 The Sine Rule

p
= sin B
c

. _ . p = c sin B

Page 65 : Wed Jan 02 02:41:42 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

p
= sin C
b

. _ . p = b sin C

. _ . c sin B = b sin C
or
c
b
=
sin C sin B
in a similar way it can be proved that:
c
a
=
sin C sin A
and the Sine Formula is:
b
c
a
=
=
sin A sin B sin C
18. In the case of triangle ABC where C is obtuse as in Fig 15:

8-1-4-1 Fig 15 Sine Rule - Obtuse Angled Triangle

Again, let p be the perpendicular to BC produced,


p
= sin B
c

Page 66 : Wed Jan 02 02:41:42 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

. _ . p = csin B
and
p
= sin ACD
b

. _ . p = b sin ACD
but
sin ACD = sin (180ACD) = sin C

. _ . p = c sin B = b sin C, as before or

b
sin B =

c
sin C

The Cosine Rule


19. When the Sine Rule is not applicable, as, for instance, when two sides and an included angle are given, the Cosine Rule
may be used. Consider the triangle ABC in Fig 16:

8-1-4-1 Fig 16. The Cosine Rule

From A draw AD = (= p) perpendicular to BC.


Let DC = n and BD = an.
Then by the principle of Pythagoras,
p 2 = b 2 n2
and
p2 = c2 (a n)

Page 67 : Wed Jan 02 02:41:42 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

. _ . b2 n2 = c2 (a 2an + n 2 )
. _ . b2 n2 = c2 a2 + 2an n 2 )
. _ . b2 = c2 a2 + 2an
but n = b cos C
b2 = c2 a2 + 2ab cos C
c2 = a2 + b2 2ab cos C
20. In the case of triangle ABC where C is an obtuse angle, as in Fig 17:

8-1-4-1 Fig 17 Cosine Rule - Obtuse Angled Triangle

p 2 = b 2 n2
p2 = c2 (a+n)

. _ . b2 n2 = c2 a2 2an n2
. _ . b2 = c2 a2 2an

Page 68 : Wed Jan 02 02:41:43 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

where
n = b cos ACD

= b cos ACB

= b cos C
. _ . b2 = c2 a2 2a(b cos C)
. _ . b2 = c2 a2 + 2ab cos C
. _ . c2 = a2 + b2 2ab cos C
which is identical to the previous formula.
21. By the same method it can be shown that
b2 = a2 + c2 2ac cos B
and
a2 = b2 + c2 2bc cos A
Thus given any two sides and their included angle the third side can be found. It may then be more convenient to apply the Sine
Rule to find any other unknown elements.
22. When three sides a, b and c are given the cosine form may be re-arranged as follows:
c2 = a2 + b2 2ab cos C
2ab cos C = a 2 + b2 c2

cos C =

a2 + b 2 c 2
2ab

Similarly cos A and cos B may be found.

8-1-4-1 Fig 2a - Acute

Page 69 : Wed Jan 02 02:41:43 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

8-1-4-1 Fig 2b - Obtuse

8-1-4-1 Fig 2c - Reflex

8-1-4-1 Fig 10a

8-1-4-1 Fig 10b

Page 70 : Wed Jan 02 02:41:43 2002

Geometry and Trigonometry

8- 1- 4- 1

DO NOT DISTRIBUTE

AP 3456

Geometry and Trigonometry


Chapter 2 - The Circle
Definitions Relating to the Geometry of a Circle
1.

Some important definitions relating to the circle are explained with reference to Fig 1.

8-1-4-2 Fig 1 The Properties of a Circle

a. The Chord of a circle is any straight line which divides the circle into two parts, and is terminated at each end by the
circumference. AB in Fig 1 is a chord.
b. A Segment of a circle is a figure bounded by a chord and the arc which it cuts off. In Fig 1 the chord AB divides the
circle into two segments.

Page 71 : Wed Jan 02 02:41:44 2002

Geometry and Trigonometry

8- 1- 4- 2

DO NOT DISTRIBUTE

AP 3456

(1) The minor segment is ABC.


(2) The major segment is ADB.
c. An Arc of a circle is a portion of the circumference. EF in Fig 1 is an arc.
d. A Sector of a circle is a figure which is bounded by two radii and the arc between them. OEF in Fig 1 is a sector.
e. A Tangent to a circle is a straight line which meets the circle in one point called the point of contact and does not cut the
circle when produced. A tangent is at right angles to the radius drawn from the point of contact. GHI in Fig 1 is a
tangent.
f. The Angle in a Segment is the angle subtended at a point on the arc of a segment by the chord of the segment. In Fig 1
the angle in the major segment is ADB and the angle in the minor segment is ACB.
g. The Angle at the Centre is the angle subtended at the centre of a circle by a chord or by an arc. EOF in Fig 1 is the
angle at the centre subtended by EF.

Theorems
2. The angle at the centre of a circle subtended by an arc is double the angle at the circumference subtended by the same arc.
In Fig 2 AOB = 2 ADB and the reflex AOB = 2 ACB. Some important results follow from this theorem:
a. All angles in the same segment of a circle are equal.
b. The opposite angles of a quadrilateral in scribed in a circle are together equal to 180; that is, they are supplementary.

8-1-4-2 Fig 2 The Angles in a Circle

c. The angle in a semi-circle is a right angle.


d. In equal circles, arcs which subtend equal angles either at the centres or at the circumference are equal.
e. In equal circles, chords which cut off equal arcs are equal.
3.

The angle between a tangent and a chord drawn through the point of contact is equal to the angle in the alternate segment.

Page 72 : Wed Jan 02 02:41:44 2002

Geometry and Trigonometry

8- 1- 4- 2

DO NOT DISTRIBUTE

AP 3456

In Fig 3 DBC = DEB and ABD = DFB.

8-1-4-2 Fig 3 The Angle between Tangent and Chord

4.

The ratio of the circumference of a circle to its diameter is denoted by , so that C/d =
or C = d

Since diameter is equal to two times the radius,


C=2r

Circular Measure
5. The magnitude of an angle is commonly expressed in degrees which are obtained by the division of a right angle into 90
parts. There is another method which is of great practical importance and in which the unit employed is an absolute one.
Consider Fig 4. Suppose the line OA in Fig 4 rotated about the point O to the position OB, so that the length of the arc AB is
equal to the radius of the circle. The angle AOB subtended by the arc AB is called a radian. The radian is the unit of
measurement in circular measure. Hence a radian may be defined as the angle subtended at the centre of a circle by an arc equal
in length to the radius.

8-1-4-2 Fig 4 The Radian

Page 73 : Wed Jan 02 02:41:44 2002

Geometry and Trigonometry

8- 1- 4- 2

DO NOT DISTRIBUTE

6.

AP 3456

The length of an arc when the angle is given in radians can be calculated as follows:
Length of an arc for 1 radian = r

Length of an arc for 1 radian = r

Arc = r
7. The Relationship Between Radians and Degrees. Since an arc of r units in length subtends an angle of 1 radian, the number
of radians subtended by the circumference of a circle is given by the number of times the radius is contained in the
circumference.
ie
C=2r
The number of radians for one revolution

2r
= 2 radians
r

From this, radians = 180

Page 74 : Wed Jan 02 02:41:44 2002

Geometry and Trigonometry

8- 1- 4- 2

DO NOT DISTRIBUTE

AP 3456

and 1 radian = 180/ = 57.3


Radians may be converted to degrees by multiplying by and dividing by 180.
Examples:
Convert to radians 45, 30 taking as 3.1416

a.

45 x 3.1416 3.1416
=
= 0.7854 rad
180
4

b.

30 x 3.1416 3.1416
=
= 0.5236 rad
180
6

Conversions are easily carried out if an electronic calculator is available which will enter an accurate value for at the touch of a
button.

Angular Rotation
8. A straight strip of tape stuck on the face of a gear wheel from the axis to perimeter will enable the observation that in one
revolution of a gear wheel an individual tooth is rotated through 360. Since 360 = 2 radians then one revolution is also 2
radians. In circular measure, therefore, all even multiples of correspond to complete revolutions. eg 4 radians will be 2
revolutions, 6 radians will be 3 revolutions etc.
9. If a shaft or pulley is rotating at 3 revolutions per second then the angular rotation must be 3 x 2 radians per second. In
general terms, a rotation of n revs per second will give an angular velocity of 2 n radians per second.
10. The Relationship Between Angular and Linear Velocity. Let QMN of Fig 5 represent a flywheel which has an angular
velocity of radians per sec. This means that any radius OQ rotates through an angle of radians in 1 sec. Any point P on OQ
will also have the same angular velocity. Since arc = r , the arc traced out by Q in 1 sec = . OQ, and the arc traced out by P in
1 sec = . OP. In general, if the point is at a distance r from the centre of rotation, the linear velocity of that point will be r.
Let V be the linear velocity of a point, then V = r. Although all points on the flywheel have the same angular velocity, the
linear velocity of any point will depend on its distance from the centre of rotation.

8-1-4-2 Fig 5 Angular and Linear Velocity

Page 75 : Wed Jan 02 02:41:44 2002

Geometry and Trigonometry

8- 1- 4- 2

DO NOT DISTRIBUTE

AP 3456

Geometry and Trigonometry


Chapter 3 - Spherical Trigonometry
Introduction
1. Many of the problems encountered in navigation require the solution of triangles on the Earth's surface. In all but very small
triangles the effect of curvature must be taken into account.

GEOMETRY OF THE SPHERE


2. The following properties and definitions apply to the geometrical sphere and, although the shape of the Earth is not truly
spherical, it is considered to be so for most practical navigational purposes.
a. The section of the surface of a sphere made by any intersecting plane is a circle
b. The axis of a circle on a sphere is the diameter of the sphere at right angles to the plane of the circle (PP' in Figs 1 and
2).
c. The two points at which the axis of the circle intersects the surface of the sphere are called the poles of the circle (P and
P' in Figs 1 and 2).
d. If the plane passes through the centre of the sphere the section is called a great circle. All other sections are called small
circles.
e. A quadrant is the great circle arc drawn from any point on a great circle to its pole (Fig 1).
f. Only one great circle passes through two points which are not at opposite ends of a diameter of a sphere. The shorter arc
of this great circle is the shortest distance between these two points, measured over the surface of the sphere.

8-1-4-3 Fig 1 Great Circle

Page 76 : Wed Jan 02 02:41:44 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

8-1-4-3 Fig 2 Small Circle

Spherical Distance
3. The spherical distance between two points on the surface of a sphere is the length of the shorter great circle arc joining them.
It is measured by the angle which that arc subtends at the centre of the sphere, expressed in degrees, minutes and seconds, or in
radians. In Fig 3, the spherical distance BC is measured as BC, eg BC = 42 27' means that the arc BC subtends an angle of
42 27' at the centre of the sphere.

8-1-4-3 Fig 3 Spherical Distance

Page 77 : Wed Jan 02 02:41:45 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

4.

AP 3456

Angular measurement of spherical distance is convenient for the following reasons:


a. The actual length of the arc is readily obtained, given the radius of the sphere, since the length of the arc = radius of the
sphere x the angle subtended at the centre (in radians). In the case of the Earth, by definition, an arc of length 1 nautical
mile subtends an angle of 1'. Thus, if B and C are two points on the surface of the Earth, the length of BC is given directly
by converting BC to minutes, eg:
BC = 42 27 0 = 2,547 nm
b. Angular measurement of spherical distance can be made without reference to the size of the sphere; this is of value when
dealing with an abstract quantity such as the celestial sphere.
c. The use of angular measurement of the sides simplifies the solution of the various formulae used in spherical
trigonometry.

Spherical Angle
5. A spherical angle is formed at a point where two great circles intersect and is measured by the angle between the great
circles at that point. This is the equivalent of measuring the angle between the planes of the two great circles in a plane mutually
perpendicular to them both.

8-1-4-3 Fig 4 Spherical Angle

Page 78 : Wed Jan 02 02:41:45 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

6. In Fig 4, BC is the spherical angle formed by the great circles AB and AC at the point A and is measured by bc = BC =
arc BC. Now BOC is a plane perpendicular to the planes of the great circles ABD and ACD and is contained in the plane of the
great circle XBCY. Then OA is perpendicular to the plane of XBCY and A is a pole of that great circle. By definition, BC is a
measure of the spherical angle at A, from which it follows that the length of arc BC is also a measure of A. Thus the spherical
angle formed at a point may be measured by the arc intercepted between those great circles along the great circle to which that
point is a pole.

The Spherical Triangle


7. A spherical triangle is formed by the intersection of three great circles on the surface of a sphere. Fig 5 shows such a
triangle, ABC. The arcs BC, CA and AB form the sides of the triangle and are denoted by a, b and c respectively. The angles
b C and BCA form the angles of the triangle and are denoted by A, B and C. Note that by this convention, a is the side
BC, AB
opposite angle A, b is opposite angle B and c is opposite angle C.

8-1-4-3 Fig 5 Spherical Triangle

Page 79 : Wed Jan 02 02:41:45 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

The Polar Triangle


8. Consider the spherical triangle ABC in Fig 6. Point A' is a pole of the great circle of which side a is a part and is that
particular pole which lies on the same side of the great circle as angle A. Similarly, B' and C' are the poles appropriate to sides b
and c respectively.

8-1-4-3 Fig 6 Polar Triangle

Page 80 : Wed Jan 02 02:41:45 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

9. The 3 points so obtained are joined by arcs of great circles B'A', A'C' and C'B' giving a second spherical triangle A'B'C'.
The original triangle is called the primitive and the second, the polar triangle. It should be noted that in many cases the shape of
the polar triangle might bear little resemblance to that of its primitive.
10. From para 2c, if A' is the pole of arc a then the arc BA' is a quadrant. But point A' lies on the arc b', therefore B is also a
pole of arc b'. Similarly, A and C are poles of arcs a' and c' respectively; thus triangle ABC is the polar triangle of A'B'C'. So, if
one triangle should be a polar triangle of another, the latter will be the polar triangle of the former.

Relationship between the Primitive and Polar Triangles


11. In Fig 6, let D and E be the points where the arc a' is intercepted by the arcs c and b respectively. Then, from para 6, since
A is a pole of a', the spherical angle A is measured by the arc DE. But B'E and C'D are both quadrants:
. _ . B0 E + C0 D = 180
also
B0 E + C0 D = B0 C0 + DE

. _ . DE = 180 B 0 C0
= 180 a0
or, angle A is the supplement of the angle subtended by arc a'. Similarly it can be proved that:
A0 = 180 a, B 0 = 180 b, C 0 = 180 c

Page 81 : Wed Jan 02 02:41:45 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

and
a0 = 180 A, b 0 = 180 B, c0 = 180 C

Properties of Spherical Triangles


12. The following statements are rules of spherical trigonometry and are stated without the proof which may be found in any
basic spherical trigonometry primer. For a spherical triangle:
a. By convention, each side is less than 180.
b. From a above it follows that each angle must be less than 180.
c. Any two sides are together greater than the third side.
d. If two sides are equal, the angles opposite those sides are equal; conversely, if two angles are equal, then the sides
opposite those angles are equal.
e. The greater side is opposite the greater angle; conversely, the greater angle is opposite the greater side.
f. The sum of all the sides is less than 360
g. The sum of all the angles lies between 180 and 540.

Determination of Spherical Triangles


13. The spherical triangle has 6 parts, ie 3 sides and 3 angles. In general, if any 3 parts are known the triangle is fixed. The
following combination of 3 parts all determine unique triangles:
a. Two sides and the included angle.
b. Two angles and the included side.
c. Three sides.
d. Three angles.
In the following cases there may be 1 or 2 solutions.
a. Two sides and a non-included angle.
b. Two angles and a non-included side.
For example, in Fig 7, given c, b and B, there are 2 possible triangles, ABC1 and ABC2. In Fig 8, given B, C and c, there are
again 2 possible triangles, ABC1 and ABC2.

8-1-4-3 Fig 7 Two Possible Triangles given Two Sides and a Non-included Angle

Page 82 : Wed Jan 02 02:41:46 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

8-1-4-3 Fig 8 Two Possible Triangles given Two Angles and a Non-included Side

Page 83 : Wed Jan 02 02:41:46 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

Symmetrical Equality
14. The 2 triangles in Fig 9 obey all the normal rules of congruency, ie the sides and angles of one are equal to the
corresponding sides and angles of the other. However, triangle ABC cannot be superimposed on AB'C' since its curvature is in
the opposite sense. Hence, the 2 triangles cannot be truly congruent since they fail in this respect. They are, therefore, said to be
symmetrically equal.

8-1-4-3 Fig 9 Symmetrically Equal Triangles

15. Points to Note. The following points of difference between plane and spherical triangles should be noted:
a. Given 2 angles of a spherical triangle, the third angle is still undetermined since, from para 12g, the sum of the three
angles may be anywhere between 180 and 540. This is in contrast to the plane triangle in which the third angle may be
obtained by subtracting the sum of the two known angles from 180.
b. Since 3 angles determine a unique spherical triangle (para 12d) it follows that similar triangles do not occur on the same
sphere.
c. In plane trigonometry, the angles of an equilateral triangle are all 60. In an equilateral spherical triangle, however,
whilst the 3 angles are all equal, their value is not restricted to 60.

FORMULAE FOR THE SOLUTION OF SPHERICAL TRIANGLES


16. Spherical triangles may be solved by various formulae as listed below.

The Sine Formula


17. From C (Fig 10) drop a perpendicular to the plane OAB, meeting the plane in H. Draw a perpendicular from C to OA,
meeting that line in J. Then JH will be perpendicular to OA. Draw a perpendicular from C to OB, meeting that line in K. Then
KH will be perpendicular to OB.

8-1-4-3 Fig 10 Proof of the Sine Formula

Page 84 : Wed Jan 02 02:41:46 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

18. From Fig 10:

CJ = OC sin CJ
OJ = OC cos CJ
JH = CJ cos CJH
= OC sin CJ cos CJH
CH = OC sin CJ sin CJH
Let OC = 1 unit

CJ = b (angle subtended at the centre)


CJH = A (CJ is in the plane of arc b, JH is in
the place of arc c, CJH is the angle
between the planes)
OJ
CJ
JH
CH

= cos b and
= sin B
= sin b cos A
= sin b sin A

.......... (1)

=
=
=
=

.......... (2)

Similarly,

OK
CK
KH
CH

cos a and
sin a
sin a cos B
sin a sin B

Equating (1) and (2):

sin b sin A = sin a sin B

Page 85 : Wed Jan 02 02:41:46 2002

.......... (3)

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

By repeating this construction in the plane OCB it may be shown that:

.......... (4)

sin b sin C = sin c sin B


From (3) and (4) it follows that:

sin b
sin c
sin a
=
=
.......... (5)
sin A sin B sin C
Or, the sines of the angles are proportional to the sines of the sides opposite. This expression is known as the Sine Formula

The Cosine Formula


19. Fig 11 shows the plane OAB of Fig 10. By definition BC = c (angle subtended at the centre). Draw JL perpendicular to
OB; draw HM perpendicular to JL. Then OK = OJ cos c + JH sin MJH. But MJH = 90 MJO = BA = c.

.......... (6)

OK = OJ cos c + JH sin c

8-1-4-3 Fig 11 The Plane OAB of Fig 10

Substituting the values for OK, OJ, and JH in (6)

cos a = cos b cos c + sin b sin c cos A

.......... (7)

By repeating the construction in the planes of OCB and OCA, 2 further expressions are obtained, viz:

cos b = cos a cos c + sin a sin c cos B


cos c = cos a cos b + sin a sin b cos C

.......... (8)
.......... (9)

Equations 7, 8 and 9 are of the same form and permit the determination of 1 side knowing the other 2 sides and the included
angle.

Examples of the Use of the Sine Formula


20. The Sine formula can be used in the following cases:
a. To find a non-included angle, given 2 sides and a non-included angle.
b. To find a non-included side, given 2 angles and a non-included side.
In accordance with para 13, these cases are ambiguous and may yield 2 possible results, as demonstrated below.

Page 86 : Wed Jan 02 02:41:46 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

21. Example 1, Sine Formula. From Fig 12, find C, given A = 38 42', a = 76 18', c = 57 25'.
From the Sine Formula

sin C sin A
=
sin c
sin a

sin C =

sin C =

sin A sin c
sin a

sin 38 42 0 sin 57 25 0
sin 76 18 0

Using calculator or logarithms (log sin C = 1.73422)

. _ . C = 32 50 0 or 147 10 0
In this particular case, one result may be eliminated by applying the rule that the greater side must be opposite the greater angle
(para 12c): a is greater than c hence A must be greater than C; therefore C cannot have a value 147 10'. The requirements are
satisfied by 1 triangle only and, therefore, C = 32 50'.
22. Example 2, Sine Formula. From Fig 13, find b, given c = 28 25', B = 62 07' and C = 33 42'.
sin c
sin b
=
sin B sin C

8-1-4-3 Fig 12 Example 1 - Sine Formula

Page 87 : Wed Jan 02 02:41:47 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

sin b =

sin b =

sin B sin c
sin c

sin 62 07 0 sin 28 25 0
sin 33 42 0

So, using calculator or logarithms (log sin b= 1:87973)


b = 49 180 or 130 42 0
In this case, B is greater than C, thus b must be greater than c. Both values of b satisfy this requirement, hence the ambiguity is
unresolved and both results must be accepted.

8-1-4-3 Fig 13 Example 2 - Sine Formula

Page 88 : Wed Jan 02 02:41:47 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

Examples of the Use of the Cosine Formula


23. The Cosine formula may be used as follows:
a. To find the third side, given 2 sides and the included angle.
b. To find the third angle, given 2 angles and the included side (by transposition to the polar triangle).
Both these cases determine unique triangles; hence no ambiguity will arise.
24. Example 1, Cosine Formula. From Fig 14, find c, given a = 47 15', b = 115 20' and C = 82 38'. From the Cosine
formula:
cos c = cos a cos b + sin a sin b cos C

cos c = cos 47 15 0 cos 115 20 0 + sin 47 15 0 sin 115 20 0 cos 82 38 0


Now 115 20' is in the second quadrant. It may thus be written that:
cos 115 20 0 = cos 64 40 0
sin 15 20 0 = sin 64 40 0

. _ . cos c = cos 47 15 0 cos 64 40 0 + sin 47 15 0 sin 64 40 0 cos 82 38 0

= 0.29044 + 0.08510

. _ . cos c = 0.20534

Page 89 : Wed Jan 02 02:41:47 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

c = 180 78 09 0
= 101 51 0

8-1-4-3 Fig 14 Example 1 - Cosine Formula

25. Example 2, Cosine Formula. From triangle ABC in Fig 15a, find C, given A = 47 15', B = 115 20', c = 82 38'. In this
case two angles and the included side are given and the Cosine formula is not directly applicable. However, the polar triangle
may be derived thus; from the rules of para 11:
a0 = 180 A = 180 47 15 0 = 132 45 0
b0 = 180 B = 180 115 20 0 = 64 400
C0 = 180 c = 180 82 38 0 = 97 220
c0 = 180 C0 .

8-1-4-3 Fig 15 Example 2 - Cosine Formula Indirect Application

Page 90 : Wed Jan 02 02:41:47 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

The given quantities are now in terms of 2 sides and an included angle and:
cos c0 = cos a0 cos b0 + sin a0 sin b0 cos C0

cos c0 = cos 132 45 0 cos 64 40 0 + sin 132 45 0 sin 64 40 0 cos 97 22 0

cos c0 = cos 47 15 0 cos 64 40 0 sin 47 15 0 sin 64 40 0 cos 82 38 0

= 0.29044 0.08510

= 0.37554

c0 = 112 03

10
2

. _ . C0 = 180 112 03

= 67 56

Page 91 : Wed Jan 02 02:41:47 2002

10
2

10
2

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

The Haversine Formula


26. In the preceding examples it has been necessary to consider the sign of the various functions. This is an inconvenience and
calculations would be much simplified if only positive values occurred. The Haversine (half-reverse-sine) formula may be used
to achieve this object.
27. The expression

1 - cos A
is known as the haversine of an angle A, written hav A and has special properties. Thus:
2

When
A = 0 cos A = 1 and hav A = 0

A = 90 cos A = 0 and hav A =

1
2

A = 180 cos A = 1 and hav A = 1

A = 270 cos A = 0 and hav A =

1
2

A = 90 cos A = 0 and hav A =

1
2

When

A = 180 cos A = 1 and hav A = 1

A = 270 cos A = 0 and hav A =

1
2

A = 360 cos A = 1 and hav A = 0


The value of the haversine never exceeds 1 and is always positive irrespective of whether the angle is positive or negative and
since:

hav a =

1 cos a
1 cos A
and hav A =
2
2

cos a = 1 2 hav a
and
cos A = 1 2 hav A)

Page 92 : Wed Jan 02 02:41:48 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

Substituting for cos a and cos A in the Cosine formula (para 19):
cos a = cos b cos c + sin b sin c cos A

1 - 2 hav a = cos b cos c + sin b sin c (1 - 2 hav A)


or
1 - 2 hav a = cos b cos c + sin b sin c - 2 sin b sin c hav A
Now:
cos b cos c + sin b sin c = cos (b - c)
and
cos (b - c) = 1-2 hav (b - c)

. _ . 1 - 2 hav a = 1 - 2 (b - c) - 2 sin b sin c hav A


From which:
hav a = hav (b - c) + sin b sin c hav A
28. Since the haversine is always positive, the value of hav (b - c) is positive no matter what values are assigned to b and c and
the equation may be simplified by writing (b c), meaning the difference between b and c. So:

hav a = hav (b c) + sin b sin c hav A

.......... (10)

Similarly it may be shown that:


hav b = hav (a c) + sin a sin c hav B
and
hav c = hav (a b) + sin a sin b hav C
By convention, only angles and sides up to 180 are considered. Terms sin a, sin b and sin c will, therefore, always be positive;
hence every term in the Haversine formula is positive.

Example of the Use of the Haversine Formula


29. The Haversine formula is used to
a. Find the third side, given 2 sides and the included angle

Page 93 : Wed Jan 02 02:41:48 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

b. Find the third angle, given 2 angles and the included side (by transposition to the polar triangle).
From triangle ABC in Fig 16, find c given a = 47 15', b = 115 20' and C = 82 38'.

8-1-4-3 Fig 16 Application of the Haversine Formula

hav c = hav (47 15 115 200 ) +


sin 47 15 0 sin 115 20 0 hav 82 38 0

hav c = hav 68 05 0

+ sin 47 15 0 sin 64 40 0 hav 82 38 0

= 0.31337 + 0.28931

. _ . c = 101 51 0

The Cosecant Formula


30. Given the 3 sides of a spherical triangle, the 3 angles may be determined by substitution in the Haversine formula. This is
transposed for convenience to give the Cosecant formula:
hav a = hav (b c) + sin b sin c hav A

Page 94 : Wed Jan 02 02:41:48 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

hav A =

hav a hav (b c)
sin b sin c

hav A = hav a - hav (b c) cosec b cosec c .......... (11)


The other 2 forms are:
hav B = hav b hav (a c) cosec a cosec c
and
hav C = hav c hav (a b) cosec a cosec b
From these equations, all 3 angles may be determined. Alternatively, given 3 angles, the 3 sides can be obtained by
transposition to the polar triangle.

Example of the Use of the Cosecant Formula


31. From the triangle ABC in Fig 17a, find side a given A = 82 30', B = 60 52' and C = 45 02'. The corresponding sides of
the polar triangle (Fig 17b) are:
a0 = 180 82 30 0 = 97 30 0
b0 = 180 60 52 0 = 119 08 0
c0 = 180 45 02 0 = 134 58 0
A0 = 180 a

8-1-4-3 Fig 17 Application of the Cosecant Formula

Page 95 : Wed Jan 02 02:41:48 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

Then
0

hav A0 = [hav a hav (b c0 )] cosec b cosec c0


0

= [hav 97 30 hav (119 08 134 580 )]


0

(cosec 119 08 cosec 134 58 0 )


0

= [hav 97 30 hav 15 50 0 ]
0

(cosec 60 52 cosec 45 02 0 )
0

(Because 60 52 is 180 119 08 0 and 45 02 0 is 180 134 58 0 )

= [0.56526 0.01897]
0

(cosec 60 52 cosec 45 02 0 )
Using calculator or logarithms (log hav A' = 1:94642)
. _ . A0 = 140 10 0
and
a = 180 140 10 0

Page 96 : Wed Jan 02 02:41:49 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

= 39 50 0
Sides b and c can be found in a similar manner.

The Half-log Haversine Formula


32. By rewriting equation (11), substituting

1 - cos a
for hav a and 1
2

- cos (b
2

c) for hav (b c), the following expression is

obtained:
1
hav A = [1cos a (1cos(b c))] (cosec b cosec c)
2

1
= cos[(b c) cos a](cosec b cosec c)
2
Now

(b c)+a
(b c)a
cos(b c)cosa= 2sin
sin
2
2

= 2sin

a (b c)
a + (b c)
sin
2
2

But:

hav =

1 cos 2sin2 2

=
= sin2
2
2
2

So:
p
sin = hav
2
Therefore:

sin

p
a + (b c)
= hav [a + (b c)]
2

and

p
a (b c)
sin
= hav [a (b c)]
2
Therefore:

Page 97 : Wed Jan 02 02:41:49 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

hav A =

hav [a + (b c)]

hav [a (b c)]

(cosec b cosec c)
Similarly:
hav B =

hav [b + (a c)]

hav [b (a c)]

(cosec a cosec c)

hav C =

hav [c + (a b)]

hav [c (a b)]

(cosec a cosec b)
This equation is easier to manipulate than the Cosecant formula since a straight multiplication is the only operation required. A
calculator or logarithms will produce the relevant results. It should be noted that:

log

1
hav [a + (b c)] = log hav [a + (b c)]
2

An alternative to this formula is called the All Natural Haversine formula where

hav A =

hav a hav (b c)
hav (b + c) hav (b c)

Example of the Use of the Half-log Haversine Formula


33. Using triangle ABC in Fig 17 again, find side a given A = 82 30', B = 60 52' and C = 45 02'. The corresponding sides of
the polar triangle are:
a0 = 180 82 30 0 = 97 30 0
b0 = 180 60 52 0 = 119 08 0
c0 = 180 45 02 0 = 134 58 0
b0 c0 = 15 50 0

hav A0 =

hav(97 30 + 15 500

Page 98 : Wed Jan 02 02:41:50 2002

hav(97 30 15 500 (cosec 119 08 cosec 134 58 0 )

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

hav (113 20 )

hav (81 40 ) (cosec 60 52 cosec 45 02 0 )

(Because 60 52 is 180 119 08 0 and 45 02 0 is 180 134 58 0 )


Using calculator or logarithms (log hav A' = 1.94642)
. _ . A0 = 140 10 0 ,
and
a = 180 140 10 0
= 39 50 0
sides b and c can be found in similar manner.

The Four Parts Formula


34. A case that arises frequently in practice is that in which, given 3 consecutive parts of a spherical triangle (say 2 sides and an
included angle), it is required to know a further angle. This could be solved by using the Haversine and Cosecant formulae in
succession, but such a method is laborious.
35. Consider any spherical triangle ABC in which a, c and B are known.

.......... (12)

Then: cos a = cos b cos c + sin b sin c cos A

cos b = cos a cos c + sin a sin c cos B

sin b =

sin a
sin B
sin A

Substituting for sin b and cos b in equation (12):


cos a = cos c (cos a cos c + sin a sin c cos B)

+ sin c cos A

sin a
sin B
sin A

= cos a cos 2 c + cos c sin a sin c cos B

+ sin a sin c sin B cot A

Page 99 : Wed Jan 02 02:41:50 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

2
cos a (1 cos c) = sin a sin c (cos c cos B + sin B cot A)
2
2
Now: 1 cos c = sin c
2
So: cos a sin c = sin a sin c (cos c cos B + sin B cot A)

cot a sin c = cos c cos B + sin B cot

.......... (13)

Likewise:
cot a sin b = cos b cos C + sin C cot A
cot b sin c = cos c cos A + sin A cot B
cot c sin a = cos a cos B + sin B cot C
cot b sin a = cos a cos C + sin C cot B
cot c sin b = cos b cos A + sin A cot C
36. The following rules, using Fig 18 as an example, may assist in memorizing these formulae:
a. The four parts follow consecutively around the spherical triangle, eg A, c, B, a in the example.
b. c and B are known as inner parts.

8-1-4-3 Fig 18 Illustration of Four Parts Formula

c. A (to be found) and a are known as outer parts.


d. The equation always follow the form cot sin cos, cos sin cot.
e. The sides always appear in the first 3 terms, the angles always appear in the last 3 terms.

Page 100 : Wed Jan 02 02:41:50 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

f. Each inner part appears twice in the equation.


g. Each outer part appears only once.
h. The outer parts appear at the each end of the equation.
Applying these rules to the Fig 18:
The sequence is: cot sin = cos cos + sin cot
B and c must both appear twice, a appears only once and at the left-hand end since it is a side and must be in the first 3 terms;
Therefore, A must be at the right hand end. Thus cot a sin c = cos c cos B + sin B cot A which matches equation (13) and is
correct.

Example of the Use of the Four Parts Formula


37. The Four Parts formula can be used as follows:
a. To find an angle, given 2 sides and the included angle
b. To find a side, given 2 angles and the included side.
38. In the spherical triangle ABC at Fig 19, find B given that A = 18 55', b = 123 59', c = 36 58'.

8-1-4-3 Fig 19 The Four Parts Formula - Example

B and B are the outer parts.

Page 101 : Wed Jan 02 02:41:51 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

. _ . cot b sin c = cos c cos A + sin A cot B

cot B =

cot b sin c cos c cos A


sin A

cot 56 01 0 sin 36 580 cos 36 58 0 cos 18 55 0


sin 18 55 0

Using calculator or logarithms

0.40536 0.75584
sin 18 55 0

1.16120
sin 18 55 0

. _ . cot B = 0.55411
B = 15 36 0
B = 180 15 36 0
B = 164 24 0

Tangent Formula or Napier's Analogies


39. Although the lengthy working is omitted, from the Sine and Cosine formulae it can be proven that for any spherical triangle
ABC:
1 sin12(a b)
1
tan (A B) = cot C
..........(14)
2
2 sin(a + b)
Also
1 cos12(a b)
1
tan (A + B) = cot C
..........(15)
2
2 cos12(a + b)
40. From any polar triangle A'B'C':
0

C = (180 c )

Page 102 : Wed Jan 02 02:41:51 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

1
1
. _ . cot C = cot (90 c0 )
2
2

1
= tan c0
2

(A B) = (180 a ) (180 b )
0

= (a b0 )

tan

1 0
1
(A B) = tan (a b0 )
2
2
1 0
= tan (a b0 )
2

Similarly
0

(a b) = (A B0 )

sin

1 0
1
(a b) = sin (A B0 )
2
2
1 0
= sin (A B0 )
2

1
1
0
(a + b) = (180 A + 180 B0 )
2
2

1 0
= 180 (A + B0 )
2

sin

1 0
1
(a + b) = sin 180 (A + B0 )
2
2
1 0
= sin (A + B0 )
2

Substituting these values in equation (14):

Page 103 : Wed Jan 02 02:41:52 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

sin12 (A B0 )
1 0
1
tan (a b0 ) = tan c0
0
2
2
sin12 (A + B0 )

1 0
1 sin12 (A B0 )
. _ . tan (a b0 ) = tan c0
2
2 sin12 (A0 + B0 )
Thus in any spherical triangle ABC:
1 sin12(A B)
1
..........(16)
tan (a b) = tan c
2
2 sin12(A + B)
Similarly, for equation (15):
1 cos12(A B)
1
..........(17)
tan (a + b) = tan c
2
2 cos12(A + B)

Examples of the Use of the Tangent Formulae


41. The tangent formulae are used:
a. Given 2 sides and the included angle, to find the other 2 angles.
b. Given 2 angles and the included side, to find the other 2 sides
c. Given 2 sides and the angles opposite, to find the other unknowns.
42. Example 1, Tangent Formula. In Fig 20 (a repeat of Fig 16), find A and B, given a = 47 15', b = 115 20' and C = 82 38'.

8-1-4-3 Fig 20 Example 1 of the Tangent Formulae

Page 104 : Wed Jan 02 02:41:52 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

0
1
82 380 sin 12 (47 15 115 20 )
tan (A B) = cot
2
2 sin 12 (47 150 + 115 20 0 )

A useful feature of the tangent formulae emerges here. Since b > a, sin2 (a b) is negative. However, with b > a, it follows that
B > A and tan (A - B) is also negative. The negative sign appears on both sides of the equation and may, thus, be disregarded. It
is therefore permissible to write:
1 sin 12(b a)
1
tan (B A) = cot C
2
2 sin 12(b + a)
That is to say, in the application of the tangent formulae to any example the order may be changed so that the smaller quantity is
subtracted from the larger and negative angles do not occur. This operation must be performed throughout all terms in the
equation.
Then
0

1
0 sin 34 02 2
tan (B A) = cot 41 19
0
2
sin 81 17 12

1
0 cos 34 02 2
tan (B + A) = cot 41 19
0
2
cos 81 17 12

Page 105 : Wed Jan 02 02:41:52 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

Using a calculator or logarithms:


10
1
(B A) = 32 47
2
2

10
1
(B + A) = 80 52
2
2

1
1
B = (B A) + (B + A)
2
2

1
1
A = (B + A) (B - A)
2
2

. _ . A = 113 40 0
and
B = 48 05 0
43. Example 2, Tangent Formula. In Fig 21, find B, given A = 38 42', a = 76 18', C = 32 50'. and c = 57 25'.

8-1-4-3 Fig 21 Example 2 of the Tangent Formulae

Page 106 : Wed Jan 02 02:41:52 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

sin 12(a c)
1
tan (A C) = cot B
2
sin 12(a + c)

tan 12(A C) sin 12(a + c)


1
cot B =
2
sin 12(a c)

tan 12 (38 42 32 500 ) sin 12 (76 18 + 57 25 0 )


0

sin 12 (76 18 57 250 )

tan 2 56 0 sin 66 51 12

sin 12 (76 18 57 250 )

From which, using a calculator or logarithms:


B = 147 57 0

Right-angled Spherical Triangles


44. Consider a triangle ABC in Fig 22 in which the spherical angle A = 90. Then, by the Four Parts formula:

8-1-4-3 Fig 22 Right-angled Spherical Triangle

Page 107 : Wed Jan 02 02:41:53 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

cot a sin c = cos c cos B + sin B cot A


but:
cot A = cot 90 = 0

. _ . cot a sin c = cos c cos B


and:
cos B = cot a tan c
Now:
cos B = sin (90 B)
and:
cot a = tan (90 a)

sin (90 B) = tan (90 a) tan c

.......... (18)

From the Cosine Formula:


cos a = cos b cos c + sin b sin c cos A

Page 108 : Wed Jan 02 02:41:53 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

but:
cos A = 0
so:
cos a = cos b cos c

sin (90 a) = cos b cos c

.......... (19)

45. By taking each form of the Cosine and Four Parts formulae in turn a series of expressions can be obtained as follows:
Sin (90 C) = tan (90 a) tan b
Sin (90 C) = cos (90 B) cos c
Sin (90 B) = cos (90 C) cos b
Sin (90 A) = tan (90 B) tan (90 C)
Sin c = tan (90 B) tan b
Sin c = cos (90 a) cos (90 C)
Sin b = tan (90 C) tan c
Sin b = cos (90 B) cos (90 a)

Napier's Rules of Circular Parts


46. The foregoing rules are difficult to memorize and are conveniently summarized in Napier's Rules of Circular Parts. Fig 23
shows a right-angled spherical triangle with the appropriate circular parts written alongside. Note that:

8-1-4-3 Fig 23 Diagram for Napier's Rules of Circular Parts for a Right-angled Spherical Triangle

Page 109 : Wed Jan 02 02:41:53 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

a. The parts are written down in the order in which they appear in the triangle.
b. The right angle is not counted as a circular part and is represented in the diagram by the double line.
c. The circular parts corresponding to the other 2 angles are the complements of those angles.
d. The circular part corresponding to the side opposite the right angle is the complement of that side.
47. Provided that the circular parts are written down in accordance with the above principles, any one of the formulae in para
45 may be derived on sight from the following 2 rules:
a. The sine of the middle part is equal to the product of the tangents of the adjacent parts.
b. The sine of the middle part is equal to the product of the cosines of the opposite parts.
For example, select any part as the middle part. Let this be c in Fig 23.
Then:
b and (90 B) are the adjacent parts
. _ . sin c = tan b tan (90 B)
(90 a) and (90 C) are the opposite parts
. _ . sin c = cos (90 a) cos (90 C)

Examples of the Use of Napier's Rules.


48. Napier's Rules are especially useful when it is required to solve the spherical triangle for any other part, given:
a. Two sides and a non-included angle.

Page 110 : Wed Jan 02 02:41:53 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

b. Two angles and a non-included side.


This is done by dividing the triangle into 2 right-angled triangles and applying the preceding rules. In para 22, an attempt was
made to solve such a case by use of the Sine formula and it was apparent that ambiguity could arise. This, unfortunately, is also
possible using Napier's Rules.
49. Example Using Napier's Rules. In the spherical triangle ABC in Fig 24, find side b, given that B = 62 07', C = 33 42' and
c = 28 25'. To simplify the calculation, first construct AD, a perpendicular drawn from A to arc BC. Let this be designated d.
In order to obtain b, side d is required.

8-1-4-3 Fig 24 Example of Napier's Rules

sin d = cos (90 B) cos (90 c)


= cos 27 53 0 cos 61 35 0

8-1-4-3 Fig 25 Circular Parts for Triangle ABD

Page 111 : Wed Jan 02 02:41:54 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

Using a calculator or logarithms

d = 24 52

10
10
or 155 07
2
2

But c is opposite the right-angle so d cannot possibly be larger than c.


Hence:

d = 24 52

10
2

Continuing, from Napier's Rules using the circular parts diagram for triangle ADC at Fig 26:

8-1-4-3 Fig 26 Circular Parts for Triangle ADC

Page 112 : Wed Jan 02 02:41:54 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

sin d = cos (90 C) cos (90 b)

= sin C sin b

sin b = sin d cosec C

= sin 24 52

10
cosec 33 42 0
2

Using a calculator or logarithms:


b = 49 180 or 130 42 0
both of which are valid.

Right-sided Spherical Triangles


50. A spherical triangle (Fig 27) in which 1 side has a value of 90 (sometimes called a 'quadrantal triangle') may be solved by
Napier's Rules because, if a triangle is right-sided, it follows that its polar triangle is right-angled. To save the labour of
conversion to the polar form in such cases, the following rules for the circular parts of right-sided triangles are stated without
proof:

8-1-4-3 Fig 27 Right-sided Spherical Triangle

Page 113 : Wed Jan 02 02:41:54 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

a. The parts are written down in the order in which they appear in the spherical triangle.
b. The right side is not counted as a circular part.
c. The circular parts corresponding to the other 2 sides are the complements of those sides.
d. The circular part corresponding to the angle opposite the right side is the complement of that angle.
Fig 28 shows the circular parts diagram in this case.

8-1-4-3 Fig 28 Circular Parts Diagram for Right-sided Triangle

Page 114 : Wed Jan 02 02:41:54 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

51. Napier's Rules for right-sided spherical triangles are the same as those given in para 47, viz:
a. The sine of the middle part is equal to the product of the tangents of the adjacent parts.
b. The sine of the middle part is equal to the product of the cosines of the opposite parts.
The exception is that when the adjacent or opposite parts are both sides or both angles a negative sign is added to the equation.
eg sin (90 A) = tan (90 b) tan (90 c)
but sin (90 b) = + cos (90 c) cos B
52. Example of a Right-sided Spherical Triangle. In Fig 29, find A, given a = 90, c = 73 19' and b = 54 32'.
By Napier's Rules:
sin (90 A) tan (90 b) tan (90 c)
. _ . cos A = tan 35 28 0 tan 16 41 0

8-1-4-3 Fig 29 Right-sided Triangle Example

Page 115 : Wed Jan 02 02:41:54 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

Using a calculator or logarithms


A = 102 19.7 0

Summary of Formulae
53. Table 1 summarizes the use of the various formulae covered in this chapter

Table 1 Summary of Formulae

Page 116 : Wed Jan 02 02:41:55 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

8-1-4-3 Fig 17 Application of the Cosecant Formula

Page 117 : Wed Jan 02 02:41:55 2002

Geometry and Trigonometry

8- 1- 4- 3

DO NOT DISTRIBUTE

AP 3456

8-1-4-3 Fig 17 Application of the Cosecant Formula

Introduction to Calculus
Chapter 1 - Functions and Limits
Functions
1.

In Section 2, Chapter 3 (Graphs) it was shown that the relationship between two variables, x and y say, can be expressed in

Page 118 : Wed Jan 02 02:41:55 2002

Introduction to Calculus

8- 1- 5- 1

DO NOT DISTRIBUTE

AP 3456

an equation such as y = mx + c. The principle is not confined to the linear relationship but may also be extended to such
equations as:
y = sin x, y = ex etc
Since values are attributed to x it is known as the independent variable; the corresponding values of y may then be determined,
and y is therefore known as the dependent variable.
2. The dependence of y upon x is expressed mathematically in the phrase 'y is a function of x' and is usually written as y = f(x),
in which f(x) is a shorthand way of indicating some expression in terms of x.
Thus in the expression
y = x2 4x +3,

f(x) is x 4x + 3

similarly in
y = sin 2x,

f(x) is sin 2x

and in
y = e2x ,

f(x) is e

2x

In each case by plotting the graphs of these functions a smooth curve is obtained whose shape depends upon the nature of f(x).
3. In each of the above examples an explicit statement has been made, ie y is equal to some function of x. Such functions are
known as explicit functions.
4. It is however possible to write a function such as 9x + 6xy + 4y 2 = 1 in which although there is no direct statement of y in
terms of x, it is evident nevertheless that corresponding values of y could be determined by giving values to x. Such a function
is known as an implicit function.

Gradients
5. Suppose that an object is moving in a straight line in such a way that its distance, s metres, from a fixed point on the line at
any time, t seconds after it started moving, is governed by the equation:
s = 12 + 10t t 2

ie s = f(t)

By giving a series of values to t and calculating the corresponding values of s then a graph of the function can be plotted
showing how s changes as t changes. Such a graph is shown in Fig 1.
6. Information about the speed at which the object is moving can be obtained from this graph by constructing chords. For
example over the period of 5 seconds the increase in s is indicated by PF = 25 metres and the object's average speed over the
period is therefore 25/5 m/sec = 5 m/sec. Letting FAP be called then tan = PF/AP = 25/5. So the average speed during the
5 secs is given by the slope, or gradient, of the chord AF. Similarly the object's average speed over the first 4 secs is given by
the gradient of the chord AE = 24/4 = 6 m/sec.

8-1-5-1 Fig 1 Distance/Time Graph s = 12 + 10t - t

Page 119 : Wed Jan 02 02:41:55 2002

Introduction to Calculus

8- 1- 5- 1

DO NOT DISTRIBUTE

AP 3456

Thus it can be inferred that the average speed over any selected period of time will be given by the gradient of the chord
spanning that part of the curve. For example the average speed of the object during the third second of its movement is given by
the gradient of the chord CD, ie 5 m/sec.
7. Whereas, for reasonably long intervals of time, it is possible to measure the gradient of the chord directly from the graph, if
it becomes necessary to determine the gradient over a short period, such as KL, then this method will be difficult and inaccurate.
However, it is possible to obtain the desired result by using the actual function:
s = 12 + 10t t 2
As an example suppose that it is required to find the average speed of the object over the period of time from t = 3 secs to t = 3.1
secs.
After 3.1 secs
2

s = 12 + 10.(3.1) (3.1) m

= (43 9.61) m

= 33.39 m
After 3 secs
s = (12 + 30 9) m

= 33 m
Therefore, in 0.1 secs the object covered 0.39 m at an average speed of 3.9 m/sec.

Page 120 : Wed Jan 02 02:41:55 2002

Introduction to Calculus

8- 1- 5- 1

DO NOT DISTRIBUTE

AP 3456

8. By shortening the interval of time to 0.01 secs, ie from 3 to 3.01 secs, and substituting these figures in the function, the
average speed becomes 3.99 m/sec. Taking an even shorter interval from 3 secs to 3.001 secs yields an average speed of 3.999
m/sec. If the same exercise is repeated for time intervals just prior to 3 secs the following results are obtained:
a. From 2.9 to 3 secs: 4.1 m/sec
b. From 2.99 to 3 secs: 4.01 m/sec
c. From 2.999 to 3 secs: 4.001 m/sec
From the figures it can be inferred that at the precise time of 3 secs the actual or instantaneous speed was 4 m/sec.

8-1-5-1 Fig 2 Two Chords on Magnified Section of s = 12 + 10t - t

9. Fig 2 shows a magnified section of the graph with just two of the chords drawn. The gradient of the chord PD represents the
average speed between 2.9 and 3.0 secs; the gradient of DQ represents the average speed between 3.0 and 3.1 secs. The chord
PD has been extended to M and the chord QD projected back to L. Between 2.9 and 3.1 secs the chord PM rotates about an axis
through D until it is aligned with LQ. At some instant during this rotation the chord will take up the position of the tangent to
the curve at D. It can be inferred that this will occur at the time t = 3 secs; thus the gradient of the tangent at a point on a
distance/time graph measures the actual speed at that instant ie the rate of change of s compared with the rate of change of t at
that instant.

Infinitesimals and Limits


10. A shorter method of arriving at this conclusion without using specified intervals was devised by Newton. He suggested that
a small increase in any quantity like s might be indicated by a special symbol s (delta s) which has no specified size but
represents a minutely small change in s. A similar small change in t would be denoted by t, and in x by x, etc.
11. Fig 3 shows a section of the curve
s = 12 + 10t t 2
PM represents the distance covered, s, at time OM (t). QN represents the distance (s + s) covered in time ON (t + t). In both
cases s and t are very small. Thus the graph shown is a greatly magnified portion of a very small arc of the curve. The
gradient of the chord PQ represents the average speed between time t and (t + t) and can be measured as QR/PR = s/t.

Page 121 : Wed Jan 02 02:41:55 2002

Introduction to Calculus

8- 1- 5- 1

DO NOT DISTRIBUTE

AP 3456

12. Since Q is on the curve:


s + s = 12 + 10(t + t) (t + t )
= 12 + 10t + 10t t 2 2tt (t)

(1)

and for P
s = 12 + 10t t 2 (2)
Subtracting (2) from (1)
s = 10t 2tt (t)

Dividing by t
s/t = 10 2t t
Thus a formula has been derived for calculating the average speed over any period of time however small.

8-1-5-1 Fig 3 Section of the Curve s = 12 + 10t - t

For example between t = 3 and t = 3.0001 secs ie t = 0.0001 secs:


s/t = 10 6 0.0001 = 3.9999 m/sec

Page 122 : Wed Jan 02 02:41:56 2002

Introduction to Calculus

8- 1- 5- 1

DO NOT DISTRIBUTE

AP 3456

If a value of 0.000001 secs had been used in the formula then s/t would have been 3.999999.
13. Thus it will be seen that in the expression:
s/t = 10 2t t
if the value of t is allowed to grow smaller and smaller, ie approaches zero, then s/t approaches the value 10 2t. This is
written as:
s
Lim
= 10 2t
t!0 t
This is read as 'The limit of delta s by delta t as delta t tends to zero equals 10 2t'.
14. If the value of 3 secs is now substituted into this expression then 10 2t = 4 which is the value for the gradient that was
deduced earlier, ie the actual speed at the instant of t = 3 secs. To indicate that this is the actual gradient at an instant
then:
ds
s
Lim
is replaced by
dt
t!0 t
Thus in summary if an object is moving so that the distance s metres covered in time t seconds is a function of the time, (ie s =
2
f(t) and f(t) = 12 + 10t t ), then its speed at any time, t, is equal to the gradient of the tangent to the distance/time graph at time
t and is defined by ds/dt which may be calculated from the expression ds/dt = 10 2t. The notation ds/dt therefore, is a measure
of the rate at which s is changing compared with the rate at which t is changing at an instant of time 't'.

Differentiation
Chapter 2 - Differentiation
Gradients
1. In Chapter 1 it was shown that given the relationship between distance gone and time it was possible to find the rate of
change of distance with time, ie speed, either over a specified interval or at a particular instant, by determining the gradient of
the appropriate chord or tangent.
2.

The technique is not restricted to the distance/time problem but has a general applicability whenever one parameter is
n
changing in response to changes in another. The nature of the change may be expressed in polynomial terms eg y = ax , in
kx
trigonometric terms eg y = p sin x, in exponential terms eg y = a , in logarithmic terms eg y = c log x, or indeed in many other
forms.
3.

3
As an example Fig 1 shows the graph representing the function f(x) = x .

Let P (x,y,) be any point on the curve. NM represents a small change x, in x, and RQ represents the consequential change y,
in y. Thus Q is the point (x + x, y + y). As both P and Q lie on the line then:
for P
y = x3

(1)

and for Q

Page 123 : Wed Jan 02 02:41:56 2002

Differentiation

8- 1- 5- 2

DO NOT DISTRIBUTE

AP 3456

y + y = (x + x)

= x3 + 3x2 x +3x( x) + x3

(2)

Subtracting (2) (1)


2

y = 3x2 x + 3x(x) + x3
Dividing by

x
y
= 3x2 + 3xx + x2
x

Then by definition the gradient of the tangent to the curve:


dy
lim y
= x&!0
dx
x

8-1-5-2 Fig 1 f(x) = x

dy
= 3x2 as the 3xx and x2 terms are eliminated.
ie dx

4.

Clearly the gradient varies from point to point, as can be seen from the graph. The way in which the gradient varies is given

dy
2
by dx ie by the function 3x .

2
Thus the value of the gradient can be determined by substituting the appropriate value of x into the expression 3x .

Page 124 : Wed Jan 02 02:41:56 2002

Differentiation

8- 1- 5- 2

DO NOT DISTRIBUTE

AP 3456

eg
gradient at x = 0 = 0
gradient at x = 1 = 3
gradient at x = 2 = 12

The Differential Coefficient (Derivative)


dy

5. dx is called the differential coefficient of y with respect to x, or the derivative of y with respect to x. The process of
dy
dy
d
d
(y) in which dx is an operator like the
obtaining dx is called differentiating y with respect to x. Sometimes dx is written dx
p
symbol , and means simply 'the derivative of'. An alternative way of expressing the notion uses the functional notation. Thus
if f(x) is a function of x, then f' (x) means the derivative of that function of x.

Differentials
dy

dy

Although yx is a quotient (ie it stand for y x), dx is not. Strictly speaking dx ought to be regarded as a single symbol,
2
however it is often convenient to treat it as if were a quotient. For example having found that when y = x , dy
dx = 2x, this result
2
2
could be written as dy = 2xdx or d(x ) = 2xdx. In this notation, dy, dx and d(x ) are best regarded as infinitesimal increments in
2
y, x, and x and are called the the differentials of those quantities.
6.

The General Case


7.

3
The procedure used to find the differential coefficient of y = x can be extended to the general case ie:
y = f(x)

(3)

If (x,y) is any point on the curve then an increase in x, ie x causes an increase in y, ie y, such that:
y + y = f(x + x)

(4)

Subtracting (4) (3):


y = f(x + x) f(x)
Hence
y f(x + x) f(x)
=
x
x
and so
dy
lim f(x + x) f(x)
= x!0
dx
x

Successive Differentiation

Page 125 : Wed Jan 02 02:41:57 2002

Differentiation

8- 1- 5- 2

DO NOT DISTRIBUTE

AP 3456

3
8. When y = x was differentiated the result was
dy
= 3x2
dx
d
which is itself a function of x. This of course can itself be differentiated thus: dx

d
The expression dx
d2 y


dy
dx

= 6x is usually written as


dy
dx

= 6x

d2y
or as f "(x). Similarly the result of further differentiation of 6x would be
dx 2

d4 y

written as dx3 = 6, and dx4 = 0


Recalling the distance/time example, it will be remembered that differentiating the expression resulted in the rate of change of
distance with time ie speed. Differentiating the result will yield the rate of change of speed with time ie acceleration.

Standard Derivatives
9. The method used above is known as differentiation from first principles. In practice there are are rules to allow the
derivatives of certain functions to be determined without recourse to formal working. For example it can be shown that:
d
n
dx ax

= naxn1 where a and n are constants which may be positive or negative, fractions or integers.

10. Sum of Terms. Where f(x) is the sum of a number of terms eg:
y = ax3 + bx2 + cx + d
where a, b, c, and d are constants, then f '(x) is the sum of the derivatives of each individual term. Thus in this case:
2

f' (x) = 3ax + 2bx + c


Note that the derivative of a constant = 0.
11. Product Rule. It may be that it becomes necessary to differentiate an expression which is the product of two functions of
the same variable, eg:
2

y = (x + 1)(x 3)
In this case it would be possible to multiply out the expression without much difficulty and then differentiate the sum of the
terms as outlined in paragraph 10. However this may not be convenient, especially if there are several functions rather than just
two. In this situation the product rule can be used.
12. Let one function be u and the other v so y = uv. Then by the product rule:
dy udv vdu
=
+
dx
dx
dx
ie the result is the first function multiplied by the derivative of the second function, plus the second function multiplied by the
derivative of the first function. Thus in the example:
2

y = (x + 1)(x 3)

Page 126 : Wed Jan 02 02:41:57 2002

Differentiation

8- 1- 5- 2

DO NOT DISTRIBUTE

AP 3456

2
Let (x + 1) = u and (x 3) = v
Then:
dy udv vdu
=
+
dx
dx
dx
Thus:
dy
2
= (x + 1) 2x + (x 3).1
dx

= 2x2 + 2x + x2 3
= 3x2 + 2x 3
= u'vw + uv'w +uvw'
13. This method can be extended to cover more than two factors. Thus, d(uvw)
dx
For example:
2

y = (x + 1)(3x + 4)(x x 1)
dy
2
3
3
2
2
= 1.(3x + 4)(x x 1) + 6x(x + 1) (x x 1) + (3x 1)(x + 1)(3x + 4).
dx
14. Function of a Function - The Chain Rule. Consider an expression such as:
2

y = (3x + 2)

2
2
Here, y is a function of (3x + 2) and (3x + 2) is a function of x. As with the product rule, in some cases the function may be
simplified into the sum of several functions which may then be differentiated individually. However this will be tedious if the
power is greater than, say, 2. The chain rule can be used to solve this problem as follows:
Let
3x2 + 2 = u
Then
y = u2

._.

dy
= 2u
du

and as

Page 127 : Wed Jan 02 02:41:57 2002

Differentiation

8- 1- 5- 2

DO NOT DISTRIBUTE

AP 3456

u = 3x2 + 2

du
= 6x
dx

._.

dy du dy
: =
= 2u x 6x
du dx dx

Then substituting back for u:


dy
2
= 2(3x + 2) 6x
dx

= (6x + 4).6x

=36x3 + 24x
15. Quotient of 2 Functions. If y = u/v where u and v are functions in x then:
udv
dy vdu
= dx 2 dx
v
dx
As an example consider the function:
2

y=

(x + 1)
(3x + 2)

Then
2

u = (x + 1)

._.

du
= 2x
dx

and
v = (3x + 2)

._.

dv
=3
dx

Thus

Page 128 : Wed Jan 02 02:41:58 2002

Differentiation

8- 1- 5- 2

DO NOT DISTRIBUTE

AP 3456

dy (3x + 2).2x (x + 1).3


=
2
dx
(3x + 2)

6x2 + 4x 3x2 3
(3x + 2)

3x2 + 4x 3
(3x + 2)

Summary
16. A summary of the results derived above together with other standard derivatives is shown in Table 1.

Table 1 - Standard Differential Coefficients


Type of Function

Standard Type

Standard Differential
Coefficient

y = f(x)

dy
dx

Algebraic

y = axn

naxn1

Trigonometric

y = sin x
y = cos x

cos x

Logarithmic

y = loge x

Exponential

y = ekx

Sum of two or more


functions

u+v

Quotient of two
functions
Function of a function

u
v

F[f(x)]

Reduce the power by 1


and multiply by the
original power.

sin x
sec2 x

y = tan x

Product of two functions uv

Comments

1
x

kekx

du
dx

+ dv
dx

udv
+ vdu
dx
dx

v du
dx

Multiply the original


function by the
differential coefficient
of its index.
The differential
coefficient of a sum is
the sum of the
differential
coefficients.
Multiply each function
by the differential
coefficient of the other
and add the results.

u dv
dx
v2

df
d[f(x)]

df(x)
dx

Use the chain rule.

Introduction to Calculus
Chapter 3 - Integration
PRINCIPLES OF INTEGRATION

Page 129 : Wed Jan 02 02:41:58 2002

Introduction to Calculus

8- 1- 5- 3

DO NOT DISTRIBUTE

AP 3456

Introduction
1.

Differentiating means solving the problem:

given y = f(x),

nd

dy
dx

The reverse problem is:

given

dy
= f(x),
dx

nd y

This reverse process is called integration.

Indefinite Integrals
2.

Consider the following:


dy
= axn
dx

(1)

From the discussion on differentiation in Chapter 2 it will be apparent that the function whose derivative with respect to x is ax
is of the form:
y = bxn+1

(2)

If (2) is differentiated with respect to x the result is:


dy
n
= (n + 1) bx
dx
Comparing (1) and (3), they will be the same if (n + 1)b = a, ie if b =

y=

axn+1
n+1

(3)

a
n + 1.

Substituting this value of b in (2):

(4)

Although (4) is certainly one solution to the problem it is not a unique solution. Since the derivative of any constant is zero, the
derivative of (4) will be unchanged if any constant, c, is added to the right-hand side of the equation. Therefore, the general
solution is :

y=

axn+1
+c
n+1

(5)

Because of the presence of the arbitrary constant, c, (5) is known as the indefinite integral of (1).
3. Integration Symbol. It is convenient to have a symbol to denote the indefinite integral of a function, thus (5) may be
rewritten as:

Page 130 : Wed Jan 02 02:41:58 2002

Introduction to Calculus

8- 1- 5- 3

DO NOT DISTRIBUTE

AP 3456

Z
In this notation the

axn dx =

axn+1
+ c (6)
n+1

and the dx are used as brackets to denote that everything between them is to be integrated with respect to
R
n
x. The quantity so bracketed is known as the integrand. Thus ax is the integrand of axn dx, while the right-hand side of (6)
is the integral. Formula (6) holds for all values of n, integral and fractional, positive and negative, with the single exception of n
= 1. This case will be dealt with later.
4. The Constant of Integration. When a function is differentiated, the result represents the gradient of the graph of that
function. Consequently as the integration process is the reverse of differentiation, an integral represents a function with the
given gradient. Recalling that the equation of a straight line is y = mx + c it will be remembered that the coefficient of x, ie m,
equates to the gradient of the line. There is, however, an infinite family of parallel lines, all with the same gradient, m, varying
in the value of the constant c. Thus the knowledge of the gradient is insufficient to describe uniquely a particular straight line.
So when a function is integrated an arbitrary constant must be included to take account of the infinite number of 'parallel'
functions. As an example consider the following function:

y=

2xdx

To perform the integration the power of x has to be increased by 1, and then the integrand has to be divided by the new power.
Finally the arbitrary constant

8-1-5-3 Fig 1 Graphs of y = x + c

must be added. Thus:


y = x2 +c
2
Fig 1 shows the graphs of y = x + c for a variety of values of c.

Page 131 : Wed Jan 02 02:41:59 2002

Introduction to Calculus

8- 1- 5- 3

DO NOT DISTRIBUTE

AP 3456

For any given value of x all of the curves have the same gradient, ie they all satisfy the condition dy/dx = 2x. In order to
determine which graph is the solution to the particular problem then more information is required. For example it may be known
2
that y = 3 when x = 0, hence 3 = 0 + c, thus c = 3. Therefore, the required solution is y = x + 3.
2
5. As a practical example suppose that a body moves with an acceleration of 3ms and it is necessary to find an expression
for its velocity after t seconds.
dv
=3
dt
ie

v=

3dt = 3t + c

The reason that this is an inadequate description of the velocity is that although acceleration information was provided, no
information was given concerning the initial velocity of the body. Therefore, no definite value for the velocity at any given time
1
can be deduced. If the initial velocity was, say, 2 ms
then the velocity at any time, t, becomes 3t + 2 ms (ie c = 2).

Standard Integrals
6. Just as with differentiation, there are a number of standard integrals which are used. In general an unfamiliar expression
must be converted into a standard form, or a variation or a combination of standard forms, before the integration can be
accomplished. Similar rules to those used in differentiating apply in integrating; thus the integral of a sum of a set of functions
becomes the sum of the integrals of each individual function. If an integrand has a constant then this is taken out before
integration is performed, thus:
Z

3x2 dx = 3

x2 dx

Usually products or quotients must be simplified into simple functions before integration can take place. Thus, for example:
Z

(x + 2)(x 3)dx

x3
x2

6x + c
3
2

(x x 6)dx

and
Z

Page 132 : Wed Jan 02 02:41:59 2002

x(x 1)
1

x2

x2 x
1

x2

dx

dx

Introduction to Calculus

8- 1- 5- 3

DO NOT DISTRIBUTE

AP 3456

x2 x

7.

1
2

dx

2x2
2x2

+c
5
3

n
In paragraph 3 it was shown that the integral of a simple function in x, ax is given by:
axn+1
n+1

However it was stated that this formula did not apply when n = 1. This is because the denominator of the expression would
become 1 + 1 = 0 and dividing by zero has no real meaning. The paradox can be resolved by recalling that differentiating

loge x yields

1
x

Therefore the converse means:


Z
8.

1
dx = loge x
x

A list of the more common standard integrals is shown in Table 1.

Definite Integrals
9. Fig 2 shows an arc, ZB, of the curve y = f(x). P is the point (x, y) and Q the point [(x + x), (y + y)], where x and y are
very small quantities.
10. The elemental strip LPQM is part of the area (A) between the curve and the axes of x and y. Let the area, LPQM, be
denoted as A. The mean height of the arc PQ lies between y and y + y.
Suppose it equals y + Ky where K < 1

Table 1 - Some Standard Integrals


dy
dx

xn
cos x
sin x
sec2x
ekx
1
x

y=

dy
dx

dx

x n+1
n + 1

sin x
cos x
tan x
e kx
k

Comments
Increase the index by 1, and divide by the new
index.
Inverse of differentiation.
Inverse of differentiation.
Write down the function ekx and divide it by the
differential coefficient of the index of e.

logex

Then the area LPQM = A = x(y + Ky)

Page 133 : Wed Jan 02 02:41:59 2002

Introduction to Calculus

8- 1- 5- 3

DO NOT DISTRIBUTE

AP 3456

. _ . (A/x) = y + Ky
In the limit as x 0 then (A/x) and y 0

._.

dA
= y and
dx

dA
dx =
dx

y dx

ie
A=

y dx =

f (x) dx

Suppose
Z

f(x)dx = F(x) + C

Then
A = F(x) + C
11. If it is required to find the area between the curve, the x axis, and the ordinates at x = a and x = b, ie the area DCBA in Fig 3
then:
For the ordinate at
x=b

A1 = F(b) + C

x=a

A2 = F(a) + C

and at

Subtracting these, DCBA = A1 A2 = F(b) F(a)


This is written as:
Z

b
a

f (x) dx = F(b) F(a)

or in words 'the integral f(x)dx between the limits x = a and x = b'. 'a' and 'b' are called respectively the lower and upper limits of
the value of x. Notice that the constant of integration has disappeared; this is because it would appear in both F(b) and in F(a)
and is thus cancelled in the subtraction. Because such integrals are evaluated between defined limits, they are called definite
integrals.
12. In summary the method is as follows:
a. Integrate the function, omitting the constant of integration.
b. Substitute the value of the upper limit for x; repeat for the value of the lower limit. Subtract the results to give F(b)
F(a).

Page 134 : Wed Jan 02 02:41:59 2002

Introduction to Calculus

8- 1- 5- 3

DO NOT DISTRIBUTE

AP 3456

Example
Z

2
1

x4
x dx =
4
3

2
1

24
4

14
4

= 3:75

8-1-5-3 Fig 2 Area Under the Curve of f(x)

8-1-5-3 Fig 3 Area Under the Curve - The definite Integral

Page 135 : Wed Jan 02 02:42:00 2002

Introduction to Calculus

8- 1- 5- 3

DO NOT DISTRIBUTE

AP 3456

APPROXIMATE NUMERICAL INTEGRATION


Introduction
13. The integration process is often complex, but provided that the requirement is for a numerical answer to definite integration
then there are a number of methods available which yield approximate results, most of which are suitable for computer
implementation if necessary. Two such methods, the trapezoidal rule and Simpson's rule, will be described.

Trapezoidal Rule
14. In the trapezoidal rule the x axis is divided into equal intervals, h, and the top of each arc section is approximated by the
chord as in Fig 4. Thus a series of trapezia are formed whose top coordinates have the values y1, y2, etc.
15. The area of the first trapezium is
1
h(y1 + y2 )
2
and so
Z

Xn
X1

1
1
1
ydx = h(y1 + y2 )+ h(y2 + y3 )+......+ h(yn1 + yn )
2
2
2

1
1
= h y1 + y2 + y3 +......+ yn1 + yn
2
2

In general a small value of h will give a better solution than a large one, but the best procedure is to repeat the computation with
successively smaller values of h until two results agree within the required level of precision.
R1
1
16. Example. Compute 0.5 x 2 dx using the trapezoidal rule.

Page 136 : Wed Jan 02 02:42:00 2002

Introduction to Calculus

8- 1- 5- 3

DO NOT DISTRIBUTE

AP 3456

First compute with h = 0.1, say:

x1 = 0.5

y1

0.3535

x2 = 0.6

y2

0.7746

x3 = 0.7

y3

0.8367

x4 = 0.8

y4

0.8944

x5 = 0.9

y5

0.9487

x6 = 1.0

y6

0.5000

Sum

4.3079
Z

1
0.5

x2 dx = 0.1 4.3079 = 0.43079

Repeat with h = 0.05

x1 = 0.50

y1

= 0.3535

x2 = 0.55

y2

= 0.7416

x3 = 0.60

y3

= 0.7746

x4 = 0.65

y4

= 0.8062

x5 = 0.70

y5

= 0.8367

x6 = 0.75

y6

= 0.8660

x7 = 0.80

y7

= 0.8944

x8 = 0.85

y8

= 0.9220

x9 = 0.90

y9

= 0.9487

x10 = 0.95

y10

= 0.9747

x11 = 1.00

y11 = 0.5000
Sum

= 8.6184
Z

1
2

dx = 0.05 x 8.6184 = 0.43092

0.5

To three decimal places the result is 0.431 which compares very well with the correct value of 0.43096.

8-1-5-3 Fig 4 The Trapezoidal Rule

Page 137 : Wed Jan 02 02:42:00 2002

Introduction to Calculus

8- 1- 5- 3

DO NOT DISTRIBUTE

AP 3456

Simpson's Rule
17. In the trapezoidal rule the curve y = f(x) is approximated by a series of straight lines. It can, however, be approximated by
any suitable curve and in the case of Simpson's rule, a parabola is used. Rather than joining pairs of points, a parabola is traced
through three points on the line as shown in Fig 5.
18. The result for an integration interval divided into 2 parts with 3 ordinates is:
Z
19. Example. Compute

R1

0.5

1
2

x3
x1

1
ydx = h(y1 + 4y2 + y3 )
3

dx using Simpson's rule with h = 0.25.

x1 = 0.5

y1 = 0.7071

x2 = 0.75

y2 = 0.8660

x3 = 1.00

y3 = 1.0000

and the integral is given by:


1
0.25(0.7071 + 4 0.8660 + 1.0000) = 0.4309
3
which is a better result than that given by the trapezoidal rule with h = 0.10.
20. Simpson's rule will usually give a more accurate result than the trapezoidal rule for the same interval, h, but it is often
necessary to sub-divide the curve into more than one set of three ordinates. A different parabola is fitted over each section. For
example, in Fig 6, seven ordinates are used, the parabolas are P1, P2, P3 and the integral is given by:
1
1
1
h(y1 + 4y2 + y3 ) + h(y3 + 4y4 + y5 ) + h(y5 + 4y6 + y7 )
3
3
3

1
h(y1 + 4y2 + 2y3 + 4y4 + 2y5 + 4y6 + y7 )
3

The principle is capable of extension to any ODD number of ordinates.

Page 138 : Wed Jan 02 02:42:00 2002

Introduction to Calculus

8- 1- 5- 3

DO NOT DISTRIBUTE

AP 3456

8-1-5-3 Fig 5 Simpson's Rule - Fitting a Parabola Through Three Points

8-1-5-3 Fig 6 Simpson's Rule - Seven Ordinates

Introduction to Statistics
Chapter 1 - The Scope of Statistical Method

Page 139 : Wed Jan 02 02:42:01 2002

Introduction to Statistics

8- 1- 6- 1

DO NOT DISTRIBUTE

AP 3456

Introduction
1. The word statistics is used in two distinct ways. It is used to mean either sets of figures, usually tabulated, in which sense it
is short for statistical data, or to mean the methods whereby the significant details may be extracted from such sets of figures. In
this sense it is short for statistical method, and it is with this meaning that this Section is concerned.
2. A good definition of the subject is: a body of methods for making wise decisions in the face of uncertainty. Defined in this
way the subject may be regarded as an extension of the idea of common sense, which is the name each person gives to his own
method of making wise decisions in everyday matters. The fact that the answers provided by common sense on the one hand
and by statistical method on the other often seem to be poles apart, is attributable either to the inadequacies of common sense or
to an incorrect use of statistical method.
3. As a broad generalization it may be stated that statistics takes over from common sense where the complexity of the problem
warrants it, and where the quantities involved can be expressed in the form of numbers. If these conditions are fulfilled then the
use of statistical method will give an economy of effort and a precision which is unobtainable in any other way.
4. There are very few aspects of human activity in which uncertainty does not play a part, so that the potential uses of
statistical method are very large in number. In general, practical problems have solutions which are more or less probable, and
probability theory forms the basis of statistics. An understanding of at least the elementary ideas of probability is, therefore, a
prerequisite for the understanding of statistical method.
5. It is important to notice that the word, "wise" appears in the definition of statistics, and not the word "right". That the latter
word is inadmissible follows, of course, from the fact that we are accepting uncertainty as a basic ingredient of the problem. The
wise decision that is made will be based on the most probable occurrence, but the most probable occurrence is not bound to
occur. Our decisions, therefore, will sometimes turn out to be wrong, no matter how elegant the mathematics used in the
solution of the problem, and this fact must be accepted.
6. Quite frequently decisions will prove to be wrong because they were based on inadequate data, and the point must be made
that statistical analysis does not bring anything out of the data that is not already there. Statistics provides an objective way of
testing the data and of obtaining answers free from personal prejudice and preconceived notions. The use of statistical method,
in other words, makes it possible to interpret correctly the influence of chance in the evidence available.
7. It is clearly important that any experiment or series of trials should be designed to provide relevant evidence in sufficient
quantity to do what is required. The only safe way to ensure this is to bring the statistician into the enquiry from the very
beginning, so that he will not only analyse the data but in fact will also state what data should be collected. Far too often the
statistician is called in too late, so that the investigator is faced with the invidious choice of either giving incomplete answers or
of repeating all or part of the investigation in order to obtain adequate data. This procedure is likely to be far more costly in the
long run than a properly planned attack on the problem in the first place. The best results are invariably obtained by the
statistician and subject specialist working together from the beginning of the enquiry.

The General Method of Statistics


8. Although there are many distinct statistical techniques, they all have certain features in common. The following remarks are
of general application.
9. The Presentation of Data. The raw data of statistics consists of a more or less haphazard collection of numerical data. An
important first step before any statistical analysis can be started is to tabulate the data in some way which is meaningful for the
investigation in hand. Suitable tabulation will, in fact, probably suggest the profitable line of attack, a process which may be
aided by a graphical presentation of the data.
10. Population. Any collection of data which is to be analysed by a statistical process represents a finite selection of values
from a practically if not theoretically, infinite population. In this context, the word population is part of statistical jargon, and
refers to the totality of values which it is possible to conceive within the restrictions laid down for the data. It is important to
note that the population may be one of people, of measurements, of bombs dropped on a target under specified conditions, or of
any data whatsoever that may be given numerical values.
11. Sample. The concept of population is always to some extent an abstraction, since it is not possible to obtain access to all
members of it (if it were, statistical analysis would be unnecessary). The data that is available is a sample taken from the
population being studied. A problem will always be posed in terms of populations and not of samples. The solution to the
problem, however, must proceed through an analysis of samples. For this to be a valid procedure it is clearly necessary that the
sample should be a fair representation of the population that is being studied. It may then be assumed that parameters calculated
from the sample are reliable estimates of the corresponding parameters of the population, and that conclusions based on the
sample will be valid for the population also.

Page 140 : Wed Jan 02 02:42:01 2002

Introduction to Statistics

8- 1- 6- 1

DO NOT DISTRIBUTE

AP 3456

12. Sampling. It will be clear from para 11 that the technique of sampling, whereby the sample to be used in the investigation
is to be chosen, plays a vital part in statistical work. Every collection of data is a fair sample from some population, but the
important thing is to ensure that the collection selected is a fair sample of a specified population. This is by no means an easy
thing to ensure. Broadly, two different sampling techniques are used, the choice in a particular case being dependent upon the
extent of one's knowledge of the system studied. If a great deal is known about the population it may be possible to select a
sample which conforms to the same pattern as the population; this technique is used extensively in public opinion polls. If it is
not possible to do this, or if one is uncertain about the completeness of one's knowledge, then a technique of random selection, in
which every member of the population has an equal chance of being selected for the sample, will be used. This has the extreme
merit, if done correctly, of removing unsuspected bias, which may arise in any subjective method of sampling. A common
method of ensuring randomness in the sample is through the use of special tables of random numbers, which have been
thoroughly tested and found free from bias.
13. Statistical Significance. It is not unusual in statistical work to find that when a random sample is taken with the aim of
providing a hypothesis it turns out that the sample data does not wholly support that hypothesis. The difference could be due to:
a. The hypothesis being wrong, or
b. The sample being biased.
Clearly, tests are needed to determine which is the more likely possibility. Tests of significance are very important to
statisticians but are outside the scope of the chapter.
14. Proof and Disproof. It is perhaps clear already from the foregoing discussion that statistical method never provides a
definite proof of any hypothesis, though it may provide very strong evidence indeed in favour of it. Any process which involves
extrapolation from sample to population, and usually from past to future time, must involve uncertainty, and no matter how
improbable an event there is always the possibility that it will happen. The gibe that "you can prove anything by statistics"
shows a complete misunderstanding of the methods of statistical inference. Nothing is "proved" or "disproved" by statistics.

Some Uses of Statistics


15. The point has been made that the opportunities for the application of statistical method are virtually unlimited, so that any
list of uses will necessarily be incomplete. The following selection of topics, however, gives a good idea of the great scope of
statistical method.
16. The Measurement of the Inexplicable. Many problems are concerned with quantities which do not take unique values under
the conditions which it is possible to specify. In such cases residual uncertainties may be very important, and it is necessary to
be able to assess their likely magnitude. Probability theory provides a way of doing this. We may distinguish two important
types of problem:
a. Those in which the uncertainties give rise to a random departure from a true value or from a desired result. We are here
concerned with the determination of errors.
b. Other cases, in which there is no true or desired value, but in which random variability is an essential feature of the
system. Measurements of biological quantities and of human attainment come into this category.
The emphasis here is on the use of statistics to calculate meaningful parameters which may be used to describe the population.
Some distribution of values is obtained, and in practice the essential features of this distribution are established by comparison
with a similar theoretical distribution.
17. The Identification of Important Factors. The problem here is to determine which of several factors that might be expected
to affect performance do in fact have a significant effect. In some cases the important factors may be seen without the aid of
statistical analysis. In other cases, because of interactions between factors, the issue may be by no means clear, and statistical
techniques then provide a way of separating the effects of individual factors and thus establishing their relative importance.

Some Misuses of Statistics


18. No discussion of the uses of statistics would be complete unless accompanied by a discussion of misuses, for abuse of
statistical method occurs all too frequently. It is because of this that statistics are viewed with such suspicion by so many
people.
19. In general we may lay down the principle that from a given set of data one set of conclusions relative to a particular issue

Page 141 : Wed Jan 02 02:42:01 2002

Introduction to Statistics

8- 1- 6- 1

DO NOT DISTRIBUTE

AP 3456

will be more likely than any other set. But the fact is that other conclusions, incompatible with the first, will frequently be
drawn, often because a certain conclusion is desired and sometimes, as in misleading advertising, because no other conclusion is
acceptable. It is often difficult or impossible for a person not intimately engaged in an investigation to trace invalid reasoning,
and hence the belief arises that a judicious use of statistics will enable conflicting conclusions to be drawn from the same set of
data. Again the possibilities are legion, but the following examples illustrate some of them.
20. Deceptive Presentation. Cases of presentation with intent to deceive are common features of everyday life, and very often
take the form of the omission of relevant data. Thus a poster supporting an anti-immunization campaign announced boldly that
in a certain period of time 5,000 cases of diphtheria occurred among immunized children. The public was expected to infer that
immunization failed, but the poster did not disclose the highly relevant information that in the same period 75,000 cases
occurred among non-immunized children, nor that non-immunized children were 6 times as likely to get diphtheria and 30 times
as likely to die from it.
21. Sampling Errors. The importance of sampling has already been emphasized, and it must be obvious that bad sampling will
lead to invalid conclusions. For example, taking the announcement of births from the columns of the "Times" gave a sex ratio of
1,089 males per 1,000 females. However the Registrar General found during the same period a ratio of 1,050 females per 1,000
males. The reason for the discrepancy is no doubt that the sample from the "Times" was not a fair sample of the whole
population, perhaps because parents are more inclined to announce the births of their sons and heirs than of their daughters.
22. False Correlations. The technique of correlation is very easily misapplied, and correlation between two variables should
not be sought unless there are reasons, stemming from knowledge of the system studied, to expect it. It is easy to think of
variables which, though unrelated, will show strong correlation, often because both happen to vary in a certain way with the
passage of time. Such correlations are called nonsense correlations.
23. Statistics versus Experience. It is the subject expert who will make the decision, making full use both of his own specialist
knowledge and of the results of the statistical analysis. If a statistical inference runs counter to what the specialist expects then
he should query it. The inference may be correct, but a little probing will be most worthwhile.

Introduction to Statistics
Chapter 2 - Descriptive Statistics
Introduction
1. Statistics is concerned with the mathematical analysis of numerical data. The numerical data is in the form of a set of
observations of the variable (or variables) under consideration. A variable (or variate) is a quantity which assumes different
measurable values, eg height, weight, examination marks, temperature, intelligence, length of life, etc. Any set of observations
of the variable is considered, for statistical purpose, as a sample drawn from some infinitely large population. When each and
every member of a population has an equal chance of being selected for a sample, the sample is called a random sample. The
principle task of statistical analysis is to deduce the properties of the population from those of a random sample. In this chapter
we discuss how samples and populations can be described; in particular we will look at averages and the bunching of the
samples and population about these averages.

AVERAGES
Types of Average
2. The Arithmetic Mean. The arithmetic mean is commonly referred to as "the average". The mean is the sum of all the values
of a variable divided by the number of variables. The algebraic form of expression is:

X=

X1 + X2 +XN
X
=
N
N

where
X = the mean

Page 142 : Wed Jan 02 02:42:01 2002

Introduction to Statistics

8- 1- 6- 2

DO NOT DISTRIBUTE

AP 3456

X1 .... XN = the values of the dierent variables in a distribution

N = the number of variables

= the symbol instructing the addition of all the values


Example: A sample consists of the data 5,8,9,6,12,14.
The mean is

X=

5+8+9+6+12+14 54
=
=9
6
6

3. The Median. If the sample observations are arranged in order from the smallest to the largest the median is the middle
observation. If there are two middle observations, as in the case of an even number of observations, the median is halfway
between them.
Examples:
a. Given sample 1,14,9,6,12. Arranged in order 1,6,9,12,14, the median is 9.
b. Given sample 20,7,11,10,13,17. Arranged in order 7,10,11,13,17,20, the median is 12.
4. The Mode. The mode is the observation which occurs most frequently in a distribution. If each observation occurs the same
number of times there is no mode. If two or more observations occur the same number of times, and more frequently than any
other observations, then the sample is said to be multi-modal.
a. Given sample 16,13,18,16,17,16 the mode is 16.
b. Given sample 4,7,4,9,3,7 then the modes are 4 and 7.
c. Given sample 3,7,12,11,16,20 there is no mode.
The mode is seldom used but has been included for completeness.

MEASURES OF DISPERSION
Introduction
5. Knowledge of the average of a distribution provides no information about whether figures in a distribution are clustered
together or well spread out. For example two groups of students have examination marks of 64%, 66%, 70% and 80%, for the
first group and 37%, 61%, 88% and 94% for the second group. Both groups have a mean mark of 70% but the marks of the
second group have a much greater dispersion than those of the first group. Clearly it would be useful to have a way of
measuring dispersion (or variance) and expressing it as a simple figure. The most commonly used measures are:
a. Range
b. Quartile Deviation
c. Standard Deviation
d. CEP (used in particular applications)

Page 143 : Wed Jan 02 02:42:01 2002

Introduction to Statistics

8- 1- 6- 2

DO NOT DISTRIBUTE

AP 3456

Range
6. Range is the difference between the highest and lowest values. Unfortunately range is too much influenced by extreme
values so that one value differing widely from the remainder in a group could give a distorted picture of the distribution. Range
also fails to indicate the clustering of values into particular groups or areas.

Quartile Deviation
7. Quartiles are the values of the items one quarter and three quarters of the way through a distribution. If the top and bottom
quarters are cut off extreme values are discarded and a major disadvantage of range as a measure of dispersion is avoided.
Quartile Deviation =
Third Quartile - First Quartile
2
As with Range, this method fails to indicate clustering.

Standard Deviation
8. Standard deviation is the most important of the measures of dispersion. The standard deviation () is found by adding the
square of the deviations of the individual values from the mean of the distribution, dividing the sum by the number of items in
the distribution, and then finding the square root of the quotient. (Scientific calculators include a facility for finding and other
statistical parameters from sets of data.)

(x -x)
N

The more that values of individual items differ from the mean, the greater will be the square of these differences, giving rise to a
large measure of dispersion. The main disadvantage of using standard deviation as a measure of dispersion therefore is that it
can give disproportionate weight to extreme values because it squares the deviations eg a value twice as far from the mean as
2
another is weighted by a factor of 4, (2 ). Nevertheless standard deviation is the best and most useful measure of dispersion
within a set of observations.

Circular Error Probable (CEP)


9. A term commonly encountered in weapon effects planning is Circular Error Probable (CEP). The CEP is the radius of the
circle, centered on the mean impact point, within which 50% of weapons fall. Strictly, CEP should only be used when the
distribution of impacts is known to be circular, but this restriction is often ignored in practice. As a rough guide, the CEP is
approximately 1.18 times the standard deviation of the linear errors in weapon impacts, although this convention is only
justifiable if the errors in range and deflection are normally distributed (see paras 25-27).

FREQUENCY DISTRIBUTION
Introduction
10. It is very difficult to learn anything by examining unordered and unclassified data. Table 1 displays such raw data.

Table 1 - Weekly Mileages Recorded by 60 Salesmen


504
623
507
477
429
535

592
548
477
522
491
517

671
467
501
627
497
612

498
487
562
556
510
491

601
399
555
622
603
432

Page 144 : Wed Jan 02 02:42:01 2002

532
482
642
521
547
508

Introduction to Statistics

8- 1- 6- 2

DO NOT DISTRIBUTE

577
562
688
417

444
685
492
512

556
432
486
563

AP 3456

639
642
467
612

444
562
474
375

723
662
433
578

Raw data is simply a list of data as received, in this case from sixty individual salesmen. Little of use can be learned from data
presented in this form.

Ungrouped Frequency Distribution


11. The first step in making the raw data more meaningful is to list the figures in order from the lowest mileage to the highest.
At the same time it may be convenient to annotate those figures that occur more than once with the frequency of occurrence.
The result is the distribution shown in table 2. The symbol for frequency is '' The sum of the frequencies () must equal the
total number of items making up the raw data.

Table 2 - Ungrouped Frequency Distribution


Mileage Mileage Mileage Mileage Mileage
375
1
482
1
508
1
555
1
622
1
399
1
486
1
510
1
556
2
623
1
417
1
487
1
512
1
562
3
627
1
429
1
491
1
517
1
563
1
639
1
432
2
492
1
521
1
577
1
642
2
433
1
497
1
522
1
578
1
662
1
444
2
498
1
532
1
592
1
671
1
467
2
501
1
535
2
601
1
685
1
477
2
507
1
548
1
612
2
723
1

Grouped Frequency Distribution


12. Though the ungrouped frequency distribution is an improvement in the presentation there are still too many figures for the
mind to grasp the information effectively. More simplification is necessary in order to compress the data. This can be done by
grouping the figures and showing the frequency of the group occurrence. The result is shown in table 3.

Table 3 - Grouped Frequency Distribution


Mileage
375 - under 425
425 - under 475
475 - under 525
525 - under 575
575 - under 625
625 - under 675
675 - under 725
Total Records

3
9
18
12
9
6
3
60

13. Effect of Grouping. As a result of grouping it is possible to see from table 3 that mileages cluster around 475-525.
Although grouping highlights the pattern of a distribution it does lead to the loss of information about where in the group the 18
occurrences lie. The increased significance of the table has therefore been paid for but the cost is worthwhile. The loss of
information also means that calculations made from grouped frequency distribution cannot be exact.

Class Limits
14. The boundaries of a class are called the class limits. Care must be taken in deciding class limits to ensure that there is no
overlapping of classes or gaps between them. For example if the class limits in table 3 had been 375-425 and 425-450 which
group would a mileage of 425 have gone into? Likewise if the class limits had been 375-424 and 425-449 a mileage of 424

Page 145 : Wed Jan 02 02:42:02 2002

Introduction to Statistics

8- 1- 6- 2

DO NOT DISTRIBUTE

AP 3456

would have no group to fit into.


15. Discrete and Continuous Data. In defining class limits it should be remembered that discrete data increases in jumps. For
instance data relating to the number of children in families will be in whole units because 1 and 2 children are not possible.
Continuous data however may include fractions.
16. Class Intervals. Class intervals define the width of a class. If the class intervals are equal the distribution is said to be an
equal class interval distribution.
17. Unequal Class Intervals. Some data is such that if equal class intervals were used a very few classes would contain all the
occurrences whilst the majority would be empty. An example of this is the distribution of salaries using a class interval of
1000. In this situation the class intervals should be arranged so that over-full classes are subdivided and near empty ones
grouped together.
18. Choice of Classes. The construction of a grouped frequency distribution always involves a decision as to what classes to
use. The following suggestions should be borne in mind:
a. Class intervals should be equal wherever possible.
b. Class intervals of 5, 10 or multiples of 10 are more convenient than, say, 7 or 11.
c. Classes should be chosen so that occurrences within the classes tend to balance around the mid point.

Construction of a Grouped Frequency Distribution


19. To construct a grouped frequency distribution directly from raw data the following steps should be taken:
a. Pick out the highest and lowest figures (375 and 723) and on the basis of these decide upon the list and the classes.
b. Take each figure in the raw data and insert a check mark (1) against the class into which it falls.
c. Total the check marks to find the frequency of each class. See table 4.

GRAPHS OF OBSERVATION
The Histogram
20. A Histogram is a graph of a frequency distribution. It is shown in Fig 1 and is constructed as follows:
a. The horizontal axis is a continuous scale running from one end of the distribution to the other. The axis should be
labelled with the name of the variable and the unit of the measurement.

Table 4 - Direct Construction of Grouped Frequency Distribution


Class

Check Marks

375 - under 425


425 - under 475
475 - under 525
525 - under
575 - under
625 - under
675 - under

575
625
675
725

111
1111 1111
1111 1111
1111 111
1111 1111 11
1111 1111
1111 1
111
Total Records

3
9
18
12
9
6
3
60

8.1.6.2. Fig 1 Histogram of Data from Table 3

Page 146 : Wed Jan 02 02:42:02 2002

Introduction to Statistics

8- 1- 6- 2

DO NOT DISTRIBUTE

AP 3456

b. For each class in the distribution a vertical column is constructed with its base extending from one class limit to the
other and its area proportional to the frequency of the class.

The Frequency Polygon


21. If the mid points of the tops of the blocks of a frequency histogram are joined by straight lines, the resulting figure is a
frequency polygon as illustrated in Fig 2. The area enclosed by the horizontal axis, the polygon and any two ordinates
represents approximately the number of observations in the corresponding range. The total area enclosed by the polygon
represents the total number of observations if the frequency scale is used, and unity if the probability scale is used.

8-1-6-2 Fig 2 Frequency Polygon

The Frequency Curve


22. If the number of observations is greatly increased and the size of the class interval is correspondingly reduced, then the
frequency histogram and the frequency polygon will tend to a smooth curve as in Fig 3. By increasing the number of
observations indefinitely the whole population instead of just a sample will have been analysed. Thus the smooth curve obtained
by this process represents the distribution of the complete population in the same way as the histogram or frequency polygon
represents the distribution of the sample

8-1-6-2 Fig 3 Frequency Curve

Page 147 : Wed Jan 02 02:42:02 2002

Introduction to Statistics

8- 1- 6- 2

DO NOT DISTRIBUTE

AP 3456

23. If the original sample is random, and large enough, then it is unlikely that the smooth curve will be very different from
drawing a smooth curve through the mid points of the histogram blocks. Such a smooth curve is known as a frequency curve if
the frequency scale is used or a probability curve if the probability scale is used. The area under the curve between any two
ordinates represents, as accurately as any estimate can, the number of observations within the corresponding range, while the
same area for a probability curve represents the probability that a single observation, taken at random from the complete
population, will lie within the corresponding range. The latter idea is more useful since it applies to the whole population and is
no longer confined to the sample.
24. Histograms frequently display a pattern in which there is a high column in the centre with decreasing columns spread
symmetrically either side. If the class interval is small enough the frequency curve looks like the cross section of a bell. This
pattern occurs frequently in statistical work.

The Normal Curve of Distribution


25. The task of handling statistical data can be simplified if a mathematical curve can be found approximating to that which
would be produced by plotting the actual data on a graph. By substituting the mathematical graph for the real one it is possible
to make calculations to reveal facts about the distribution of the raw data which would otherwise have been difficult to
determine. The normal curve of distribution satisfies this requirement. It has the following features:
a. It is symmetrical
b. It is bell shaped
c. Its mean lies at the peak of the curve
d. The two tails continuously approach, but never cross, the horizontal axis (x). See Fig 4.
The formula for the curve can be ignored because any mathematical data relating to it can be found in mathematical tables.

8-1-6-2 Fig 4 Normal Curve of Distribution

Page 148 : Wed Jan 02 02:42:03 2002

Introduction to Statistics

8- 1- 6- 2

DO NOT DISTRIBUTE

AP 3456

26. Areas Below the Normal Curve of Distribution. The 'y' axis in Fig 4 is at the peak of the curve and passes through the
mean value of the distribution. The total area beneath the curve is unity, representing the fact that a random variable is certain to
lie between + and -. If 1 lengths are now marked off on the x axis from the mean value of the curve, the area enclosed by
the curve and the 1 boundaries is 68.26% of the total area. Tables are available to give the areas lying under the curve between
any two lines on the x axis designated in terms of .
27. Areas and Frequencies. Areas under the normal distribution curve are proportional to frequencies. An area of 95% of the
total area is equivalent to a frequency figure which indicates that 95% of all occurrences lie between the two 2 values.
Similarly 99.75% of all occurences fall within the 3 values (see Fig 5). The height of the curve at a particular point has no
practical relevance - areas under the curve must always be used to give the frequency of occurence in a specified range of values.

8-1-6-2 Fig 5 Approximate Areas Beneath the Normal Curve

Page 149 : Wed Jan 02 02:42:03 2002

Introduction to Statistics

8- 1- 6- 2

DO NOT DISTRIBUTE

AP 3456

Introduction to Statistics
Chapter 3 - Elementary Theory of Probability
Mathematical Definition of Probability
1. Suppose that a certain experiment can have just n possible results, all of them equally likely (eg in drawing a card from a
pack there are just 52 possible results and all of them are equally likely) and suppose that a certain event can occur as m of these
n possible results (eg picking an ace can occur as 4 of the 52 possible results). Then the probability of the event occuring in any
one experiment is defined as m/n
2. Thus the probability of drawing an ace is 4/52 or 1/13. We may also say that the chances are 1 in 13 or that the odds are 12
to 1 against.
3. If the probability that an event will occur is m/n, then the probability that it will fail to occur is (n-m)/n = 1 - m/n. Thus if p
is the probability of success and q the probability of failure, we have
q = 1 p
4.

or p = 1q

or p+q = 1

It is certain that an event will either occur or fail to occur. The probability of either a success or a failure is n/n = 1. Hence

Page 150 : Wed Jan 02 02:42:03 2002

Introduction to Statistics

8- 1- 6- 3

DO NOT DISTRIBUTE

AP 3456

probability = 1 denotes certainty and, similarly, probability = 0 denotes impossibility.

Interdependence of Events
5.
a. Mutually Exclusive Events. Two events are mutually exclusive if the occurrence of one prevents the occurrence of the
other. If a die is thrown, the occurrence of a 6 prevents the occurrence of a 5.
b. Independent Events. Two events are independent if the occurrence or non-occurrence of one has no effect on the
probability of the occurrence of the other. If two dice are thrown, the result of one throw has no effect on the result of the
other.
c. Dependent Events. Two events are dependent when the occurrence or non-occurrence of one event has some effect on
the probability of occurrence of the other. If two cards are drawn from a pack, the probability that the second card is an ace
is 3/51 or 4/51 depending on whether the first card was or was not an ace.
Notice that mutually exclusive events occur as alternative results of the same experiment whereas independent and dependent
events occur as the simultaneous or consecutive results of different experiments.

Calculation of Probabilities
6. Theorem I - Addition of Probabilities. If two events are mutually exclusive, then the probability of either one or the other
event occurring is the sum of the probabilities of the individual events.
Proof. Let the probability of E1 be m1/n and let the probability of E2 be m2/n. Then out of n equally likely events, m1 are E1
and m2 are E2, ie m1+m2 events out of n are either E1 or E2. Hence the probability of either E1 or E2 occurring is:
m1 + m 2
m
m
= 1+ 2
n
n
n
7. Theorem II - Multiplication of Probabilities
a. Independent Events. If two events are independent, then the probability of both one and the other happening is the
product of the probabilities of the individual events.
b. Dependent Events. If two events are dependent, then the probability of both the first event and the second event
happening is the product of the probability of the first event and the conditional probability of the second event on the
assumption that the first event has happened.

Proof. Let the probability of E1 be m1/n1 and let the probability of E2 be m2/n2. If E1 and E2 are independent, then the
probability of E2 is independent of the success or failure of E1, but if they are dependent then the probability m2/n2 is to be
regarded as the conditional probability of E2 on the assumption that E1 has happened. Then the total number of possible results
of the two experiments together is n1n2. Of these possible results, E1 and E2 can occur together in m1m2 ways. Hence
probability of both E1 and E2 occurring is:
m1 m2
m
m
= 1 2
n1 n2
n1
n2
8.

Conclusion both Theorem I and Theorem II extend to any number of events.

Example 1. A card is drawn from a pack. What is the probability of it being either an ace, the king of clubs or a red queen?
4

The probability of drawing an ace is 52

Page 151 : Wed Jan 02 02:42:03 2002

Introduction to Statistics

8- 1- 6- 3

DO NOT DISTRIBUTE

AP 3456

The probability of drawing the king of clubs is 52


2

The probability of drawing a red queen is 52


Hence the required probability is:
1
2
7
4
+
+
=
52 52 52 52
Example 2. Two dice are thrown and a card is drawn from a pack. What is the probability that both dice will show sixes and
that the card will be the ace of spades?
1

The probability of a six in one throw is 6 and the probability of drawing the ace of spades is 52
Hence the required probability is:
1
1 1 1

=
6 6 52 1872
Example 3. Two dice are thrown. What is the probability of the total throw being 10? The possible successes are (4, 6), (5, 5),
(6,4). The result of one die is independent of the result of the other.

Probability of (4, 6) is

1
1 1
=
6 6 36

Probability of (5, 5) is

1
1 1
=
6 6 36

Probability of (6, 4) is

1
1 1
=
6 6 36

Any combination of the pair excludes every other combination. Hence the required probability is:
1
1
1
1
+
+
=
36 36 36 12
Example 4. Two cards are drawn from a pack. What is the probability that they will both be aces?
4

The probability that the first card is an ace is 52 and the conditional probability that the second card shall also be an ace is 51
Hence the required probability is:
1
1
1

=
13 17 221
Example 5. Five balls are drawn from a bag containing 6 white balls and 4 black balls. What is the probability that 3 white balls
and 2 black balls are drawn?
Although the rules for combining probabilities are important, it sometimes pays to work from first principles, ie direct from the
mathematical definition of probability as in this example.

Page 152 : Wed Jan 02 02:42:03 2002

Introduction to Statistics

8- 1- 6- 3

DO NOT DISTRIBUTE

5 balls can be selected from 10 in

AP 3456

10
5

ways, that is in

10.9.8.7.6 2.9.7.2
=
= 2.9.7.2 ways.
1.2.3.4.5
1
This is the total number of possible selections.

6
3 white balls can be selected from 6 in
ways, ie
3
6.5.4
= 5.4 ways.
1.2.3
2 black balls can be selected from 4 in


4
ways, ie
2
4.3
= 2.3 ways.
1.2

Hence the number of ways in which 3 white balls and 2 black balls can be selected is 5.4.2.3, from which the required
probability is:
5.4.2.3 10
=
2.9.7.2 21

Limitations of Mathematical Definition


9. It is not always possible to use the mathematical definition of probability because the definition requires that the experiment
should have a finite number of possible different results. In practice there are often an infinite number of possible results (eg in
dropping a bomb on a target). Moreover, the definition depends on the results being equally likely, which to a certain extent
begs the question unless we can satisfy ourselves intuitively that the results are equally likely. Thus when the number of
possible results is infinite or when we cannot be sure that they are all equally likely, we cannot use the mathematical definition
and instead we use the following one.

Statistical Definition of Probability


10. Suppose that a certain experiment is carried out n times and that a certain event occurs as m of the n results. Then the
probability of the event occurring as the result of any one experiment is defined as:
p = lim

n!1

m
n

In practice, we cannot let n but instead we take n to be as large as we conveniently can. The resulting value obtained for the
probability is then the best available estimate in cases where the mathematical definition cannot be used.
11. It can be shown that there is little theoretical difference between the two definitions and that therefore, the theorems proved
on the basis of the mathematical definition still hold for probabilities obtained statistically.
Example 6. The operational requirement for a guided weapon demands that the weapon should have a reliability of 90%. If the
weapon can be broken down into 120 functionally-tested components of equal complexity and reliability, determine the
reliability demanded from each component.
The reliability of a weapon or a component is the probability that the weapon or component will be completely serviceable. In
this case the statistical definition of probability clearly applies. Now if RC is the reliability of a component, then the probability

Page 153 : Wed Jan 02 02:42:04 2002

Introduction to Statistics

8- 1- 6- 3

DO NOT DISTRIBUTE

AP 3456

that 120 such components will all be serviceable together may be obtained using the Multiplication Theorem for independent
events.
0.90 = R 120
C or RC =

120

0.90 = 0.9991

Thus each component must have a reliability of 99.91 %.

Probability of At Least One Success


12. It is important to distinguish between:
a. The probability of at least one success.
b. The probability of one success only
Suppose it is required to get a minimum of one hit on a target. It is not sufficient to calculate the probability of one hit only
because this would exclude the possibility of 2, 3 or more hits which must also count as successes. The criterion of success is at
least one hit.
Example 7. Three ballistic missiles are launched at a certain target. From the statistical analysis of performance trials of the
1
missile, it is estimated that the probability of hitting the target with a single missile is 6 . Calculate the chance of scoring:
a. One hit only.
b. At least one hit.
For a: The chance of a hit with the first missile but not with the other two is:
1 5 5
25
=
6 6 6 216
The chance of a hit with the second missile but not with the other two is:
25
5 1 5
=
6 6 6 216
The chance of a hit with the third missile but not with the other two is:
5 5 1 25

6 6 6 216
Hence the chance of one hit only is:
25
25
75
25
+
+
=
216 216 216 216
For b: In the same way, the chance of two hits only is:
15
1 1 5
3=
6 6 6
216
and the chance of three hits is:

Page 154 : Wed Jan 02 02:42:04 2002

Introduction to Statistics

8- 1- 6- 3

DO NOT DISTRIBUTE

AP 3456

1
1 1 1

6 6 6 216
Hence the chance of at least one hit is:
15
1
91
75
+
+
=
216 216 216 216
1

Alternatively, since the chance of a hit with one missile is 6 the chance of missing with one missile is 116 = 56. Thus the chance
3
of missing with all three missiles is 56 = 125
216, and the chance of failing to miss with all three missiles, that is the chance of at
least one hit is:
1

125
91
=
216 216

To generalize, let the chance of success in one attempt = p, then the chance of failure in one attempt = 1 p, ie the chance of
n
n
failure in n attempts = (1 p) , and finally the chance of at least one success in n attempts = 1 (1 p) .
1

Example 8. If the probability of obtaining a hit with a single missile is assessed as 20 , how many missiles must be launched to
give a 75% chance of at least one hit?
If n is the number of missiles which must be launched we require that :

1
1
20

= 0.25

ie

19
20

= 0.75

or 0.95n = 0.25
Taking logarithms,
n log 0.95 = log 0.25

._. n =

log 0.25
log 0.95

= 27
Example 9. The data given below refer to an interceptor fighter armed with 2 air-to-air guided weapons. Determine the overall
effectiveness of the weapon system.

Aircraft serviceability

Page 155 : Wed Jan 02 02:42:04 2002

- 0.90

Introduction to Statistics

8- 1- 6- 3

DO NOT DISTRIBUTE

Aircraft reliability in flight


Missile reliability
Missile lethality

AP 3456

- 0.80
- 0.70
- 0.50

The probability that the aircraft will be both serviceable and reliable in flight, and therefore able to deliver the weapon is:
0.90 0.80 = 0.72
The probability that a single weapon will inflict the required damage on the target is:
0.70 0.50 = 0.35
Thus the probability that at least one weapon will inflict the required damage is:
2

1 (1 0.35) = 0.58
ie the overall effectiveness is:
0.72 0.58 = 0.42 or 42%

Reliability
13. The accurate assessment of the reliability of complex and expensive systems or equipment can be a vitally important factor
in planning the purchase or deployment of resources. In order to quantify reliability for analysis it is necessary to be able to
attach a numerical value to it. Reliability is defined as the probability that an item will not fail during a given period of time.
The probability of an item not failing is denoted by p. The probability of an item failing is denoted by q. It follows that p+q = 1.
It is important that when a figure for reliability is quoted the time period to which it relates should also be quoted. It should be
noted that reliability is a probability and is therefore expressed as a fraction of 1 or a percentage.
14. Combination. The overall reliability (R) of a combination of components may be calculated on the assumption that the
quantities p and q are independent of the other components in the system. This being so, probabilities must be combined by the
multiplication rule to give the probability that all will occur together.
a. Components in Series.
- - - (1) - - - (2) - - - ..... - - - (n) - - -

The system will survive only if all the components survive.


. _ . R = p1 p2 ......pn
b. Components in Parallel

Page 156 : Wed Jan 02 02:42:04 2002

Introduction to Statistics

8- 1- 6- 3

DO NOT DISTRIBUTE

AP 3456

The system will fail only if all the components fail.

. _ . 1 R = q1 q2 ......qn

or R = 1 q1 q2 ......qn

or R = 1 (1 p1 )(1 p2 )......(1 pn).


A parallel system may be referred to as a redundant system.
c. Components may be combined partly in series and partly in parallel.

These systems may be reduced to a system of units in series by obtaining the overall reliability of each parallel branch. Thus in
the above example, if R23 is the overall reliability of the parallel components 2 and 3
R23 = 1 q2 q3

Page 157 : Wed Jan 02 02:42:05 2002

Introduction to Statistics

8- 1- 6- 3

DO NOT DISTRIBUTE

AP 3456

and the system reduces to


- - - (1) - - - (R23) - - - (4)
whence R = P1 x R23 x p4
= p1 p4 [1 (1 p2 )(1 p3 )]
15. Where redundant components are used on a mutually exclusive basis either one is used or another, and in this case the
addition rule for combining the probabilities must be used. For example a navigation system consists of mode 1 (reliability p1)
and mode 2 (p2) which will only be switched in by a switch of reliability p3 if mode 1 fails. The system will survive only if
mode 1 survives or if mode 1 fails and the switch operates and mode 2 survives. In terms of reliabilities the switch and mode 2
are only called upon to survive if mode 1 fails.
. _ . R = p1 + 1(1 p1 )(p3 )(p2 )
The term (1 p1)(p3)(p2) is the reliability contribution of the standby equipment. The overall reliability of the system can be
worked out by the series/parallel calculation

giving R = 1 (1 p1)(1 p3 p2)


= p 1 + p 2 p 3 p 1 p2 p3

PHYSICS
Heat
Chapter 1 - The Nature of Heat
Temperature and Heat
1. Heat is a form of energy possessed by a body by virtue of its molecular agitation. The heat content of an object is not
measured simply by its temperature; heat content is also a function of mass. One unit of heat is the kilocalorie (kcal), and may
be defined as the quantity of heat required to raise the temperature of 1 kilogram of pure water from 14.5 C to 15.5C at a
pressure of 1 atmosphere. The temperature range needs to be specified, as the quantity of heat required to raise the temperature
by 1C depends slightly on which C is chosen.
2.

Because heat is a form of energy it may equally, and perhaps more properly, be expressed in the SI unit of energy, the joule

Page 158 : Wed Jan 02 02:42:05 2002

Heat

8- 2- 1- 1

DO NOT DISTRIBUTE

AP 3456

(J). Indeed, by international agreement the kilocalorie is now defined as 4186.8 joules.

Specific Heat
3. Materials other than water require different quantities of heat to change their temperature by 1C. All materials may thus be
attributed a value, known as the specific heat, which reflects this variation and is defined as the amount of heat required to raise
the temperature of 1 kilogram of the substance by 1C. This may be expressed in the equation:

Q
where Q
m
c
t

=
=
=
=

mct joules
quantity of heat
mass of substance (kgs)
specific heat (J kg-1 C-1 )

change in temperature (C)

4. The value of specific heat depends upon the external conditions under which the heat is applied. Two variables are
normally taken into consideration leading to two values, one at constant pressure and one at constant volume. In the case of
solids and liquids which are generally heated at constant pressure, then only the constant pressure value is normally quoted, and
in any case the difference between the two values is negligible for all normal purposes. However in the case of gases the two
specific heats are quite different.

Change of State
5. When a material changes state from a solid to a liquid or from a liquid to a gas, or vice versa, then energy must either be
added to the substance or be released from the substance. For example, in order to change 1 kg of ice into water, approximately
3
335 x 10 J of heat need to be added. During the change of state the temperature does not rise, ie 1 kg of ice at 0C changes to
3
1 kg of water at 0C. Conversely to freeze 1 kg of water then approximately 335 x 10 J of heat need to be removed. The heat
which is required to change the state of a substance, without any temperature change, is known as latent heat. Where the change
is between solid and liquid it is known as the latent heat of fusion; where the change is between liquid and gas it is known as
latent heat of vaporization. In both cases the values quoted refer to 1 kg of the substance at the normal melting and boiling
points. (Heat energy which causes a change of temperature without giving rise to a change of state is defined as sensible heat.)
6. Supercooling. If a liquid is cooled slowly and is kept motionless, its temperature can be reduced to well below its normal
freezing point. This is known as supercooling. A supercooled liquid is in an unstable state and any disturbance will cause some
of the liquid to solidify, thereby releasing latent heat. The temperature of the supercooled liquid is raised to its freezing point by
the release of this latent heat and the normal process of solidification takes place.

Heat Transfer
7.

Heat may be transferred from one place to another by three mechanisms:


a. Conduction. If one end of, say, a metal rod is heated, then the atoms and electrons at that end will acquire higher kinetic
and potential energy than those in other parts of the rod. In random collisions these energetic atoms and electrons will
transfer energy to their neighbours which in turn will become more energetic, collide with their neighbours and transfer
some of their energy. Thus in this way the thermal energy which was applied at one end of the rod will be transferred along
the rod. This process is known as conduction.
b. Convection. Convection is the transfer of heat from one place to another occasioned by the movement of the heated
substance. Convection only occurs in liquids and gases.
c. Radiation. In radiation the heat is transferred in the form of electromagnetic waves. The intervening medium plays no
part in the process and indeed radiation can take place in a vacuum. Heat is radiated more efficiently by dull surfaces and
dark colours than by polished ones and light colours; thus the most efficient radiating surface is matt black.
Correspondingly, radiated heat energy is most efficiently absorbed by matt black surfaces, and most efficiently reflected by
light shiny ones.

Heat

Page 159 : Wed Jan 02 02:42:05 2002

Heat

8- 2- 1- 2

DO NOT DISTRIBUTE

AP 3456

Chapter 2 - Temperature and Expansion


The Concept of Temperature
1. Temperature may be defined as the degree of hotness of a body measured according to some fixed scale. The sense of touch
readily distinguishes between hot and cold bodies and thus between higher and lower temperatures. Temperature may also be
regarded in another way; if two bodies are placed in thermal contact, then heat will flow from the body at higher temperature to
that at a lower temperature.
2. The direction in which the heat flows does not depend on the quantity of heat in either body. For example, the water in a
tank may be at a lower temperature than a hot soldering iron, but owing to its greater volume it may contain a greater quantity of
heat. If the soldering iron is immersed in the water, heat will pass from the iron to the water. Once the temperatures are equal
no more heat will be transferred.

Temperature Scales
3. The temperature of an object is measured by a thermometer which makes use of those properties of liquids, gases and other
substances, which vary continuously with temperature and are independent of previous treatment. Common methods involve the
measurement of the expansion of solids, liquids or gases as they are heated, (liquid-in-glass thermometers and bi-metallic strips),
on the measurement of gas pressure as a gas is heated under constant volume (constant volume gas thermometer), and in the
change in colour of emitted light as an object is heated (optical pyrometer).
4. All of these thermometers require to be calibrated according to a defined scale. Historically two fixed temperature points
have been defined; zero degrees Celsius (originally centigrade), and 100 degrees Celsius, or their Fahrenheit equivalents of 32F
and 212F.
5. The lower point was defined by the temperature at which pure water and ice exist in thermal equilibrium; the upper point at
which pure water and steam exist in thermal equilibrium. Both points were defined at a pressure of 1 atmosphere (1.01325 x
5 -2
10 Nm ).
6. Temperature scales are now defined relative to the Kelvin scale. The temperature at which all molecular agitation due to
heat energy ceases is defined as Absolute Zero. This corresponds to a temperature of approximately 273 on the Celsius scale
and it is assigned a value of 0 K. The size of the degree Kelvin is identical to that of the degree Celsius, thus the conventional
fixed points of 0C and 100C are 273 K and 373 K respectively. Conventionally, the degrees sign is omitted when referring to
Kelvin temperatures but it may still be seen to be used in many instances.
7. Temperature Conversion. Conversion of temperature values between Celsius and Fahrenheit scales may be accomplished
using the two following equations:

F=

C=

9
C + 32
5

5
( F 32)
9

Temperature Measurement
8.

Temperature is measured using a thermometer, or for high temperatures, a pyrometer; various types are described below.

9. Liquid-in-Glass Thermometers. The liquid-in-glass thermometer is the simplest instrument for the measurement of
temperature. It depends on the fact that as the temperature of a liquid changes, so the volume of that liquid expands or contracts.
For most purposes mercury is used as the liquid, however as it solidifies at 39C there is a limit to the lower end of its useful
range. Alcohol may be used for lower temperature measurement, but it is limited at the higher temperatures as it has a boiling
point of 78C.
10. Bi-metallic Strip. The bi-metallic strip thermometer consists of two strips of dissimilar metals welded together, and usually
formed into a helix. One end of the helix is fixed while the other is free to rotate. As the two metals have different coefficients
of expansion, they will expand or contract at different rates as temperature changes. This will be manifested in the helix coiling
and uncoiling in response to temperature changes. A pointer is attached to the free end of the helix and this moves over a

Page 160 : Wed Jan 02 02:42:05 2002

Heat

8- 2- 1- 2

DO NOT DISTRIBUTE

AP 3456

graduated scale. This type of thermometer is frequently used for outside air temperature measurement. A non-coiled version is
often used as the sensing element in thermostatic controls in which the bending of the bi-metallic strip makes or breaks an
electrical circuit.
11. Pyrometers. Conventional thermometers are not suitable for the measurement of very high temperatures, such as those
found in jet pipes. The instrument used for high temperature measurement is called a pyrometer and three types are described
below:
a. Thermocouple. The principle of operation of the thermocouple is illustrated in Fig 1. A and B are junctions of dissimilar
metals, G is a sensitive galvanometer. If the temperature of the two junctions is different, a current will flow from the iron
to the copper at the colder junction and from the copper to the iron at the hotter junction. The size of the current is
measured by the galvanometer; a higher current indicating a greater temperature difference between the two junctions. The
advantage of this system is that the cold junction and the galvanometer can be remote from the hot, sensing, junction. The
main disadvantage of the system is that it has to be calibrated; both to relate the current to the temperature difference and to
determine the cold junction temperature so that actual, rather than relative, temperatures can be determined. The two metals
used in the junctions can be varied to suit the temperature range to be measured. Thermocouples are typically used in the
measurement of jet pipe temperatures.

8-2-1-2 Fig 1 Thermocouple

b. Radiation Pyrometer. The radiation pyrometer relies on the principle that materials change colour and brightness as
they are heated. This type of instrument is typically used in kilns and furnaces. An electric filament is placed between the
eye of the observer and the bright interior of the kiln or furnace. A current passing through the filament causes it to glow
and the current is adjusted until the filament brightness matches that of the furnace interior. The current is measured with an
ammeter and this current is then related to the interior temperature. As with the thermocouple, this device must first be
calibrated against known temperature sources.
c. Resistance Wire. The electrical resistance pyrometer relies on the fact that the electrical resistance of materials varies
with temperature. The resistance of metals increases with temperature increases while the resistance of non-metals
decreases with temperature increases. Over moderate temperature ranges the resistance change is proportional to
temperature change.

The Behaviour of Gases


12. It can be shown that for a given amount of gas (say n moles), the pressure (p), the volume (V), and the temperature (T), of
the gas are related by the ideal gas equation:
pV = nRT where R is the universal gas constant.
13. This equation incorporates two laws as follows:
a. Boyle's Law. This law asserts that if the temperature of a gas is kept constant then the product of the pressure and the
volume remains constant as a given amount of gas is compressed or expanded;
ie pV = constant
b. Charles' Law. This law asserts that if the pressure of a gas is kept constant then the ratio of the volume to the

Page 161 : Wed Jan 02 02:42:05 2002

Heat

8- 2- 1- 2

DO NOT DISTRIBUTE

AP 3456

temperature remains constant as a given amount of gas is heated or cooled;

ie

V
= constant
T

This behaviour of gases is used as the basis for the two types of gas thermometer: the constant volume and the constant
pressure thermometers.

8-2-1-2 Fig 2 Isothermal and Adiabatic Changes

14. Isothermal and Adiabatic Changes. Fig 2 illustrates the two ways in which the pressure of a gas changes as the volume is
changed. The difference depends upon whether the temperature changes concurrently. If a gas is compressed slowly such that
there is time for the energy transferred to it to be dissipated through the container, then the temperature of the gas will remain
constant, and the change of state is said to be isothermal. On the other hand if no energy transfer between the gas and the
surroundings is permitted, then its temperature will rise and the change in state is termed adiabatic.

Expansion of Solids and Liquids


15. Solids expand as they are heated. The thermal expansion of a solid is usually best described by the increase in linear
dimension. The increment is directly proportional to the temperature change and to the original length. Thus:
L = LT
where
L = change of length

L = original length

Page 162 : Wed Jan 02 02:42:06 2002

Heat

8- 2- 1- 2

DO NOT DISTRIBUTE

AP 3456

T = change in temperature
The constant,, is called the coefficient of linear expansion. Its value for any material varies slightly with the initial
temperature.
16. The volumetric expansion of a soild can be expressed in an analogous equation:
V = VT
where
V = change in volume

V = original volume

T = change in temperature
and the constant, , is the coefficient of volume expansion. = 3
17. Most liquids expand when they are heated, so that their density reduces with increasing temperature. However, water
exhibits a somewhat unusual variation. From 0C to 4C the volume decreases, non-linearly, as temperature increases. Above
4C the volume increases with temperature. Thus water has its maximum density at 4C.

Electro Optics
Chapter 1 - The Nature of Light
Introduction
1. The true nature of light is a question which has taxed scientists for many generations. Most theories treat light as the
transport of energy, either as a wave motion or as a stream of particles. Each of these ideas can be used to explain certain
phenomena associated with light but neither can satisfactorily explain them all. For example the wave theory can explain why
crossed light beams do not scatter each other, whereas the particle concept would not allow this to happen. Conversely the wave
theory cannot be used to explain the photoelectric effect - only the particle theory is satisfactory. Neither idea really defines
what light actually is; rather each is a model which can be used to describe and predict the behaviour of light under a particular
set of circumstances. Thus it is necessary to choose the appropriate model for the task in hand. In general, where the behaviour
of light in motion is being studied, the wave model is more useful, while the particle model is to be preferred when studying the
interaction of light with other matter, eg absorption and emission.

The Wave Model


2. When a pebble is tossed into a pond it sets into motion the water particles with which it comes into contact. These particles
set neighbouring particles in motion and so on until the disturbance reaches the edge of the pond. In fact any particular particle
only oscillates vertically about a mean position. It does not itself move to the edge of the pond - only the disturbance moves
through the water. This is a typical characteristic of wave motion. The oscillations are at right angles to the direction of
propagation of the wave, and such a wave motion is known as a transverse wave. Light waves are also transverse waves, but
rather than representing the motion of particles in a medium, the wave represents the variations in electric and magnetic field
strength. Neither of these fields can exist without the other and they are interdependent. The fields are mutually perpendicular
and both are at right angles to the direction of wave motion. This arrangement is illustrated in Fig 1. Unlike water waves which
are constrained to move along the water surface, light waves can propagate in any direction and are not dependent upon the

Page 163 : Wed Jan 02 02:42:06 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

medium; indeed, they can move through a vacuum.

8-2-2-1 Fig 1 Light Wave

3. The waves generated by tossing a pebble into a pond are small compared with, for example, sea waves. Also whereas the
distance between the crests of the pond waves may be only a few centimetres, with sea waves the separation may be several
metres. Similar variations occur in light waves and these parameters are summarised in Fig. 2, where it will be seen that light
waves have a sinusoidal form. The distance between the subsequent crests of a wave (or between any other corresponding
points) is known as the wavelength and is usually given the symbol . The vertical size of the wave is measured from the mean
level and is known as the amplitude. The time taken for corresponding points on a wave to pass a fixed point is known as the
period, and the number of corresponding points passing in unit time is the frequency (f). The wave travels at a speed (c),
which depends upon the medium through which it is passing. In free space the speed of light is approximately
3 108 ms1 :From this it will be seen that speed, frequency and wavelength are related by the equation: C = f.

8-2-2-1 Fig 2 Wave Parameters Polarization

Polarization
4. In general the electric and magnetic fields of a light wave are free to vibrate in any of the infinite planes at right angles to the
direction of propagation. Any ordinary light source consists of the superposition of a number of plane waves each with a
random plane of vibration. Such a light beam is said to be unpolarized. If, however, the electric field is constrained to lie in a
particular plane then the light is said to be polarized. Polarization can be achieved by passing the light through a suitable
filter.

The Electromagnetic Spectrum


5. Light waves can be shown to have essentially the same characteristics as many other types of radiation such as radio waves,
microwaves and X-rays. Indeed all of these are examples of electomagnetic radiation, differing primarily in their frequencies.
The electromagnetic spectrum describes a large range of such waves, illustrated in Fig 3, in which visible light occupies only a
very small band.

Wave-fronts and Rays


6. From a point source light is propagated in all directions. Travelling out from the source, wave maxima will be encountered
at regular intervals, corresponding to the wavelength. Similarly at different positions but still with the same spacing, minima
will occur, and the same will apply for any intermediate value of amplitude. Lines can be drawn joining points of equal
amplitude, analogous to contour lines, and these are known as wave-fronts. As the radiation occurs in three dimensions these

Page 164 : Wed Jan 02 02:42:06 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

wave-fronts are in fact spherical surfaces, but it is often satisfactory to treat the radiation as if it were planar in which case the
wave-fronts reduce to circles. As a further practical simplification if the wave-front is at a large distance from the source and if
only a small sector is investigated then the wave-front may be approximated by a straight line.
7. The direction of propagation of light is at 90 to the wave-fronts and a line representing this direction is known as a ray.
Rays are often used in diagrams where only the direction of propagation is of concern. Wave-fronts and rays are illustrated in
Fig 4.

8-2-2-1 Fig 3 The Electromagnetic Spectrum

8-2-2-1 Fig 4 Wave-fronts and Rays

Superposition and Interference


8. When two or more sinusoidal waves act at the same place then the result can be found by adding algebraically the individual
amplitudes. Four examples of this principle of superposition are shown in Figs 5, 6, 7 and 8. In Fig 5 the two waves have the
same amplitude and frequency and are in phase, ie the maxima and minima of each wave occur at the same time and place. In
this case the waves reinforce each other, thus resulting in a doubling of the amplitude. This is known as constructive
interference. In Fig 6 the same two waves are completely out of phase. At every point the amplitudes have the same magnitude
but the opposite sign, and so cancel each other - a situation known as destructive interference. In Fig 7 three waves having the
same frequency but different phases add to produce another sinusoid with the same frequency. Fig 8 shows the general case
where two waves have different amplitudes and frequencies.

8-2-2-1 Fig 5 Two Sinusoidal Waves, Same Amplitude, Frequency and Phase

Page 165 : Wed Jan 02 02:42:06 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

8-2-2-1 Fig 6 Two Sinusoidal Waves, Same Amplitude and Frequency, 180 degrees Out of Phase

8-2-2-1 Fig 7 Three Sinusoidal Waves, Same Amplitude and Frequency, Different Phases

8-2-2-1 Fig 8 Two Sinusoidal Waves, Different Amplitude and Frequency

Page 166 : Wed Jan 02 02:42:07 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

9. If two close sources radiate light of the same frequency and if the two radiations are in phase then as shown in Fig 9 at some
points there will be constructive interference whilst at others there will be destructive interference. The result is a radially
symmetrical interference pattern.

Diffraction
10. It is a characteristic feature of waves that they will deflect around the edges of obstacles placed in their path and spread into
the shadow zone behind the obstacle. This effect can often be seen where for example water waves pass through a narrow
harbour entrance or impinge upon a breakwater. There is no complete 'shadow' behind the wall, rather the waves appear to bend
around the obstacle. This phenomena is known as diffraction. In general the amount of diffraction is related to the wavelength
and the size of the gap through which the waves must pass. The deviations of the wave are quite large when the size of the
obstacle or gap is of the same order as the wavelength. Light, behaving as a wave form, will also experience diffraction but as
the wavelength of light is very small the effect is only pronounced when the obstacle or gap is very small.
11. The effect of diffraction is closely associated with the ideas of constructive and destructive interference developed in
paragraphs 7 to 9. An insight into the effect can be gained by studying the passage of light through a pair of closely spaced
narrow slits.
12. Fig 10 shows plane waves arriving at a screen in which there are two narrow slits which are perpendicular to the page. The
two slits act as sources of light and therefore two series of circular wave-fronts can be constructed, one set from each slit. In the
diagram the wave-fronts represent the crests of the waves. As the slits were illuminated by the same incident light the light
leaving each slit has the same wavelength, amplitude and phase. The principle of superposition can be used to predict the effects
which will be observed. At point P crests of waves from each slit arrive simultaneously and therefore constructive interference
takes place and a reinforcement of amplitude will be seen. The same argument can be applied to any point where crests
intersect. Such points have been joined by bold lines in the diagram and strong waves would be expected to be seen radiating
along these lines. If the distance between point P and each slit is measured it will be seen to be different by one wavelength
(1). The same is true for any point along the bold line through P. Other lines of constructive interference will each demonstrate
a different value for the difference in distance and these values have been indicated in the diagram. In each case the value is an
integral multiple of a wavelength (n ).

8-2-2-1 Fig 9 Constructive and Destructive Interference

Page 167 : Wed Jan 02 02:42:07 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

8-2-2-1 Fig 10 Diffraction by Two Slits

13. Point Q corresponds to the point of intersection of a crest from one slit and a trough from the other and therefore destructive
interference will occur. The dashed lines represent the lines along which destructive interference occurs. In this case the
difference in distance from any point on a dashed line to each slit is an integral multiple of half a wavelength (n /2).
14. Points like P and Q represent the extremes of complete constructive interference and complete destructive interference. In
between these points will be points where the interference is partially constructive, ie the resultant amplitude is greater then zero
but less than twice the amplitude of each wave.
1
2
15. Consider now Fig 11a in which point R is a large distance from an opaque screen in which there are two slits S and S .
There is a small angle between the lines joining the slits to point R, however, if the point is sufficiently far away then this angle
becomes so small that it may be safely ignored and the lines can be considered to be parallel. This situation is relected in Fig 11b
which is a magnification of the area close to the screen in which the lines joining the slits to R are drawn parallel. The lines are
inclined at an angle to lines normal to the screen.

Page 168 : Wed Jan 02 02:42:07 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

8-2-2-1 Fig 11 Determination of Angles for Constructive and Destructive Interference

2
2
16. A line S M is drawn which is perpendicular to the lines from each slit to R. Point M and the slit S are the same distance
1
from R and the difference between distances 11 and 12 is equal to the distance between slit S and M.
17. It has been shown in paragraphs 11 and 12 that the factor determining whether constructive or destructive interference
occurs at R is the difference between the distances 11and 12. Thus it is now necessary to relate the distance S1 M to the angle .
angle + angle = 90

and

angle + angle = 90

thus

angle = angle

Now

sin = S1 M/d

thus

sin = S1 M/d

so

S 1 M = d sin

For constructive interference to occur the difference between the distances 11 and 12 must be equal to an integral number of
wavelengths so:
11 12 = S1M = d sin = n
or sin = n /d where n is any ineger.
This equation determines the angles at which constructive interference occurs in terms of wavelength and slit spacing. It can be
shown that the equivalent equation for destructive interference is:
1
sin = (n + ) /d
2
18. As the number of equally spaced slits in the screen (and hence interference sources) is increased, the destructive
interference regions widen, as shown in Fig 12. A screen with many equally spaced slits is known as a diffraction grating.

Page 169 : Wed Jan 02 02:42:07 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

Huygens' Principle
19. A knowledge of the manner in which a wave-front propagates is necessary in order to explain the phenomena of reflection
and refraction. In 1690 the Dutch physicist Huygens proposed the following method:
To find the change of position of a wave-front in a small interval, t, draw many small spheres of radius [wave speed] t with
centres on the old wave-front. The new wave-front is the surface of tangency to those spheres.

8-2-2-1 Fig 12 Diffraction Patterns

It should be remembered that this is only a model which enables predictions to be made, it is not meant to be a description of
reality. The small spheres employed in this construction are known as wavelets. Clearly most diagrams are constrained to
showing phenomena in two dimensions only in which case the spherical wavelets reduce to circles. Fig 13 shows how Huygens'
principle is used to predict the new position of a planar wave-front after a short interval.

8-2-2-1 Fig 13 Huygens' Principle of Wave-front Propagation

Page 170 : Wed Jan 02 02:42:08 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

Reflection
20. The case of a plane wave incident on a plane mirror is shown in Fig 14. Fig 14a shows the wave-fronts approaching a
reflecting surface. One edge of the leading wave-front is just touching the surface at point P. The situation a short time later is
shown in Fig 14b where some Huygens' wavelets have been constructed (the portion of the wavelets below the surface have
been omitted as irrelevant). The new wave-front touches the surface at point P'. To the right of P' the new wave-front is parallel
to the old wave-front as it has not yet been reflected. In order to find the position of the new wave-front to the left of P', a
straight line is drawn starting from P', and tangential to the wavelet centred on P (Huygens' Principle). This straight line
represents the part of the wave that has been reflected. The two right angled triangles are congruent as they have a common
side, PP', and the short sides, PQ and P'Q' are equal (being radii of the wavelets). Thus the angles and are equal. In studying
reflection it is usually more convenient to deal with rays rather than with wave-fronts, since they more readily show the direction
of propagation. Fig 15 shows the reflection of the rays corresponding to the wave-fronts of Fig 14c. In summary there are two
Laws of Reflection as follows:

8-2-2-1 Fig 14 Reflection Using Huygen's Principle

Page 171 : Wed Jan 02 02:42:08 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

a. The angle of incidence equals the angle of reflection.


b. The incident and reflected rays and the normal to the reflecting surface lie in the same plane.

8-2-2-1 Fig 15 Reflection of a Ray

Page 172 : Wed Jan 02 02:42:08 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

Refraction
21. Huygens' Principle can also be used to predict the behaviour of light when it is transmitted at a boundary between two
mediums rather than being reflected. Consider Fig 16. AB represents a wave-front arriving at such an interface at an angle i,
the point A arriving before point B. To find the position of the wave-front after a short time interval, t, Huygens' wavelets are
constructed emanating from points A and B. From point B, which can be assumed to be in air, the light travels at velocity v1,
and a wavelet can be drawn representing the time taken for the light to travel from B to B'. In the same time interval the light
from A is travelling in the transparent medium (glass say) at a slower velocity (v2). Thus the same time interval will correspond
to a smaller wavelet. The new wave-front is now drawn from point B' to be tangential to the wavelet centred on A. Thus the
light wave has been deviated as it changes from one medium to another in which the velocity of light is different. This
phenomenon is known as refraction.
22. From Fig 16:
BB0 = v1 t and AA 0 = v2 t

8-2-2-1 Fig 16 Refraction Using Huygens' Principle

Page 173 : Wed Jan 02 02:42:08 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

Thus

BB0
v
= 1 (1)
v2
AA0

sin i =

BB0
AB0

sin r =

AA0
AB0

and

Hence
sin i
BB0
=
sin r AA0
Nothing that i = i and r = r and comparing with equation (1):
sin i
v
= 1 (2)
sin r
v2
23. It is normal to express the velocities of light in the two media as fractions of c, the velocity of light in a vacuum.

Page 174 : Wed Jan 02 02:42:08 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

Hence v1 = c/n1

and v2 = c/n2
The numbers n1 and n2 are known as the refractive indices of medium 1 and medium 2 respectively and equation (2) may be
rewritten as:
sin i n2
(3)
=
sin r n1
If the incident wave is travelling in a vacuum, (or for nearly all practical purposes in air), then n1 = 1 and equation (3) reduces
to:
sin i
sin r

= n where n is the refractive index of the second medium.

This relationship between the angle of incidence, angle of refraction and the refractive indices of the media is known as Snell's
Law of refraction. When a wave passes from a medium of high velocity to one of lower velocity then it is refracted towards the
normal and conversely when passing from a 'slower' medium to a 'faster' medium it is refracted away from the normal. As with
reflection the two rays and the normal to the interface all lie in the same plane. It is also evident that there is a corresponding
change in wavelength.

Total Internal Reflection


24. A refracted ray cannot be deviated more than 90 from the normal to the interface. Thus when r = 90, sin i has its
maximum possible value and as sin r = 1:
sin i

v1
v2

The value of the angle obtained in this case is called the critical angle. At angles of incidence greater than the critical angle all
of the incident light is reflected at the boundary. This behaviour is known as total internal reflection. The phenomenon has a
number of practical applications, perhaps the most important of which in recent years has been the development of the optical
fibre in which the light is transmitted along the fibre experiencing total internal reflection when it impinges upon the fibre
walls.

Dispersion
25. Unlike reflection, refraction is frequency dependent. This is because the velocity of a wave in a medium changes as the
frequency of the wave changes. Thus the 'bending power' of a given material is dependent upon the frequency; an effect known
as dispersion. Thus for example if a ray of light containing a mix of frequencies is refracted by a medium then each of the
component frequencies, or colours, will emerge at a different angle. Traditionally a prism has been used to demonstrate this
effect and to analyse the component frequencies of a light source. Fig 17 illustrates the arrangement. The incident light ray
experiences refraction at each glass/air interface with those components at the red end of the spectrum experiencing less
refraction than those at the blue end.

8-2-2-1 Fig 17 Dispersion of Light by a Prism

Page 175 : Wed Jan 02 02:42:09 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

The Practice Model - The Photoelectric Effect


26. Although the wave model has provided a good description of many of the ways in which light behaves, it fails to explain
the photoelectric effect.

8-2-2-1 Fig 18 Apparatus for the Investigation of the Photoelectric Effect

Page 176 : Wed Jan 02 02:42:09 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

27. The photoelectric effect involves the conversion of light into electricity and is used in solar cells and photographic light
meters for example. A simple device to illustrate the effect is shown in Fig 18. Light from a lamp illuminates a metal electrode
enclosed in an evacuated tube. Electrons are ejected from this electrode, travel to the collecting electrode (A) and then flow
around the circuit in which an ammeter can measure the current. The kinetic energy of the ejected electrons can be determined
by applying a potential difference between the emitting and collecting electrodes using an adjustable source. With the polarity
as shown the collector exerts a repulsive force on the electrons. A potential can therefore be applied which will just stop the
flow of electrons from the emitter to the collector. This potential is known as the stopping voltage.
28. It can be shown experimentally that the kinetic energy of the electrons increases linearly with the frequency of the incident
light as shown in Fig 19. Below a certain frequency the light is incapable of ejecting electrons; this frequency, f1, in Fig 19 is
called the threshold frequency.
29. The energy of waves depends upon intensity and not frequency, and therefore the wave model would predict that the kinetic
energy of the ejected electrons would increase with increasing intensity. Thus the wave model is at variance with the
experimental result.
30. The particle model assumes that monochromatic light of frequency f comprises identical particles each carrying energy hf
where h is a constant known as Planck's constant. The particles of light (known as photons) collide with electrons in the metal
and the energy hf carried by a photon is transferred to an electron. When the frequency of the light is below the threshold
frequency the energy carried by the photon is insufficient to free even the most weakly bound electrons from the metal and all of
the photon's energy is converted into heat. Once the frequency exceeds the threshold frequency then the energy is sufficient to
free electrons from the metal and also to give the electrons some kinetic energy. The higher the frequency, the higher the photon
energy and thus the higher will be the ejected electron's kinetic energy.

8-2-2-1 Fig 19 Variation of Kinetic Energy with Frequency

Page 177 : Wed Jan 02 02:42:09 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

8-2-2-1 Fig 11a

8-2-2-1 Fig 11b

Page 178 : Wed Jan 02 02:42:09 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

8-2-2-1 Fig 14a

8-2-2-1 Fig 14b

8-2-2-1 Fig 14c

Page 179 : Wed Jan 02 02:42:09 2002

Electro Optics

8- 2- 2- 1

DO NOT DISTRIBUTE

AP 3456

Electro Optics
Chapter 2 - Mirrors and Lenses
Introduction
1. This chapter will deal with the formation of images by mirrors and lenses. Images will be described as upright or inverted,
magnified or diminished, real or virtual, and sometimes reversed. Whereas most of these terms are straightforward, the terms
real and virtual need some explanation.
2. If rays of light coming from a point (the object) are caused to converge to a second point, the second point is called the
image and is a real image. If, however, rays of light coming from a point are made to appear to diverge from a second point, the
second point is a virtual image. It will become apparent that an image formed by a mirror will be real if object and image are on
the same side of the mirror, whereas with a lens the image is real if it occurs on the opposite side of the lens to the object.
Further differences are that a real image can be projected on to a screen whereas a virtual image cannot, and real images are
inverted whilst virtual images are upright.

MIRRORS
Plane Mirrors
3. Fig 1 shows a straight line object, AB, being reflected in a plane mirror MM'. The image of AB can be determined by
considering its end points. Two incident rays are drawn from each point to the mirror and produced according to the laws of
reflection. Both pairs of reflected rays are then extended behind the mirror. Thus the reflected pair of rays coming from A
converge at A' and the pair from B converge at B'. If the points A' and B' are joined by a straight line then the complete image of
AB is produced.

8-2-2-2 Fig 1 Reflection of an Object in a Plane Mirror

Page 180 : Wed Jan 02 02:42:10 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

4. Fig 2 shows the paths of the rays of light required for an eye to see the image in a mirror. Measurement will show that the
image is as far behind the mirror as the object is in front and that the size of the image is the same as that of the object. The
image is reversed and virtual. Everday experience in the use of plane mirrors will confirm that the image is upright.

8-2-2-2 Fig 2 Viewing an Image in a Plane Mirror

Curved Mirrors

Page 181 : Wed Jan 02 02:42:10 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

5. The commonest types of curved mirrors are those consisting of a portion of the surface of a sphere; they may be either
concave or convex. The centre of the sphere, of which the curved mirror is a part, is called the centre of curvature, and its radius
is called the radius of curvature. A line drawn from the centre of curvature to the centre of the mirror surface is called the
principal axis. These terms are illustrated in Fig 3.

8-2-2-2 Fig 3 Features of a Spherical Mirror

6. Consider Fig 4. If the rays of light falling on a curved mirror are parallel to the principal axis, they are reflected from a
concave mirror so that they converge at one point, and from a convex mirror such that they appear to diverge from one point.
This point is called the principal focus (F). The distance from the principal focus to the centre of the mirror is called the focal
length and is approximately equal to half the radius of curvature.

8-2-2-2 Fig 4 Spherical Mirror - Reflection of Rays Parallel to the Principal Axis

8-2-2-2 Fig 5 Comparison of Spherical and Parabolic Curved Mirrors

Page 182 : Wed Jan 02 02:42:10 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

7. If a source of light is at or near the principal focus of a spherical concave mirror, the light rays striking the mirror near its
centre are reflected parallel to the principal axis. The rays striking the edge of the mirror are reflected so that they diverge.
Parallel rays could be produced from all points on the mirror by altering the shape to a parabola as shown in Fig 5.

Images in Curved Mirrors


8. The position and size of an image produced by a curved mirror can be determined by the technique of ray tracing. The
behaviour of certain rays from an object can be easily determined and drawn as follows:
a. Rays parallel to the principal axis will be reflected through, or appear to diverge from, the principal focus.
b. Rays passing through the centre of curvature will be reflected along the same line.
9. Fig 6 illustrates the principle applied to an object reflected in a convex mirror. F is the principal focus and C is the centre of
curvature. Rays from points A and B on the object, parallel to the principal axis, are reflected as if they diverged from F. Rays
from A and B, which would pass through C if extended behind the mirror, are reflected back along the same path. The image is
virtual, upright and smaller than the object. These characteristics are true regardless of the objects distance from the mirror.

8-2-2-2 Fig 6 Formation of Image in a Convex Mirror

10. Images in concave mirrors are real unless the object is placed between the principal focus and the mirror. The image
increases in size as the object is brought from infinity towards the mirror, attaining the same size as the object when the object
reaches the centre of curvature and being magnified when the object is inside the centre of curvature.
11. Fig 7 shows how ray tracing can be used to determine the position and size of an image produced by a concave mirror. In

Page 183 : Wed Jan 02 02:42:10 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

Fig 7a the object is outside the centre of curvature; in Fig 7b the object is inside the principal focus.

8-2-2-2 Fig 7 Formation of Images in Concave Mirrors

12. The position of the image produced by a spherical curved mirror can be determined from the equation:
1 1 1
+ =
v u f
where u = object distance, v = image distance and f = focal length.
13. In order for the equation to differentiate between real and virtual images a sign convention is necessary. The real is
positive convention is normally used which has the following rules:
a. A concave mirror has a positive focal length.
b. A convex mirror has a negative focal length.
c. Real objects are assigned a positive u value.
d. Real images have positive v values.
e. Virtual images have negative v values.
The formula will automatically generate the correct sign for any derived distance.
14. Magnification. The linear magnification due to a mirror is the ratio of the height of the image to the height of the object. In

Page 184 : Wed Jan 02 02:42:11 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

Fig 8, where A'B' is the image of AB and the triangles ABP and A'B'P are similar:

m=

A0 B0 PB0 v
=
=
AB
PB
u

Thus when the image is further from the mirror than the object is the magnification will be greater than one, and vice versa.
When a real object produces a real image the magnification is positive whilst if a virtual image is produced the magnification is
negative. By rearranging the formula of paragraph 12 it can be shown that magnification can be expressed in terms of v and f, or
u and f as follows:

8-2-2-2 Fig 8 Magnification in a Concave Mirror

m=

v-f
f

m=

f
u-f

and

LENSES
Description
15. A lens is a portion of a transparent medium bounded by two curved surfaces. Most lenses are made of glass or plastic and
their surfaces are portions of spheres or cylinders. Only spherical lenses will be described here, of which there are two basic
types as follows:
a. Convex Lenses. These are thicker at the centre than at the edges and are known as converging lenses (Fig 9a).
b. Concave Lenses. These are thinner at the centre than at the edges and are known as diverging lenses (Fig 9b).

8-2-2-2 Fig 9 Convex Lenses

Page 185 : Wed Jan 02 02:42:11 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

16. Lenses have two surfaces each of which may be considered to be part of a spherical surface and therefore has a centre of
curvature. A straight line joining the two centres of curvature is called the principal axis and is perpendicular to the surfaces
where it passes through them as shown in Fig 10.

8-2-2-2 Fig 10 Principle Axis of Convex Lens

Page 186 : Wed Jan 02 02:42:11 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

17. The principal focus is the point on the principal axis to which all rays which are close to and parallel to the axis converge,
or from which they appear to diverge, after refraction. The optical centre is a fixed point for any particular lens and coincides
with the geometric centre of a symmetrical lens (see para 20). The distance from the principal focus to the optical centre of a
lens is called the focal length. These terms are illustrated in Fig 11.

8-2-2-2 Fig 11 Lens Terminology

Page 187 : Wed Jan 02 02:42:11 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

18. Refraction by Convex Lenses. A convex lens is approximately the same as two prisms placed base to base. Fig 12 shows
parallel rays of light falling on a pair of prisms. At O the ray AO is refracted towards the normal NF. As it leaves the prism at B
it is refracted away from the normal BE along the line BC. The feature to be noted is that light is bent towards the base or
thicker part of the prism. Similarly when light rays parallel to the principal axis fall on a convex lens they are refracted towards
the thick part of the lens as shown in Fig 13.

8-2-2-2 Fig 12 Refraction by Two Prisms

8-2-2-2 Fig 13 Refraction by Convex Lens

Page 188 : Wed Jan 02 02:42:11 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

19. Refraction by Concave Lenses. A concave lens is approximately the same as the prism arrangement shown in Fig 14. The
parallel rays of light are refracted towards the base of each prism and therefore diverge. Similarly the lens in Fig 15 causes light
rays parallel to the principal axis to diverge, apparently from a point F which is called the virtual focus.

8-2-2-2 Fig 14 Refraction by Two Prisms

8-2-2-2 Fig 15 Refraction by a Concave Lens

Page 189 : Wed Jan 02 02:42:11 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

Images Formed by Lenses


20. The ray tracing technique can equally be applied to the formation of images by lenses. The rays of use are similar to those
employed in the case of mirrors. A ray parallel to the principal axis will emerge to pass through the principal focus in the case of
a converging lens or will appear to have passed through the principal focus in the case of a diverging lens. As light paths are
reversible, a ray which passes through the principal focus before entering a convex lens, or which would have passed through the
principal focus had it not been intercepted by a concave lens, will emerge parallel to the principal axis. Finally a ray coincident
with the principal axis will be undeviated; in practice provided that the thickness and diameter of the lens are small compared
with its focal length, and provided that the location of the object or image point is not too far from the axis, then this non
deviation rule can be generalized to include all rays passing through the optical centre.
21. Fig 16 shows the formation of an image by a concave lens. The image is always virtual, upright and smaller than the
object. Fig 17 illustrates the formation of a real image by a convex lens and Fig 18 shows the formation of a virtual image by a
convex lens.
22. The formula:
1 1 1
+ =
u v f
is equally applicable to lenses as it is to mirrors, using the same sign convention; convex lenses having positive focal lengths,
concave lenses having negative focal lengths. The magnification formula is also the same as for mirrors.

8-2-2-2 Fig 16 Formation of Virtual Image of Concave Lens

8-2-2-2 Fig 17 Formation of Real Image by Convex Lens

8-2-2-2 Fig 18 Formation of Virtual Image by Convex Lens

Page 190 : Wed Jan 02 02:42:12 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

Lens Power
23. A thick lens with sharply curved surfaces bends light rays more than a thin flat lens does; it has a shorter focal length. The
ability of a lens to refract light rays is a measure of its power. The power is measured in dioptres (symbol D) and if the focal
length (f) is measured in metres then:

D=

1
f

The power of a convex lens is positive and that of a concave lens is negative.
24. If two lenses are placed in contact then the resultant power can be obtained by summing the powers of the individual
components lenses.

Lens Defects
25. Spherical Aberration. Particularly if a lens has a wide aperture, rays parallel to the principal axis and passing through the
periphery of the lens converge to a point which is nearer to the lens than the point to which a narrow central beam of parallel
rays converge. At P in Fig 19 the central parts of the object are blurred whilst the peripheral portions are distinct. At F the
situation will be the reverse. The distance PF is called the longitudinal spherical aberration. Spherical aberration is more
pronounced in thick lenses with short focal lengths than in thin lenses with long focal lengths.

8-2-2-2 Fig 19 Spherical Aberration

Page 191 : Wed Jan 02 02:42:12 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

26. Chromatic Aberration. Since the refractive index of a prism or lens is greater for violet light than for red light, the lens may
be considered as having a different focal length for each colour as shown in Fig 20.

8-2-2-2 Fig 20 Chromatic Aberration

27. Correcting Aberrations. Spherical aberration can be prevented by placing an adjustable diaphragm in front of the lens thus
eliminating peripheral light rays. Alternatively a compound lens can be used to correct for both spherical and chromatic
aberration.

OPTICAL INSTRUMENTS - THE SEXTANT


Introduction
28. A sextant is an instrument for measuring the altitude of a celestial body, ie the arc of the vertical circle through the body
intercepted between the body and the celestial horizon.
29. The sextant uses a pendulous reference system to define an artificial horizon from which the altitude can be measured.

Page 192 : Wed Jan 02 02:42:12 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

When the body is viewed through the sextant an image of the pendulum graticule can also be seen and the two can be made to
coincide by adjustment to the sextant optical system. The altitude so measured is then indicated on the scales of the sextant.
30. The periscope tube of the sextant fits into a mounting which is attached to the inside of the aircraft skin. The sextant can
then be raised or lowered at will through a small hole in the top of the fuselage. The sextant periscope moves in a sleeve in the
mounting and forms an air-tight seal, which enables cabin pressure to be maintained.
31. When the sextant is required for use, it is put into the raised position with the periscope tube projecting three inches from
the skin of the aircraft. When not in use the instrument is retracted into the lowered position. De-misting and de-icing
equipment enables the sextant to be operated at all temperatures likely to be met operationally.
32. To overcome the inherent difficulties of obtaining accurate single sights, a clockwork averaging device is fitted which
averages 60 readings over a period of one or two minutes.
33. A graduated azimuth ring, fitted around the lower part of the periscope tube at its junction with the sextant body, carries a
1
scale calibrated every degree, with values shown every whole 10. The scale can be read to an accuracy of 4 .

Optical Arrangement
34. Fig 21 is a simplified diagram of the sextant optical system. The image of the celestial body is reflected down the periscope
tube by the index prism, which can be rotated in the vertical plane and acts as the measuring device. At the bottom of the tube
the image is combined with that from the pendulum system in a beam splitting cube, and both images are transmitted to the
eyepiece. A tram-line graticule is included in the eye lens system to assist in location in the field of view.

8-2-2-2 Fig 21 Pendulous Reference System

Page 193 : Wed Jan 02 02:42:13 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

35. Altitude Control. To obtain coincidence between the pendulum graticule and the celestial body, the index prism is rotated
in the vertical plane. This rotation is performed in two stages:
a. In 10 steps over a range of 10 to +80, by means of a pin and hole plate assembly, the rotation of which is transmitted
to the index prism by a pair of push rods. The 10 steps are selected by the altitude coarse setting knob and indicated in the
coarse setting scale in the adjacent window.
b. The altitude fine control wheel works a micrometer screw and lever, causing rotation of the pin and hole plate assembly
over a range of 13 30'. The fine altitude setting is indicated in degrees and minutes, in two windows below that of the
coarse setting scale. The value of the fine setting is added to that of the coarse setting to obtain the altitude of the
body.

Pendulous Reference Unit


36. The detail of the pendulous reference unit is shown in Fig 22. The pendulum is suspended on a fine nylon filament and
consists of a mirror, about 0.2 in in diameter, mounted on a square plate. The movement of the plate is restricted by a guard
ring, which is so shaped that the plate cannot rotate continuously. A small steel ball is let into the surface of the glass wedge
beneath the mirror to prevent the mirror from becoming stuck to the glass by a vertical acceleration. The pendulum is
surrounded by a silicone fluid which damps its oscillations, and the expansion bellows are incorporated to allow for changes in
volume with variation in fluid temperature. Since changes of temperature alter the viscosity and refractive index of the fluid, a
thermostat controlled heater is employed to keep the fluid temperature at a minimum of 20C.

8-2-2-2 Fig 22 Pendulous Reference Unit

Page 194 : Wed Jan 02 02:42:13 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

37. Fig 23 shows the effect of an acceleration on the pendulous reference system, which causes a false zenith to be defined. A
treatment of these acceleration errors is contained in Volume 7.

8-2-2-2 Fig 23 Effect of Acceleration on Pendulous Reference

Page 195 : Wed Jan 02 02:42:13 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

Graticule System
38. The brightness of the lamp, which illuminates both the pendulum graticule and the tram-lines, is regulated by a rheostat
control. The pendulum graticule appears as an illuminated circle 45 minutes of arc in diameter, with a split crossbar. Fig 24
shows the graticule, tram-lines, and body, and the complete image obtained when the sextant is correctly adjusted.

8-2-2-2 Fig 24 Graticules

Page 196 : Wed Jan 02 02:42:13 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

8-2-2-2 Fig 7a

8-2-2-2 Fig 7b

Page 197 : Wed Jan 02 02:42:13 2002

Electro Optics

8- 2- 2- 2

DO NOT DISTRIBUTE

AP 3456

8-2-2-2 Fig 9a Convex Lenses

8-2-2-2 Fig 9b Concave Lenses

Electro Optics
Chapter 3 - Infra-Red Radiation

Page 198 : Wed Jan 02 02:42:14 2002

Electro Optics

8- 2- 2- 3

DO NOT DISTRIBUTE

AP 3456

Characteristics of Infra-red Radiation


1. Infra-red (IR) radiation is electro-magnetic radiation and occupies that part of the electro-magnetic spectrum between visible
light and microwaves. The IR part of the spectrum is further sub-divided into Near IR, Middle IR, Far IR and Extreme IR. The
position and division of the IR band, together with the appropriate wavelengths, is shown in Fig 1.
2. All bodies with a temperature greater than absolute zero (0 K, 273C) emit IR radiation and it may be propagated both in a
vacuum and in a physical medium. As a part of the electro-magnetic spectrum it shares many of the attributes of, for example,
light and radio waves; thus it can be reflected, refracted, diffracted and polarized, and it can be transmitted through many
materials which are opaque to visible light.

Absorption and Emission


3. Blackbody. The radiation incident upon a body can be absorbed, reflected or transmitted by that body. If a body absorbs all
of the incident radiation then it is termed a blackbody. A blackbody is also an ideal emitter in that the radiation from a
blackbody is greater than that from any other similar body at the same temperature.

8-2-2-3 Fig 1 Infra-red in the Electromagnetic Spectrum

4. Emissivity (). In IR, the blackbody is used as a standard and its absorbing and emitting efficiency is said to be unity; ie =
1. Objects which are less efficient radiators, ( < 1), are termed greybodies. Emissivity is a function of the type of material and
its surface finish, and it can vary with wavelength and temperature. When varies with wavelength the body is termed a
selective radiator. The for metals is low, typically 0.1, and increases with increasing temperature; the for non-metals is high,
typically 0.9, and decreases with increasing temperature.

Spectral Emittance
5. Plancks Law. A blackbody whose temperature is above absolute zero emits IR radiation over a range of wavelengths with
different amounts of energy radiated at each wavelength. A description of this energy distribution is provided by the spectral
emittance, W, which is the power emitted by unit area of the radiating surface, per unit interval of wavelength. Max Planck
determined that the distribution of energy is governed by the equation:

W =

Page 199 : Wed Jan 02 02:42:14 2002

1
2c2 h hc/ kT1
e
5

Electro Optics

8- 2- 2- 3

DO NOT DISTRIBUTE

AP 3456

where = Wavelength

h = Planck's constant

T = Absolute temperature

c = Velocity of light

k = Boltzmann's constant
6. Temperature/Emittance Relationship. This rather complex relationship is best shown graphically as in Fig 2 in which the
spectral emittance is plotted against wavelength for a variety of temperatures. It will be seen that the total emittance, which is
given by the area under the curve, increases rapidly with increasing temperature and that the wavelength of maximum emittance
shifts towards the shorter wavelengths as the temperature is increased.

8-2-2-3 Fig 2 Distribution of IR Energy with Temperature

Page 200 : Wed Jan 02 02:42:14 2002

Electro Optics

8- 2- 2- 3

DO NOT DISTRIBUTE

AP 3456

7. Stefan-Boltzmann Law. The total emittance of a blackbody is obtained by integrating the Max Planck equation which gives
the result:
W = T4

where W = Total emittance

= Stefan Boltzmann constant

T = Absolute temperature
For a greybody the total radiant emittance is modified by the emissivity, thus:
W = T4
8. Wiens Displacement Law. The wavelength corresponding to the peak of radiation is governed by Wiens displacement law
which states that the wavelength of peak radiation, m, multiplied by the absolute temperature is a constant. Thus:
m T = 2900K
By substituting m = 2900/T into Planck's expression it is found that:
Wm = 1.3 1015 T5 expressed in Watts cm 2 1
ie the maximum spectral radiant emittance depends upon the fifth power of the temperature.

Geometric Spreading
9. The laws so far discussed relate to the radiation intensity at the surface of the radiating object. In general, radiation is
detected at some distance from the object and the radiation intensity decreases with distance from the source as it spreads into an
ever increasing volume of space. Two types of source are of interest; the point source and the plane extended source.
10. Point Source. A point source radiates uniformly into a spherical volume. In this case the intensity of radiation varies as the
inverse square of the distance between source and detector.
11. Plane Extended Source. When the radiating surface is a plane of finite dimensions radiating uniformly from all parts of the
surface then the radiant intensity received by a detector varies with the angle between the line of sight and the normal to the
surface. For a source of area A the total radiant emittance is WA. The radiant emittance received at a distance d and at an angle
from the normal is given by:
WA
cos
2d2

IR Sources
12. It is convenient to classify IR sources by the part they play in IR systems; ie as targets, as background, or as controlled
sources. A target is an object which is to be detected, located or identified by means of IR techniques, while a background is any

Page 201 : Wed Jan 02 02:42:14 2002

Electro Optics

8- 2- 2- 3

DO NOT DISTRIBUTE

AP 3456

distribution or pattern of radiation, external to the observing equipment, which is capable of interfering with the desired
observations. Clearly what might be considered a target in one situation could be regarded as background in another. As an
example terrain features would be regarded as targets in a reconnaissance application but would be background in a low-level air
intercept situation. Controlled sources are those which supply the power required for active IR systems (eg communications), or
provide the standard for calibrating IR devices.

Targets
13. Aircraft Target. A supersonic aircraft generates three principle sources of detectable and usable IR energy. The typical jet
pipe temperature of 773 K produces a peak of radiation, (from Wiens law), at 3.75 microns. The exhaust plume produces two
peaks generated by the gas constituents; one at 2.5 - 3.2 microns due to carbon dioxide, the other at 4.2 - 4.5 microns due to
water vapour. The third source is due to leading edge kinetic heating giving a typical temperature of 338 K with a
corresponding radiation peak at about 7 microns.
14. Reconnaissance. Terrestrial IR reconnaissance and imaging relies on the IR radiation from the Earth which has a typical
temperature of 300 K. The peak of radiation corresponding to this temperature is about 10 microns and so systems must be
designed to work at this wavelength.

Background Sources
15. Regardless of the nature of the target source, a certain amount of background or interfering radiation will be present,
appearing in the detection system as noise. The natural sources which produce this background radiation may be broadly
classified as terrestrial or atmospheric and celestial.
16. Terrestrial Sources. Whenever an IR system is looking below the horizon it encounters the terrestrial background
radiation. As all terrestrial constituents are above absolute zero they will radiate in the infra-red, and in addition IR radiation
from the sun will be reflected. Green vegetation is a particularly strong reflector which accounts for its bright image in IR
photographs or imaging systems. Conversely, water, which is a good reflector in the visible part of the spectrum, is a good
absorber of IR, and therefore appears dark in IR images.
17. Atmospheric and Celestial Sources. Whenever an IR device looks above the horizon the sky provides the background
radiation. The radiation characteristics of celestial sources depends on the source temperature together with modifications by the
atmosphere.
a. The Sun. The sun approximates to a blackbody radiator at a temperature of 6000 K and thus has a peak of radiation at
0.5, which corresponds to yellow-green light. The distribution of energy is shown in Fig 3 from which it will be seen that
half of the radiant power occurs in the infra-red. The Earths atmosphere changes the spectrum by absorption, scattering
and some re-radiation such that although the distribution curve has essentially the same shape, the intensity is decreased and
the shorter, ultraviolet, wavelengths are filtered out. The proportion of IR energy remains the same or perhaps may be
slightly higher. Sunlight reflected from clouds, terrain and sea shows a similar energy distribution.

8-2-2-3 Fig 3 Spectral Distribution of Solar Radiation

Page 202 : Wed Jan 02 02:42:14 2002

Electro Optics

8- 2- 2- 3

DO NOT DISTRIBUTE

AP 3456

8-2-2-3 Fig 4 Spectral Energy Distribution of Background Radiation from the Sky

b. The Moon. The bulk of the energy received from the moon is re-radiated solar radiation, modified by reflection from the
lunar surface, slight absorption by any lunar atmosphere and by the Earths atmosphere. The moon is also a natural
radiating source with a lunar daytime surface temperature up to 373 K and lunar night time temperature of about 120 K.
The near sub-surface temperature remains constant at 230 K, corresponding to peak radiation at 12.6.
c. Sky. Fig 4 shows a comparison of the spectral distribution due to a clear day and a clear night sky. At night, the short
wavelength background radiation caused by the scattering of sunlight by air molecules, dust and other particles, disappears.
At night there is a tendency for the Earths surface and the atmosphere to blend with a loss of horizon since both are at the
same temperature and have similar emissivities.
d. Clouds. Clouds produce considerable variation in sky background, both by day and by night, with the greatest effect
occurring at wavelengths shorter than 3 due to solar radiation reflected from cloud surfaces. At wavelengths longer than
3, the background radiation intensity caused by clouds is higher than that of the clear sky. Low bright clouds produce a
larger increase in background radiation intensity at this wavelength than do darker or higher clouds. As the cloud formation
changes the sky background changes and the IR observer is presented with a varying background both in time and space.
The most serious cloud effect on IR detection systems is that of the bright cloud edge. A small local area of IR radiation is
produced which may be comparable in area to that of the target, and also brighter. Early IR homing missiles showed a
greater affinity for cumulus cloud types than the target aircraft. Discrimination from this background effect requires the use

Page 203 : Wed Jan 02 02:42:14 2002

Electro Optics

8- 2- 2- 3

DO NOT DISTRIBUTE

AP 3456

of spectral and spatial filtering.

IR Transmission in the Atmosphere


18. Atmospheric Absorption. The periodic motions of the electrons in the atoms of a substance, vibrating and rotating at certain
frequencies, give rise to the radiation of electro-magnetic waves at the same frequencies. However, the constituents of the
Earths atmosphere also contain electrons which have certain natural frequencies. When these natural frequencies are matched
by those of the radiation which strikes them, resonance absorption occurs and the energy is re-radiated in all directions. The
effect of this phenomenon is to attenuate certain IR frequencies. Water vapour and carbon dioxide are the principle attenuators
of IR radiation in the atmosphere. Figs 5a, 5b and 5c show the transmission characteristics of the atmosphere at sea-level, at
30,000 ft and at 40,000 ft.
19. Scattering. The amount of scattering depends upon particle size and particles in the atmosphere are rarely bigger than 0.5
microns, and thus they have little effect on wavelengths of 3 microns or greater. However, once moisture condenses on to the
particles to form fog or clouds, the droplet size can range between 0.5 and 80 microns, with the peak of the size distribution
between 5 and 15 microns. Thus fog and cloud particles are comparable in size to IR wavelengths and transmittance becomes
poor. Raindrops are considerably larger than IR wavelengths and consequently scattering is not so pronounced. Rain, however,
tends to even out the temperature difference between a target and its surroundings.
20. Scintillation. Where a beam of IR passes through regions of temperature variation it is refracted from its original direction.
Since such regions of air are unstable, the deviation of the beam is a random, time varying quantity. The effect is most
pronounced when the line of sight passes close to the earth and gives rise to unwanted modulations of the signal, and incorrect
direction information for distant targets.

8-2-2-3 Fig 5 Atmospheric Transmittance vs Altitude

Page 204 : Wed Jan 02 02:42:15 2002

Electro Optics

8- 2- 2- 3

DO NOT DISTRIBUTE

Page 205 : Wed Jan 02 02:42:15 2002

AP 3456

Electro Optics

8- 2- 2- 3

DO NOT DISTRIBUTE

AP 3456

8-2-2-3 Fig 5a Transmittance at Sea Level

8-2-2-3 Fig 5b Transmittance at 30,000 ft

8-2-2-3 Fig 5c Transmittance at 40,000 ft

Page 206 : Wed Jan 02 02:42:15 2002

Electro Optics

8- 2- 2- 3

DO NOT DISTRIBUTE

AP 3456

Electro Optics
Chapter 4 - Lasers
Introduction
1. The laser is a device that emits an extremely intense beam of energy in the form of electro-magnetic radiation in the near
ultra-violet, visible or infra-red part of the electromagnetic spectrum. The word LASER is an acronym derived from the
definition of its function, Light Amplification by Stimulated Emission of Radiation. The word has been so integrated into the
English language that it is no longer written in capital letters, as are most acronyms. Indeed, in technical circles, its use has
spawned a verb, to lase, which describes the action of using a laser. Unlike the radiation from other sources, laser light is
monochromatic (single wavelength), coherent (all waves in phase), and highly collimated (near parallel beam). Since the first
laser was constructed in 1960 in California the development has been rapid and uses have been found in a wide variety of civil
and military spheres, including surgery, communications, holography and target marking and range-finding. In order to
understand the principle of operation of a laser it is first necessary to appreciate some aspects of atomic structure and energy
levels.

Atomic Energy Levels


2. The atom consists of a central nucleus containing positively charged protons and neutral neutrons. Surrounding the nucleus
are negatively charged electrons. The number of protons and electrons are equal thus resulting in a net zero charge on the atom
as a whole. The electrons have a certain energy level due to the sum of their kinetic energy and electrostatic potential energy.
However electrons within an atom are constrained to exist in one of a series of discrete energy levels. In normal circumstances
the electrons will adopt the minimum energy levels permitted and the atom is then said to be in its ground state. In order for
electrons to enter a higher energy level, energy in one form or another has to be supplied. If such a transition to a higher energy
level occurs the atom is said to be in an excited state.
3. Conventionally the energy states of an atom can be shown on an energy level diagram as shown in Fig 1. The horizontal
lines represent the permitted energy levels, increasing upwards, separated by varying energy differences, E. The horizontal
extent of the lines has no significance. The base line is the ground state ie the lowest energy level in which atoms will normally
be found (A in Fig 1). By supplying energy it may be possible to excite an atom (B in Fig 1) into a higher energy level. This
process is known as absorption.

8-2-2-4 Fig 1 Atomic Energy Levels

Page 207 : Wed Jan 02 02:42:16 2002

Electro Optics

8- 2- 2- 4

DO NOT DISTRIBUTE

AP 3456

8-2-2-4 Fig 2 Stimulated Emission

Emissions
4. Spontaneous Emission. An atom in an excited state is unstable and will have a tendency to revert to the ground state. In
doing so it will emit the excess energy as a single quantum of energy known as a photon, a process known as spontaneous
emission. If a large population of atoms are excited into higher states, as for example in a fluorescent lighting tube, then they
will occupy a wide band of energy levels. On undergoing spontaneous emissions, some will revert to the ground state directly
whilst others will drop via intermediate levels. In either case photons will be emitted with a wide range of energy levels
corresponding to the various energy level differences. The frequency of the emitted energy is determined by the Planck-Einstein
equation:
E = hf

where E is the photon energy, f is the frequency and h is Planck's constant.

Thus in a fluorescent tube, as there are a wide variety of energy level transitions, there will be a wide variety of frequencies in
the emitted light giving the impression of white light. It should be noted that what transitions occur and when they occur is a
random process. Equally the direction in which the emitted photon is radiated is also random. Thus the radiation generated by
spontaneous emission is isotropic (ie radiating in all directions), non-coherent and covers a wide frequency band.
5. Stimulated Emission. As early as 1917 Einstein predicted on theoretical grounds that the downward transition of an atom
could be stimulated to occur by an incident photon of exactly the same energy as the difference between the energy levels. It is
this type of emission that is exploited in lasers. This process is shown in Fig 2. It should be noted that the incident photon is not
absorbed and so for each incident photon, two photons are emitted each of which can stimulate further emissions providing that
there are atoms in the higher energy level. Furthermore these emitted photons have the same energy and therefore frequency, the
same phase and are emitted in the same direction as the incident photons. These are of course the characteristics of laser
radiation.

Page 208 : Wed Jan 02 02:42:16 2002

Electro Optics

8- 2- 2- 4

DO NOT DISTRIBUTE

AP 3456

6. Population Inversion. In the normal course of events, however, most atoms are in the ground state and so incident photons
are more likely to excite a ground state atom than to induce stimulated emission. It is therefore necessary to ensure that there are
more atoms in the appropriate higher energy level than in the ground state, a situation known as a population inversion. The
process by which this is achieved will be described with reference to the ruby laser which was the first lasing medium to be used.
7. Optical Pumping. Fig 3a illustrates the normal configuration with respect to the chromium atoms within a ruby crystal. The
diagram shows a number of atoms in the ground state and a number of as yet unoccupied higher energy levels. It should be
noted that the energy levels in Fig 3 refer to the energy of the atom as a whole and not to the energy levels of the constituent
electrons. At the start of the process the ruby is subjected to a burst of intense white light generated by a system similar to a
photographic electronic flash gun. As the white light comprises a wide range of frequencies then a whole range of energies will
be imparted to the ground state atoms. Some of these atoms will therefore be excited to a range of higher energy levels (Fig
3b); a process known as optical pumping.
8. The Metastable State. From these higher energy levels spontaneous emissions will occur but whereas some will be due to
transitions to the ground state, in the case of chromium the majority will decay to an intermediate level known as a metastable
state as shown in Fig 3c from which atoms may emit photons at random. Nevertheless, this state, apart from being a preferential
level, has the additional feature that atoms tend to remain there for a longer time (by a factor of some 1000s) than they do in any
other level other than the ground state. In this way a population inversion is achieved ie there are more atoms in the metastable
state than in the ground state.
9. Lasing Action. Inevitably at some time an atom in the metastable state will make a spontaneous transition to the ground
state with the emission of a photon. This photon can now do one of two things; it can either excite a ground state atom into a
higher level or it can stimulate an excited atom in the metastable state to make a transition to the ground state. Since a
population inversion has been achieved, then on balance it is more likely to stimulate emission than to be absorbed by a ground
state atom (Fig 3d). Thus lasing will be initiated. At the end of this process all of the atoms will be back in the ground state
ready for further optical pumping to start the cycle again.

8-2-2-4 Fig 3 The Stages of Operation in a Ruby Laser

Page 209 : Wed Jan 02 02:42:16 2002

Electro Optics

8- 2- 2- 4

DO NOT DISTRIBUTE

AP 3456

10. Other Techniques. Optical pumping is not the only means of achieving a population inversion. The helium-neon laser, for
example, uses a different method. The medium in this case is a mixture of helium and neon gases of which the neon is
responsible for lasing. Energy is input to the helium by means of an electrical discharge and the energized helium atoms
transmit their excess energy not by radiation but in collisions with neon atoms. The neon atoms are excited to a high energy
level such that there is a population inversion between this level and an intermediate level rather than with respect to the ground
state. Stimulated emission therefore occurs between these two higher levels. This process is illustrated in Fig 4. Atoms in the
bottom lasing level eventually decay spontaneously back to the ground state.
11. The system so far described produces monochromatic and coherent radiation, however it is not very intense and is not
emitted as a beam. This is because the stimulating photons are incident upon the atoms from random directions and so the
emitted photons follow, likewise, random directions. In addition there will of course be a proportion of random spontaneous
emissions. It is therefore necessary to ensure that as many photons as possible are travelling in the required direction. This is
achieved by having the lasing medium within an optical resonant cavity.

The Laser
12. The features of the working laser are shown in Fig 5. The optical resonant cavity is achieved by placing mirrors at each
end of the lasing medium. These mirrors are separated by an integral number of -wavelengths of the laser radiation and are
accurately aligned perpendicular to the laser axis. One of the mirrors is semi-transparent. Photons travelling normal to the
mirrors will be reflected backwards and forwards through the cavity and in the process will stimulate further emissions which
will radiate in the same direction. The n -wavelength nature of the mirror separation ensures that the radiation stays in
phase. Off axis radiation will soon be lost to the system through the side walls allowing the axial radiation to increase rapidly in
relation to the non-axial radiation. The semi-silvered mirror allows the highly directional beam to leave the cavity.

Page 210 : Wed Jan 02 02:42:16 2002

Electro Optics

8- 2- 2- 4

DO NOT DISTRIBUTE

AP 3456

8-2-2-4 Fig 4 The Process in a Helium-Neon Laser

8-2-2-4 Fig 5 Laser Schematic

13. Q-Switching. A typical ruby laser as described will have a nominal output power of several kW and a pulse length in the
order of a millisecond. For many applications it would be beneficial to increase the power by reducing the pulse length. The
technique used to achieve this is known as Q-switching. Between the lasing medium and the fully silvered mirror is a glass cell
containing a green dye. Although the lasing action starts once the pumping commences, the green dye absorbs the red laser light
preventing the build up of energy in the resonant cavity. In doing so the molecules in the dye are raised to an excited state. The
concentration of the dye is arranged so that the dye molecules are all excited coincidently with the maximum number of atoms of
chromium being in the metastable level. At this point the dye becomes transparent to the laser wavelength and there is then a
very rapid build up of lasing action. The pulse of laser radiation is delivered in about 10 nanoseconds, before the dye molecules
return to the ground state and shut off the laser. The output power can be increased to the 100s of mW order by this technique.

8-2-2-4 Fig 3a Optical Pumping

Page 211 : Wed Jan 02 02:42:16 2002

Electro Optics

8- 2- 2- 4

DO NOT DISTRIBUTE

AP 3456

8-2-2-4 Fig 3b Atomic Excitation

8-2-2-4 Fig 3c Population Inversion

Page 212 : Wed Jan 02 02:42:16 2002

Electro Optics

8- 2- 2- 4

DO NOT DISTRIBUTE

AP 3456

8-2-2-4 Fig 3d Stimulated Emission

Acoustics
Chapter 1 - The Nature of Sound
Introduction
1. A Definition of Sound. Sound is the name given to the sensation perceived by the human ear. Every sound is produced by
the vibration of the object from which it originates; thus any explanation of the nature of sound must include a discussion of

Page 213 : Wed Jan 02 02:42:17 2002

Acoustics

8- 2- 3- 1

DO NOT DISTRIBUTE

AP 3456

vibratory motion. The audible sound spectrum is generally acknowledged to cover the frequency range from 15 to 20,000 hertz.
2. Simple Harmonic Motion. The simplest form of vibratory motion can be represented by the oscillation of a pendulum bob
swinging through a small angle. If the displacement of the bob from the central position is plotted on a graph against time, the
variation of the displacement with time gives rise to a sine curve as shown in Fig 1. This motion is called simple harmonic
motion, and could equally describe the vibration of a tuning fork. The displacement of the bob can be described by the equation:

d = a sin

2t
(1)
T

where a and T are constants as shown in Fig 1.


3. Frequency, Amplitude and Period. In Fig 1 the cycle represents one complete swing of the pendulum. The frequency of the
oscillation is defined as the number of cycles per second (1 cycle per second = 1 hertz (Hz)). During one cycle the bob twice
attains maximum deflection from the central position. The maximum distance displaced is called the amplitude, and in Fig 1 this
is shown by the constant a. The period of a vibration is the time it takes to complete one cycle. The frequency can be expressed
as:

f=

1
(2)
T

where f is the frequency and T is the period in seconds.

8-2-3-1 Fig 1 Simple Harmonic Motion

4. Fourier's Theorem. The vibration of a tuning fork is the nearest audible equivalent to the oscillation of a pendulum. It can
be shown that any vibratory motion which repeats itself regularly can be represented as the resultant or combination of simple
harmonic frequencies of suitably chosen amplitudes. These frequencies must also be integral multiples of the frequency with
which the motion repeats itself (Fouriers Theorem). Hence equation (1) is the basis for describing the vibrations of all sounding
objects.
5. Medium of Transmission. If a sound source is placed in an air-tight chamber which is slowly evacuated of air, the sound
will gradually die away as the vacuum increases. Eventually the sound will cease, although it can be seen that the source is still
vibrating. Such an experiment can be used to demonstrate that a material medium such as air, water, wood, glass or metal is
required for the sound to be transmitted.

The Propagation of Sound


6. Transverse Waves and Longitudinal Waves. The wave motion observed when the surface of a pond is disturbed is called
transverse wave motion, because the particles of water oscillate at right angles to the direction of propagation of the waves.

Page 214 : Wed Jan 02 02:42:17 2002

Acoustics

8- 2- 3- 1

DO NOT DISTRIBUTE

AP 3456

Sound waves, however, are propagated as longitudinal waves. In this kind of wave motion the particles oscillate, each about a
fixed point, in the direction of propagation of the waves. In Fig 2 the undisturbed particles of a medium are represented by
equally spaced dots. A similar set of particles is shown in Fig 3 being disturbed by the passage of a sound wave through the
medium. Each particle is displaced to the right and left of its undisturbed position as the wave passes through the medium. If
the displacement of a single particle is plotted on the vertical axis of a time graph the familiar sine wave form of simple
harmonic motion is produced as shown in Fig 4.

8-2-3-1 Fig 2 Undistributed Particles in a Medium

8-2-3-1 Fig 3 Passage of a Sound Wave Through a Medium

8-2-3-1 Fig 4 Longitudinal Wave Plotted in Graphical Form

7. Pressure Variations. When the particles of a medium are displaced by the passage of a sound wave there is a consequent
local variation in pressure. It is these small changes in pressure which actuate the human ear and mechanical devices such as
microphones. Fig 5 shows the pressure variations that accompany the passage of a sound wave; they consist of alternate
compressions and rarefactions.

The Properties of Sound Waves


8. Reflection. Like light waves, sound waves are reflected from a plane surface such that the angle of reflection is equal to the
angle of incidence. It can also be shown that sound waves come to a focus when they are incident on a concave reflector.
9. Reverberation. If sound is generated within a large enclosed space it can be heard directly from the source and indirectly
from reflected and diffused (multiple reflection) sound waves. The indirect sounds are called reverberations and continue for a
finite time after the sound source has been silenced.

Page 215 : Wed Jan 02 02:42:17 2002

Acoustics

8- 2- 3- 1

DO NOT DISTRIBUTE

AP 3456

10. Refraction. Sound waves travel faster in warm air than in cold air and are therefore refracted when there is a temperature
gradient. Refraction also occurs in water and other media because of changes in the velocity of sound.
11. Interference. If two sound sources are of the same frequency and intensity, and are initially in phase (ie coherent) they will
interfere with each other and will cancel or reinforce according to the path difference. If the path difference is an odd number of
half wavelengths cancellation occurs and no sound is heard. If it is an even number of whole wavelengths the sounds will
reinforce and a louder sound is heard.

8-2-3-1 Fig 5 Pressure Variation in a Medium

12. The Wave-Front. If a single pulse of noise, such as an explosion, occurs in a medium, the paths followed by the sound
waves can be traced by placing a number of recording microphones in the vicinity and noting the time taken for the sound to
reach each microphone. If the microphones are located at specified ranges from the source, the points in space reached
simultaneously by the sound can be plotted. These points are considered to lie on a surface called the wave-front, and in a
homogeneous medium the direction of propagation is perpendicular to this surface. It can also be observed that the sound is
propagated outwards at a constant velocity.
13. Diffraction. It is a common experience that it is possible to hear sound even when the source is behind an obstruction. This
bending of sound waves (or indeed any other type of wave) around such an obstacle is known as diffraction. Although the
mathematical treatment is rather complex, a satisfactory explanation of the phenomenon can be made using Huygens principle.
Huygens principle states that all points on a wave front can be considered as point sources from which secondary wavelets are
generated. After a time interval a new position of the wave-front will be established as the surface of tangency to these
secondary wavelets. The way that this principle accounts for the bending of sound around an obstacle is shown in Fig 6.

8-2-3-1 Fig 6 Huygens' Principle - 'Bending' Sound Around an Obstacle

Page 216 : Wed Jan 02 02:42:17 2002

Acoustics

8- 2- 3- 1

DO NOT DISTRIBUTE

AP 3456

The Velocity of Sound


14. Since the pressure variations produced by a sound wave are so rapid that no transfer of heat can occur, the process is
considered to be adiabatic. The velocity of sound (c) in a gas is found to be given by:
r

c=

(3)

where is the ratio of the specific heat at constant temperature to that at constant volume, p is the pressure, and is the density.
p

15. Since = RT in an ideal gas, where R is the specific gas constant and T is the temperature in K, equation (3) can be written
as:
c=

RT

(4)

p
Therefore in a given gas, since and R are constants, c R T within the range in which the gas obeys the ideal gas equation.
A working expression for the speed of sound in air at a temperature of tC is given by the equation:
ct = (330 + 0.61lt) ms

In equation (3) both and p are constant for a given gas at a specific temperature, and from this it can be deduced that the
velocity of sound in air is independent of pressure.
16. The velocity of sound in water is covered in Chapter 4.

The Intensity of Sound

Page 217 : Wed Jan 02 02:42:18 2002

Acoustics

8- 2- 3- 1

DO NOT DISTRIBUTE

AP 3456

17. The intensity of sound at any place is defined as the rate of flow of energy across unit area perpendicular to the direction of
propagation. If a sound source is emitting J joules of energy per second uniformly in all directions it can be calculated that the
energy passing through unit area is proportional to the inverse square of the distance from the source. The intensity of sound is
further attenuated by the absorption of energy by the medium through which it is propagated.

The Doppler Effect


18. The Doppler effect occurs when there is a relative velocity between the sound source and the observer. Fig 7 shows how
the change of wavelength and hence frequency occurs when a source of sound is moving either towards or away from the
observer. The circles represented by A, B, C etc correspond to successive wave-fronts generated at a, b, c etc by the moving
source. It is clear that to the observer at X, passage of the wave-fronts will be more frequent (ie the observer will hear a higher
pitched sound than was generated), while to the observer at Y, passage of the wave-fronts will be less frequent (ie the observer
will hear a lower pitched sound).
19. The velocity of sound in air of uniform temperature is constant, irrespective of any movement of the source or the observer.
However, any movement of the source will alter the wavelength of a sound in air and hence change the pitch of the sound heard
by the observer. If the component of velocity of the source towards the observer is Vs, then the frequency, f, of the note heard
by the observer is given by:
f0 = fo.

c
c Vs

where fo = frequency of the note if heard from a stationary source, and c = velocity of sound in air. Any movement of the
observer alters the velocity of the sound relative to the observer, and this also results in a change of pitch. If Vo is the
component of velocity of the observer towards the source, then the frequency, f, of the note heard by the observer is given by:
f0 = fo.

c + Vo
c

If both the source and the observer are moving then the frequency, f, of the note heard by the observer is given by:

f= fo.

c + Vo
c Vs

8-2-3-1 Fig 7 The Doppler Effect

Page 218 : Wed Jan 02 02:42:18 2002

Acoustics

8- 2- 3- 1

DO NOT DISTRIBUTE

AP 3456

Acoustics
Chapter 2 - Acoustic Measurement
Introduction
1. Sound consists of small variations in pressure above and below the ambient pressure with respect to time. It may be
depicted diagrammatically as a sine curve reflecting the gradual increase in pressure to a maximum and its subsequent decrease
to a minimum. The maximum excursion of pressure difference from the ambient level is called the amplitude, but for practical
purposes some kind of average value over time is required that gives equal weighting to both the rarefaction and compression
phases. This average value is known as the root-mean-square (rms) value, and its derivation is described below.
2. Root-Mean-Square Value. Fig 1a shows a graph of the sinusoidal sound pattern with amplitude on the y axis. The first
stage, (Fig 1b), is to square all the values of amplitude which has the effect of making the negative values positive. The squared
values are then averaged, (Fig 1c), to produce a level value. Finally the square root of the average is taken (Fig 1d). As with all
2
other pressures, the units are newtons per square metre (Nm ) or Pascals (Pa).

8-2-3-2 Fig 1 Root-Mean-Square Derivation Sound Intensity

Sound Intensity
3. Sound intensity is a measure of the power transmitted per unit area, the area being at right angles to the direction in which
the sound is propagating. For unhindered sound, away from the source, the intensity is proportional to the square of the pressure
ie
2
I = kp where k is a constant determined by the medium through which the sound is travelling (eg for air at atmospheric
pressure and 20C, k = 1/410).
The term sound intensity is used in some reference books, but it is not used as a measure in underwater acoustics.

Sound Pressure Level


4.

In underwater acoustics, measurements are always in terms of sound pressure level (SPL). Sound pressure levels cover a

Page 219 : Wed Jan 02 02:42:18 2002

Acoustics

8- 2- 3- 2

DO NOT DISTRIBUTE

AP 3456

-5
very wide range of values, for example the range from the threshold of hearing to the onset of pain is from 2 x 10 Pa to 20 Pa.
Using a linear scale over this range would be cumbersome and so a logarithmic (decibel) scale is used relative to a datum
-5
pressure level. The datum pressure that is chosen for sound in air is 2 x 10 Pa, which is equivalent to the lowest sound
pressure at 1000 Hz, detectable on average by people with normal hearing. In underwater measurement the datum pressure level
-6
is the micro-pascal, (1 x 10 Pa).
5. As the energy contained in a wave is proportional to the amplitude squared, and if p is the sound pressure to be measured,
then if the transmission was to take place in water:

SPL = 10 log

= 20 log

p2
(1 10

6 2

dB

p
dB
1 106

6. Note that a significant change in sound pressure level is represented by only a small change in decibel notation. If the
pressure from a noise source is doubled, this equates to 10 log 2 in dBs, ie 3 dBs. Thus, if the noise level from, say, a jet aircraft
with four engines running was 140 dBs, the same aircraft running on two engines would generate a level of 137 dBs. A one or
two dB difference in the measurement of acoustic pressure may therefore be significant in target detection.

8-2-3-2 Fig 1a

8-2-3-2 Fig 1b

Page 220 : Wed Jan 02 02:42:18 2002

Acoustics

8- 2- 3- 2

DO NOT DISTRIBUTE

AP 3456

8-2-3-2 Fig 1c

8-2-3-2 Fig 1d

Acoustics
Chapter 3 - Oceanography
Introduction
1. Maritime Patrol Aircraft are tasked with the location of submarines using acoustic sensor equipment. The aim of this
chapter is to outline those aspects of the oceanic environment which have a bearing on the transmission of sound. Details of
how acoustic systems are affected by this environment will be covered in Chapter 4.

Page 221 : Wed Jan 02 02:42:19 2002

Acoustics

8- 2- 3- 3

DO NOT DISTRIBUTE

AP 3456

GEOLOGY AND STRUCTURE


General
2. The oceans are not simply depressions in the Earths surface filled with water, rather they have a fairly complex structure.
Fig 1 shows a cross section of the Earths crust between Africa and South America and just into the eastern Pacific Ocean.
3. In the centre of the Atlantic Ocean is what amounts to a mountain chain - the Mid-Atlantic Ridge. This is part of a ridge
system which extends from North of Iceland southwards into the South Atlantic, eastwards into the Indian Ocean, then
northwards again in the East Pacific before disappearing under the North American continent. Its total length is over 50,000
kms and its width between 1000 and 4000 kms. The ridge generally rises to about 2 kms, occasionally 5 kms, above the flanking
plain, and the crest is normally about 2 kms below sea-level. The Mid-Atlantic Ridge has an axial rift and it is here that new
ocean crust is created. This crust formation leads to the Atlantic widening at a rate of 2-4 cm/year.
4. Adjacent to and either side of the ridge lie the abyssal plains which are in general relatively flat areas. This flatness is
primarily due to a thick layer of sediment overlaying the rough topography which has been generated at the ridge. In places this
flatness is punctuated by abyssal hills and seamounts. The majority of these are ocean floor volcanoes and most do not rise
above sea-level; where they do they form oceanic islands. The Pacific Ocean is particularly well endowed with seamounts some
of which are very large. Mauna Loa for example is about 100 kms across at its base and rises to about 9 kms above the plain, ie
about as high as Mt. Everest. Many of these seamounts have flat tops and are known as guyots. It is believed that this flat
topped effect was caused by wave erosion at a time when the volcano was at or above sea-level; with the passage of time they
have subsided below the surface.
5. The margins of the Atlantic and Pacific Oceans are rather different. The Pacific margin is characterized by seismic and
volcanic activity whereas the Atlantic margin is aseismic. This is because the Pacific margin coincides with the edge of a crustal
plate, whereas in the case of the Atlantic the ocean and the adjacent continents are on the same plate. This difference gives rise
to characteristically different ocean floor topography.

Atlantic Margin
6. Continental Shelf. Fig 2 shows the typical shape of an Atlantic style ocean margin. The continental shelf is geologically
part of the continent rather than the ocean; the underlying rock type is not oceanic. The shelf is normally coated with sediments
derived from the adjacent land.

8-2-3-3 Fig 1 Cross Section of the Earths Crust Between Africa and South America Showing Ocean Features

8-2-3-3 Fig 2 Typical Topography of an Atlantic Style Ocean Margin

Page 222 : Wed Jan 02 02:42:19 2002

Acoustics

8- 2- 3- 3

DO NOT DISTRIBUTE

AP 3456

The width of the shelf can extend to as much as 1500 kms although for the most part it is less than this, typically a few hundred
kms. The surface is generally flat with an average gradient of only 0.1. At the shelf break the water depth varies between 20
and 500 m with an average of around 130 m.
7. Continental Slope. The continental slope marks a sharp increase in gradient to about 4 on average. The width of the slope
is between 20 and 100 kms and the base of the slope lies at a depth of between 1.5 and 3.5 kms. Frequently the slope is cut by
submarine canyons along which sediment is transported to the deep ocean. These canyons, which are somewhat similar to
V-shaped river valleys on land, often start on the continental shelf and commonly coincide with the mouths of major rivers. The
end of the continental slope marks the end of the continental crust and the beginning of the oceanic crust.
8. Continental Rise. The continental rise has a much gentler gradient than the slope, about 1, and is built up of the sediments
which have flowed down the slope. Typically the continental rise extends for about 500 kms and reaches a depth of some 4 kms
before merging into the abyssal plain.

Pacific Margin
9. The most significant difference between an Atlantic and a Pacific type margin is the presence of a very deep trench in the
latter at the outer edge of the continental slope. In general the shelf is narrower in the Pacific, about 50 kms. The shelf break
tends to be more abrupt and the slope often has a steeper gradient, up to about 10 in places. The continental rise is missing and
is replaced by a trench which marks the site where the oceanic crust is being subducted beneath the continental crust. It is this
subduction which gives rise to seismic and volcanic activity. The trenches can reach depths of over ll kms, and a depth of 8 kms
would not be untypical. In places the trench may be partly filled with land derived sediments.

SEAWATER
Physical Properties
10. Salinity. Salinity is a measure of the dissolved solids in sea water. It is usually expressed in values of parts per thousand by
1
weight (ie gm kg ) and the symbol o/oo is used. Surface salinity represents a balance between an increase due to evaporation
and freezing on the one hand, and a decrease due to precipitation, ice melting, and river influx on the other hand. Despite these
effects salinity values do not vary greatly; between 30 and 40 O /oo at the extremes. Minimum values occur in coastal regions
with large river discharges and the maximum values occur in the Red Sea and Persian Gulf. In the open ocean surface salinity
values are even more conservative, varying between about 33 and 37 O /oo. The maximum values tend to occur around the
tropics, where the effect of evaporation is at its highest. In low and middle latitudes salinity decreases with depth in the first 600
- 1000 m, a zone known as a halocline, below which it becomes virtually constant at 34.5-35 O /oo.

8-2-3-3 Fig 3 Depth/Temperature Profiles for Different Latitudes

Page 223 : Wed Jan 02 02:42:19 2002

Acoustics

8- 2- 3- 3

DO NOT DISTRIBUTE

AP 3456

11. Pressure. Assuming constant density and a constant value for g (the acceleration due to gravity), then pressure varies
virtually linearly with depth. For 99% of the oceans, density remains within 2% of its mean value, and variations in g are very
5
2
much smaller than this so the linear relationship is a reasonable model. The change is approximately 10 Nm (1 bar or 1
atmosphere) per 10 m (33 ft).
12. Temperature. The only agent for heating the oceans is incoming solar energy and the majority of this is absorbed within
metres of the ocean surface. All of the infra-red radiation is absorbed within one metre and only about 2% of the incident energy
reaches 100 m. A small amount of heat is transmitted to depth by conduction but mixing caused by wind and waves is the main
mechanism for the transfer of heat. This turbulence generates a mixed surface-layer which may be up to 200 m thick depending
upon surface conditions, and therefore upon the season.
13. Permanent Thermocline. Below this mixed layer and down to about 1000 m the temperature falls rapidly. This region is
known as the permanent thermocline. Underneath this seasonal variations are virtually non-existent and there is a much
shallower temperature gradient with temperature falling to between 0.5C and 1.5C. This 3-layer model of the ocean
temperature structure is illustrated in Fig 3 showing the variations with latitude.
14. Season and Latitude Effects. In high latitudes (above 60N) the permanent thermocline is missing. In mid-latitudes the
classic profile may be amended by a seasonal thermocline when surface waters are heated in spring and summer.
15. Diurnal Effects. Diurnal variations in temperature are insignificantly small, usually less than 0.3C in the open oceans
although perhaps up to 3C in shallow coastal waters.

Currents
16. Deep Currents. As has been seen sea water is not homogenous - its temperature may vary considerably, and there may be
minor variations in salinity. In the Antarctic and around Greenland large masses of water are generated which are cold due to
the interaction with the atmosphere, saline due to the freezing-out of a proportion of the fresh water, and therefore somewhat
denser than average. This cold, dense water sinks and moves towards and across the Equator at depth. Clearly there will be
boundaries between this water and the other water masses that it encounters with different temperature and salinity
characteristics. Similar boundaries exist between other water masses with different properties. Examples are to be found where
water from the Baltic enters the North Sea and where Mediterranean water enters the Atlantic Ocean. These boundaries are
known as fronts and are analogous to meteorological fronts where different air masses meet.

8-2-3-3 Fig 4 Generalized Surface Currents

Page 224 : Wed Jan 02 02:42:19 2002

Acoustics

8- 2- 3- 3

DO NOT DISTRIBUTE

AP 3456

17. Surface Currents. Sea currents are generated by winds when some of the energy of air movement is transferred by friction
to the ocean surface. This leads to generalized zonal currents similar to the idealized coriolis-induced global patterns of surface
winds illustrated in meteorological texts, and although this simple model is considerably modified by the presence of land
masses, the influence of these winds is clearly discernable in Fig 4. As at depth there will be fronts between water masses of
differing characteristics, reflecting their different origins and histories. Whereas some of these fronts will be more or less
permanent, others will be of a temporary nature primarily due to seasonal effects.

Waves
18. Waves are another manifestation of the transfer of wind energy to the surface water. The precise mechanism of this transfer
is complex and as yet poorly understood. In the area of formation, waves that are generated will consist of a superimposition of
waveforms of varying frequency and amplitude depending upon the wind speed and the fetch (the length of sea area affected).
Higher wind speeds lead to higher waves, but eventually a state of equilibrium will be reached with the excess energy being
dissipated in, for example, white capping. The influence of waves is only felt close to the surface and waves generated by a
severe storm would be virtually unnoticeable below about 100 m. The characteristics of waves are illustrated in Fig 5.
19. Moving away from the area of generation the short period waves are dissipated, and only the long period waves remain.
These are known as swell and can travel large distances from the generating area, typically thousands of kilometres.
20. When waves enter shallow water they are affected by the sea bed once the depth is less than the wavelength. The front
of the wave becomes steeper and eventually breaks.

8-2-3-3 Fig 5 Characteristics of an Ocean Wave

Page 225 : Wed Jan 02 02:42:20 2002

Acoustics

8- 2- 3- 3

DO NOT DISTRIBUTE

AP 3456

21. The relationship between wind speed and sea state is expressed in the Beaufort Scale which may be used for estimating
wind speed at sea. It should be noted that the scale is valid only for waves generated locally, and providing adequate time has
elapsed and there has been an adequate fetch for a fully developed sea to become established. The Beaufort Scale is shown in
Table 1.

The Biological Environment


22. Life Forms. Ninety-eight percent of marine species belong to the benthic, or sea-bed living system. But it is the remaining
2% of species, the pelagic or free swimming system, which is of concern here. The life forms can be divided into two main
groups: the plankton and the nekton. (There is a third small group, the pleuston, which includes the jellyfish, which live at the
sea surface, but they are of no relevance here.)
23. The Plankton. The plankton are the group of organisms, both plant and animal, whose powers of locomotion are
insufficient to prevent them being transported by currents. They range in size from less than 2m up to just over 2 cms. As a
large proportion of the plankton (the phytoplankton) is responsible for photosynthesis they must exist in that part of the ocean
to which light can penetrate, typically the top 100 m or so. Virtually all plankton exhibit a diurnal variation of depth, rising near
the surface at night and descending again just before sunrise.
24. The Nekton. The nekton are those creatures which can swim with sufficient power to be more or less independent of water
movement. This group includes fish, cephalopods, and whales in the open ocean. Often they exhibit the same diurnal variation
in depth as the plankton on which they feed.
25. The Deep Scattering Layer. The diurnal depth variation of most plankton and nekton has already been described and this
event occurs within the top 100 m or so of the ocean. However it is worth noting that there are some mainly carnivorous
plankton and nekton which migrate between much greater depths. Typically they will spend the day at between 400 and 1000 m
and the night at about 200-300 m. The significance of this group of organisms is that they can affect echo-sounders and SONAR
equipment. It is thought that the reflection of sound waves is caused primarily by small fish possessing air bladders and by
prawn-like creatures. This phenomenon is known as the deep scattering layer and can be observed in all oceans.

Table 1 - Sea State and Beaufort Wind Scale

Page 226 : Wed Jan 02 02:42:21 2002

Acoustics

8- 2- 3- 3

DO NOT DISTRIBUTE

AP 3456

Acoustics

Page 227 : Wed Jan 02 02:42:22 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

Chapter 4 - Sound in the Sea


Introduction
1. Due to the rapid attenuation of other energy forms in the sea, eg light, the use of sound is one of the few methods available
for underwater detection. A study of the behaviour of sound in the sea is therefore important for operators of maritime patrol
aircraft.
2. Unlike light, which is a form of electromagnetic energy, sound involves the vibration of the medium through which it is
transmitted. So whereas light can pass through a vacuum, sound cannot, generally travelling best through solids and liquids and
somewhat less well through gases.
3.

The wavelengths of sound which are relevant in the ocean range from about 50 metres to 1 millimetre. Assuming a velocity
1
of sound in sea water of 1500ms
this corresponds to frequencies between 30 Hz and 1.5 MHz. Variations in the velocity of
sound in sea water and their effect on propagation are in fact very important and will be examined in some detail.

The Velocity of Sound in the Sea


4. In Chapter 1 a simple equation relating the velocity of sound in air to specific heat, pressure and density was given.
Unfortunately this simple relationship does not apply in liquids and although a mathematical treatment can be derived, it is rather
complex. However, it is possible to deal empirically with the variation of sound velocity with salinity, pressure and temperature.
An increase in any one factor will increase the sound velocity if all other factors remain constant. In line with current practice
changes are shown in units of feet, pounds per square inch (psi) and F rather than in SI units. The velocity of sound in sea
1
1
water varies between 4700 fts
and 5100 fts .
1
5. Salinity. A 1 O /oo increase in salinity increases the sound velocity by 4.27 fts . Unfortunately there is no method of
readily determining salinity from a maritime patrol aircraft but assuming a constant value of 35 O /oo is normally sufficient.
However, caution is needed in those areas where salinity values are liable to be markedly different from the average, eg in
coastal and ice melt areas.
6. Pressure. An increase in pressure of 44 psi, corresponding to a depth change of about 100 ft, will increase the sound
1
velocity by about 1.7 to 1.8 fts . Fortunately pressure increases with depth in a linear fashion and so velocity changes are
easily predicted with fair accuracy.

8-2-3-4 Fig 1 Sound Velocity Profile for a Typical Three Layer Ocean

Page 228 : Wed Jan 02 02:42:23 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

7. Temperature. Of the 3 factors being considered, changes in temperature have the greatest effect on sound velocity. A 1 F
1
increase in water temperature will cause an increase in sound velocity of between 4 and 8 fts .
8. Sound Velocity Profile. Fig 1 shows how sound velocity varies with depth in a typical Three Layer Ocean. This variation
with depth is known as a sound velocity profile (SVP). The following points can be deduced:
a. There is an increasing velocity down through the surface mixed layer due to the effect of increasing pressure
(Temperature constant).
b. There is decreasing velocity with depth in the thermocline due to decreasing temperature having a greater effect than
increasing pressure.
c. There is increasing velocity down through the deep water due to increasing pressure (Temperature constant).
9. Temperature Gradient and Sound Velocity. Temperature/depth profiles are readily obtained by MPA crews and the
following relationship between temperature profile and sound velocity can be expected:
a. If temperature decreases with depth at a rate of 0.2 to 0.4F/100 ft its effect will be offset almost exactly by the effect of
the increasing pressure with depth. This will result in an ISOVELOCITY condition (ie no change in velocity with depth).
b. If temperature decreases with depth at a rate of less than 0.2 to 0.4F/100 ft then sound velocity will increase with depth
and a POSITIVE velocity gradient will exist.
c. If temperature decreases with depth at a rate greater than 0.2 to 0.4F/100 ft then sound velocity will decrease with depth
and a NEGATIVE velocity gradient will exist.
d. If temperature remains constant with depth this is termed ISOTHERMAL and there will be velocity gradient of +2 (ie an
1
increase in velocity of 2 fts per 100 ft increase in depth).

Refraction
10. When sound leaves a source it can be considered to move along paths known as rays - analogous to light rays. As with
light, if the medium of transmission is homogenous, then these rays will be straight lines but if the ray moves into an area where
the velocity of sound is different then the ray will be bent, a phenomenon known as refraction.
11. The effect of refraction on a ray passing through layers of different velocity is illustrated in Fig 2 where the total ray path
will be seen to be made up of a series of straight line segments.

Page 229 : Wed Jan 02 02:42:23 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

12. A useful mnemonic is the HALT rule which states that rays will be refracted away from depths of relatively high velocity
and towards depths of relatively low velocity:
High Away, Low Towards
Thus in Fig 2 the ray moving from layer 1 to layer 2 (relatively low velocity) will be refracted towards layer 2 ie 2 > 1. On
reaching layer 3 (relatively high velocity) the ray will be refracted away from layer 3 ie 3 < 2.
13. Applying this rule to the typical SVP in Fig 1 it will be seen that sound rays in the mixed surface layer will be refracted
upwards away from the Sonic Layer Depth and sound rays in the thermocline will be refracted downwards until they pass the
depth of the Deep Sound Channel Axis when they will be refracted upwards again. The direction of propagation of a sound ray
will therefore be determined by the ambient SVP.
14. Snells Law. The degree to which a ray is bent when moving from an area of one velocity to another is determined by
Snells Law which states that:
In a medium comprising discrete layers each of different sound velocity, the angles 1, 2, 3 etc of a ray incident on and
leaving a boundary between layers are related to the sound velocities, C1, C2, C3 etc of these respective layers such that:
cos 1 cos 2 cos 3
=
=
= a constant for any given ray.
C1
C2
C3
15. Limiting Ray. When sound rays are refracted within a sea layer, one ray will eventually just graze or be tangential to the
boundary with an adjacent layer, or to the sea surface or ocean bottom. This ray, which will continue to be refracted within the
layer, is known as the limiting ray. Rays which approach the layer boundary at a more perpendicular angle, will pass through it
(or be reflected or absorbed).

8-2-3-4 Fig 2 Refraction in a Layered Medium

Shadow Zones
16. The HALT rule predicts that rays will be refracted away from depths of sound velocity maxima. As a result there often
exists at these depths a region into which very little acoustic energy can penetrate; such regions are termed Shadow Zones. The
existence of such a zone, and its relative position, depends upon the SVP. Minor amounts of energy will enter the shadow zone
due to diffractive and scattering effects. The nature of shadow zones in a variety of SVPs is shown in Fig 3 a-d.
a. Fig 3a Isovelocity Conditions. In isovelocity conditions the increase in sound velocity due to increasing pressure with
depth is offset by the effect of decreasing temperature with depth. Since there is no variation in velocity there is no
refraction and sound rays travel in straight lines from the source.
b. Fig 3b Negative Velocity Gradient. In this situation the temperature decreases at a rate sufficient to overcome the
pressure effect and so velocity decreases with depth. Sound rays are refracted downwards away from the higher velocity
area. A shadow zone exists above and beyond the limiting ray.

Page 230 : Wed Jan 02 02:42:23 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

c. Fig 3c Positive Velocity Gradient. In the situation where temperature is constant or increasing with depth, or at least
does not fall at a rate to override the pressure effect then a positive velocity gradient will be established. Sound rays will be
refracted upwards away from the higher velocity area and a shadow zone exists below and beyond the limiting ray.
d. Fig 3d Positive over Negative Velocity Gradient. The combination of a positive velocity gradient above a negative
velocity gradient produces a split beam effect at the depth of the velocity gradient change. The limiting ray and all rays
above it are refracted upwards away from the higher velocity area. All rays below the limiting ray are similarly refracted
downwards. A shadow zone will exist beyond the limiting ray(s).

Sound Transmission Modes (Paths)


17. Clearly sound energy can travel directly underwater from the source to the detector, affected only by bending due to
refraction. In general however, direct rays achieve only relatively short ranges. The range is normally determined by the
limiting ray and will depend primarily on the Sonic Layer Depth, the SVP below the surface mixed layer, and the source/receiver
depth combination:
a. Effect of Sonic Layer Depth. The deeper the Sonic Layer Depth, the greater will be the Direct Path range. This is
because the limiting ray has a greater vertical distance to travel before being refracted upwards and those rays at a greater
angle than the critical angle ray will also have a greater vertical distance to travel before passing through the layer.
b. Effect of SVP Below the Surface Mixed Layer. The greater the negative velocity gradient below the surface mixed layer,
the more pronounced will be the refraction and so the more reduced will be the direct path range.
c. Effect of Receiver/Source Depth Combination. If the receiver and source are in different layers then this will curtail the
direct path range to an extent, particularly if the receiver is shallower than the source. The greatest direct path range is
normally achieved when both source and receiver are in an isothermal layer. In general though the direct path range
depends upon the velocity gradient.
18. Phenomena. The variations in sound velocity with depth in the ocean lead to a number of effects which allow much greater
sound detection ranges to be achieved than would be possible if the direct path only was available. The infinite variety of sound
velocity profiles implies an infinite variety of transmission modes. However it is possible to group them all into four basic
phenomena; other modes tend to be just variations on these four. It should be remembered though, that unless transmitter and
receiver are very close together, then the received sound has very probably been transmitted by a combination of more than one
mode.
19. Surface Duct (Surface Sound Channel). This transmission mode, sometimes known as The Mixed Layer Sound Channel,
concerns only acoustic energy contained within the surface layer. The prime requisite for the existence of a surface duct is a
layer with a positive velocity gradient. Sound rays in the layer are refracted up towards the surface (away from the higher
velocity), reflected off the sea surface downwards and are then refracted upwards again and so on. This trapping of the sound
energy within the duct can result in extended ranges. The quality of sound transmission via this mode is dependent upon the
Sonic Layer depth, the sea state and the source depth. Whether this mode can be used depends primarily upon whether the
sensor can be placed in the layer and also upon the signal frequency. Surface Duct propagation is illustrated in Fig 4, and its
usefulness may be affected by the following factors:

8-2-3-4 Fig 3 Velocity Gradients

Page 231 : Wed Jan 02 02:42:23 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

8-2-3-4 Fig 4 Surface Duct

Page 232 : Wed Jan 02 02:42:23 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

a. Sonic Layer Depth. An effective Surface Duct cannot exist unless the Sonic Layer depth is greater than 50 feet. The
deeper the Sonic Layer depth the greater the range in the duct as the sound rays have further to travel before being reflected
off the surface. Each time a sound ray is reflected off the surface some energy is lost by scattering so for any given range
the reflection losses are less for a deep layer.
b. Sea State. Wind, waves and swell all increase absorption, reflection and scattering losses as well as increasing the
ambient noise by inducing air bubbles into the water. These factors will all reduce the range achieved and the effectiveness
of the duct.
c. Velocity Gradient in the Surface Layer. In order to have an effective surface duct the sound velocity gradient within the
surface layer must be positive. The more positive the gradient the less sound energy that can escape and the better the
detection ranges. Isovelocity conditions represent the limit of a positive gradient and result in straight line propagation, the
limiting type of Surface Duct.
20. Sound Channels. The pre-requisite for a Sound Channel is a region of decreasing sound velocity overlaying a region of
increasing sound velocity. Applying the HALT rule will show that sound rays will tend to be refracted towards the depth of
minimum sound velocity. This depth is known as the Sound Channel Axis. A Sound Channel is illustrated in Fig 5. The
following two types of sound channel are recognized:

8-2-3-4 Fig 5 Sound Channel

a. Deep Sound Channel. This is also known as the SOFAR Channel (SOund Fixing And Ranging) and is the most
common type of Sound Channel. It results from the negative velocity gradient of the thermocline overlaying the positive
velocity gradient established by the pressure effect in deep water. The axis of the Deep Sound Channel is found at the depth
of minimum sound velocity, usually at the base of the permanent thermocline. The SOFAR axis depth varies
geographically. In the Atlantic Ocean it varies between 1,300 ft in high latitudes (above 60 N) and 6,600 ft in low latitudes
(about 35 N). The depth also decreases uniformly with longitude away from the Greenwich Meridian in the North
Atlantic.

Page 233 : Wed Jan 02 02:42:24 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

b. Shallow Sound Channel. This is sometimes known as a Depressed Sound Channel or Sub-Surface Duct. This type of
sound channel occurs in the upper layers of water above the permanent thermocline but is somewhat transient in nature. In
order to be effective the axis depth of a Shallow Sound Channel must be between 150 and 600 feet below the surface. The
channel must be at least 50 feet thick and the limiting rays must be at an angle of at least 2 above and below the channel
axis. Shallow Sound Channels exist in the Mediterranean and the North-East Atlantic at certain times of the year and are
present in the lower latitudes of the Atlantic all year. In the North Atlantic they are only established during the summer
months, with an axis depth of about 450 feet. The usability of both types of sound channel is determined by the proximity
of the receiver to the axis depth and by the signal frequency.
21. Convergence Zone. Sound rays leaving a source near the surface which penetrate below the surface layer, will be refracted
downwards until they pass the depth of the deep sound channel axis whence they begin to be refracted upwards towards the
surface. Provided that the water depth is great enough a sufficient number of rays will eventually reach the near surface depths
as a ring of focused energy around the source known as the annulus. This is illustrated in Fig 6. The likelihood of Convergence
Zone propagation can be determined from a sound velocity profile using the following parameters which are shown graphically
in Fig 7:
a. Velocity at Source (Vs). This is the sound velocity at the source depth.
b. Velocity at Bottom (Vb). This is the sound velocity at the ocean bottom depth and is used to determine whether or not
sound rays will be refracted back upwards towards the surface. Before any rays will be refracted upwards, the velocity at
the bottom (Vb) must exceed the sound velocity at the source depth (Vs).
c. Velocity Excess (Vx). If Vb is greater than Vs then the difference is known as the Velocity Excess. For a reliable
1
Convergence Zone this velocity excess must be approximately 33 fts .

8-2-3-4 Fig 6 Convergence Zone Propagation

Convergence Zone range in the UK MPAs normal operating areas is between 20 and 32 nautical miles. It is unusual for
Convergence Zone propagation to take place in warm or moderately warm water in depths of less than 1,200 fathoms. However,
Convergence Zone can exist in water depths of less than 300 fathoms in certain conditions. Convergence zone width is directly
proportional to the depth excess and can be roughly estimated at 10% of the range interval. It is possible to have a number of
Convergence Zones around a single source. The first zone will be the strongest in terms of intensity. The second and third
zones occur, respectively, at twice and three times the range of the first, providing that the source is strong enough.

8-2-3-4 Fig 7 Convergence Zone Velocity Profile

Page 234 : Wed Jan 02 02:42:24 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

22. Bottom Bounce Mode. Sound rays which leave a source at angles greater than that of the Limiting Ray will eventually
strike either the ocean bottom or the sea surface. The Bottom Bounce Mode refers to those rays which are reflected back and
forth between these two boundaries. Whereas some energy will be reflected off the boundary surface (either ocean bottom or
sea surface), a proportion will be lost to sensors due to scattering and absorption. The quality of this transmission mode is
primarily dependent on factors such as boundary surface roughness, bottom type, water depth and signal frequency. Bottom
Bounce Mode is particularly important in shallow water or where the Sonic Layer depth is less than 50 feet. In fact sound
propagation for frequencies between 50 Hz and 1,500 Hz in shallow water is dominated by the Bottom Bounce transmission
mode.

Leakage Paths
23. In general sound channels or ducts are low loss propagation paths. At low frequencies, however, it is found that losses
increase suddenly and the duct breaks down. The frequency at which this effect occurs is known as the cut-off frequency. In
simple terms the wavelength becomes too big to fit into the duct. The leakage path is in the direction of propagation and the
main cause of the leakage is diffraction together with some scattering losses. In Chapter 1, when diffraction was being
considered, it was shown that according to Huygens principle secondary wavelets are produced from each point on a
wave-front. Although the vast majority of the sound energy is propagated in the forward direction, these wavelets have a
component of their energy to the side and this energy can leak into shadow zones or out of a sound channel. Leakage paths
tend to complicate the relatively simple propagation paths that have been outlined. Two examples are described below:
a. Leakage Effect on Convergence Zone. Fig 8a illustrates the effect of leakage on convergence zone propagation.
Whereas some energy will enter the convergence zone path immediately, some may remain trapped in the surface duct for
some range before leaking into a different convergence zone path. This anomalous propagation can lead to confusion in
recognizing the convergence zone and the width of the annulus.
b. Leakage Effect on Bottom Bounce. Fig 8b illustrates the effect of leakage on bottom bounce propagation. Again some
energy is transmitted directly to the bottom while some remains trapped in the surface duct for a while before entering a
bottom bounce path.

Coherency Effect (Lloyds Mirror)

Page 235 : Wed Jan 02 02:42:24 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

24. Clearly it is possible for sound from one source to arrive at a receiver having followed a variety of paths.

8-2-3-4 Fig 8 Leakage Effects

In general any two sound paths will be of different lengths and so it is highly likely that the phase of the sound arriving by one
path will be different to that of sound arriving by another. If the received sound waves are in phase then there will be an
enhancement of the signal. If, however, the phases are different then there will be a reduction in the received signal with
complete cancellation if the waves are 180 out of phase. Commonly this effect is a result of interference between a direct wave
and a surface reflected wave or in shallow water between a direct wave and a bottom reflected wave. A similar effect can occur
if some sound is reflected by an object in the water and therefore arrives out of phase with the direct wave (secondary source
polarisation). Lloyds mirror effect is illustrated in Fig 9.

8-2-3-4 Fig 9 Lloyds Mirror Effect

Page 236 : Wed Jan 02 02:42:24 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

Diurnal (Afternoon) Effect


25. Solar heating near the surface water can cause short term variations in the vertical temperature structure which can in turn
bring about changes in sound propagation paths. These diurnal variations can lead to shortening of detection ranges. As the
surface water is heated during the day a negative temperature profile builds up causing refraction of the sound energy and the
establishment of new shadow zones. As the surface water cools again the temperature gradient and its associated effects
disappear.

Noise
26. Ambient Noise. Ambient noise is that unwanted sound which is inherent in the sea independent of either the target or the
aircraft. It can of course mask the noise generated by a target which the sensors are attempting to detect. Ambient noise can be
predicted by graphical or computer methods both of which depend upon a mix of theoretical and empirical data, or it can be
measured using an ambient noise meter. There are three main sources of ambient noise in the areas where the UK MPA
normally operate.
a. Shipping Traffic. Shipping tends to generate the main component of ambient noise in deep water areas, especially at
lower frequencies (below 500 Hz). It is of course most intense in the vicinity of major shipping lanes. It results from the
sound of machinery, electrical systems, hydraulics, hydrodynamic flow and propeller cavitation.
b. Sea State. This noise source is the result of sea surface agitation causing the entrapment and subsequent escape of air
bubbles. This noise is dominant at higher frequencies (above 500 Hz). Sea state noise is increased by strong surface winds
especially in shallow water.
c. Biological. Some crustaceans, fish and marine mammals contribute to ambient noise by producing sounds associated
with their normal life activities. Biological sounds are present over a large frequency spectrum and are extremely difficult
to predict with any accuracy due to such factors as changing biological cycles and movement of the source organisms; there
are often seasonal and diurnal cycles. In general biological noise is greatest in shallow water, particularly in tropical and
sub-tropical seas.
27. Self-noise. Self-noise is a noise generated by a maritime patrol aircrafts own mechanical, avionic, and electrical systems.

Sound Transmission Losses


28. Not all of the sound generated by a source reaches a sensor. There are two general ways in which acoustic energy may be
dissipated during propagation: spreading and attenuation.
29. Spreading Losses. Spreading is a geometrical phenomenon whereby a fixed amount of energy is distributed over an ever
increasing area as it moves away from the source. There are two types of spreading loss which may be encountered.
a. Spherical Spreading. If the sound velocity in the ocean is constant in both the horizontal and vertical directions then the
sound energy will radiate equally in all directions from the source. At any given distance, r, from the source, the power of
2
the source, W, will be spread over the surface of a sphere of radius r. This surface area is 4r . The sound intensity, I,

Page 237 : Wed Jan 02 02:42:25 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

which is the sound power per unit area will be given by:
W
Wm2
4r2

I=

This is an inverse square law, ie intensity is inversely proportional to the square of the range. Notice that it is frequency
independent. Working in Sound Pressure Levels in dBs then the spherical spreading loss = 20 log r and SPL is inversely
proportional to the range. If the range is doubled from r to 2r then:
Change in SPL = 20 log [(1/r)/(1/2r)]

= 20 log (1/r) 2r

= 20 log 2

= 6dB

ie a doubling of distance results in a 6dB loss.


b. Cylindrical Spreading. Cylindrical spreading occurs when the acoustic energy is trapped between two parallel
boundaries such as the upper and lower bounds of a sound channel, surface duct or a convergence zone. The energy is
constrained to spread in only two dimensions in a cylindrical manner. In this case at any given distance from the source, r,
the power of the source, W, will be spread over the surface of a cylinder or radius r. The area in this case will be 2r and
the variation of sound intensity with range is given by:

I=

W
Wm2
2r

The spreading loss in dBs = 10 log r as sound pressure is inversely proportional to the square root of r. So looking again at
the loss involved in a doubling of range:
1

Change in SPL = 20 log [( 1/r 2 ) / (1/(2r) 2 )]

= 20 log [1/r 2 (2r)2 ]


1

= 20 log 2 2

= 3dB

ie doubling of distance results in a 3dB loss.

Page 238 : Wed Jan 02 02:42:25 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

30. Spreading Loss Comparisons. Fig 10 shows a comparison between the different types of spreading loss. Since cylindrical
loss is less severe than spherical loss, ducted modes of sound propagation often yield the highest probabilities of signal
detection. However, most of the time the received signal has suffered some combination of the spreading losses. For example,
since direct path makes up the initial portion of ALL transmission modes, they all exhibit some degree of spherical spreading.
31. Attenuation Losses. Attenuation is the term applied to the linear decrease in acoustic energy per unit area of a wave-front as
the distance from the source increases. Attenuation loss includes the effects of absorption, scattering and diffractive leakage.
a. Absorption Loss. Absorption loss involves the conversion of acoustic energy into heat by molecular action. Absorption
loss is directly proportional to the range from the source. Increases in salinity and decreases in temperature increase
absorption losses. Unlike spreading losses, absorption depends upon the frequency, varying approximately as the square of
the frequency.

8-2-3-4 Fig 10 Spreading Loss Graph

b. Scattering Loss. Scattering is the random reflection of acoustic energy from the ocean surface, the ocean bottom, or
from suspended particles (volume scattering). Factors influencing the degree of each type of scattering are:
(1) Surface Scattering. The severity of surface scattering loss is dependent upon wave height, signal frequency, and
angle of incidence of the sound energy.
(2) Bottom Scattering. The severity of bottom scattering is dependent upon the bottom roughness, sediment particle
size, signal frequency, and angle of incidence of the sound energy.
(3) Volume Scattering. The severity of the scattering loss due to volume scattering is the most difficult to predict. It is
dependent upon the ratio of the particle size responsible for the reflection to the signal wavelength. It is also dependent
on the type of particle (ie solid or fluid). The most important contributor to volume scattering is biological in nature.
Apart from the effect of the Deep Scattering Layer, volume scattering strength tends to decrease with depth.
The prediction of actual losses due to scattering is problematical, but there are some general observations that can be made:
higher scattering losses are associated with higher frequencies, rougher scattering surfaces and larger scattering particles. Fluids
such as bubbles are generally more effective volume scatterers than solid particles.

8-2-3-4 Fig 3a Isovelocity Conditions

Page 239 : Wed Jan 02 02:42:25 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

8-2-3-4 Fig 3b Negative Velocity Gradient

8-2-3-4 Fig 3c Positive Velocity Gradient

8-2-3-4 Fig 3d Positive over Negative Velocity Gradient

Page 240 : Wed Jan 02 02:42:26 2002

Acoustics

8- 2- 3- 4

DO NOT DISTRIBUTE

AP 3456

8-2-3-4 Fig 8a Leakage Effect on Convergence Zone

8-2-3-4 Fig 8 Leakage Effect on Bottom Bounce

Mechanics

Page 241 : Wed Jan 02 02:42:26 2002

Mechanics

8- 2- 4- 1

DO NOT DISTRIBUTE

AP 3456

Chapter 1 - Statics
Introduction
1. Statics is the study of forces in equilibrium. A particle is in equilibrium when all the forces acting upon it are balanced; it
may be stationary or it may be in a state of unchanging motion. The term force and some associated terms are discussed below.

Forces, Moments and Couples


2. Force. A force is that quantity which when acting on a body which is free to move, produces an acceleration in the motion
of that body. For example, if a stationary football is kicked, it moves, ie it experiences an acceleration; the kick is the applied
force. If a moving football is kicked it changes speed, or direction, or both, so again it experiences an acceleration. However,
just because a body is not accelerating does not mean that there are no forces acting upon it. Although that may be the case, it
may equally be the situation that there are several forces acting such that their resultant is zero. For example an aircraft which is
flying at constant altitude, speed and direction has many forces acting upon it, principally lift, thrust, weight and drag.
3. Weight. Any body on or near the Earth is subject to the attraction of Earth's gravity, and if a body is released in this
situation it will accelerate towards the centre of the Earth. The value of the acceleration, which is usually given the symbol g, is
2
approximately 9.8 ms but varies slightly both geographically and with altitude. Since gravity imposes an acceleration on a
body it must be a force, and it is this force which is known as weight.
4. Vector Representation of a Force. Force is a vector quantity, ie it has both magnitude and direction. A vector may be
represented by a straight line drawn to scale and marked with an arrow. The direction of the line represents the direction of the
vector and its length represents the magnitude. Vectors may be added together by means of the triangle, parallelogram, and
polygon laws to give a single resultant and a single vector may be resolved into two or more components. A more complete
treatment of vectors is to be found in Vol 8, Pt 1, Sec 1, Ch 4.
5. Moment or Torque. The moment of a force about a point, or torque, is the tendency of the force to turn the body to which it
is applied about that point. The magnitude of the moment or torque is the product of the force and the perpendicular distance
from the point to the line of the force. The moment of a force about an axis is the product of the force and the length of the
perpendicular common to the line of the force and the axis.
6. Couple. A system of two equal and parallel forces, acting in opposite directions but not in the same line, is known as a
couple. The moment of a couple about any point in the plane of the forces is constant and equal to the product of one of the
forces and the perpendicular distance between their lines of action.

Centre of Gravity
7. In any rigid extended body there is a unique point at which the total gravitational force, the weight, appears to act. This
point is known as the centre of gravity.
8. The position of the centre of gravity of a flat body can be determined by suspending it at any point, P and marking the
vertical, then suspending it at a second point, Q, and again marking the vertical. The centre of gravity is at the intersection of the
two lines (Fig 1).
9. The position of the centre of gravity of a composite body which can be treated as a number of symmetrical parts can be
found as shown in the following example. A body consists of two uniform spheres mounted on a uniform cylindrical bar as
shown in Fig 2. Taking 0 as the reference point, the parts can be treated as three cylinders with centres of gravity distances 1,
11, and 19, metres respectively from the datum, and weights 10, 50, and 10 units; and two spheres of weights 200 and 25 units,
with centres of gravity 4 and 17 metres respectively from the datum. Let the centre of gravity, through which the total weight
(295 units) of the body acts be x metres from 0. Taking moments about 0:
(10 1) + (200 4) + (50 11) + (25 17)

+ (10 19) = 295x or x = 6.69 metres.

8-2-4-1 Fig 1 Determining the Centre of Gravity of a Flat Plate

Page 242 : Wed Jan 02 02:42:26 2002

Mechanics

8- 2- 4- 1

DO NOT DISTRIBUTE

AP 3456

8-2-4-1 Fig 2 Determining the Centre of Gravity of a Symmetrical Body

Equilibrium
10. Two types of equilibrium condition may be recognized, translational equilibrium and rotational equilibrium.
11. Translational Equilibrium. An object is said to be in translational equilibrium if it has constant velocity (including velocity
equal to zero). In order to achieve this condition the vector sum of all of the forces acting upon the object must be zero. It is
usually convenient to consider the components of the forces in three orthogonal directions (x, y, and z axes). In this case the
algebraic sum of the x, y and z components must each equal zero. As an example consider Fig 3a which shows a uniform
rectangular body being supported by two ropes. Three forces are acting on the body; the tension in the ropes, F 1, and F2, and
the weight, F3. All three forces may be regarded as coplanar and as acting through the centre of gravity and this simplified
situation is shown in the vector diagram, Fig 3b. For translational equilibrium the x component of F1 must be equal and
opposite to the x component of F2; F3 has no x component. This equality can be shown by constructing verticals from the ends

Page 243 : Wed Jan 02 02:42:27 2002

Mechanics

8- 2- 4- 1

DO NOT DISTRIBUTE

AP 3456

of the vectors to the x axis. Also the sum of the y components of F1 and F2 must be equal and opposite to F3. The y
components may be determined by constructing horizontals from the ends of the vectors to the y axis.
12. Rotational Equilibrium. An object is in rotational equilibrium when it rotates about an axis of constant direction at a
constant angular speed, including zero angular speed. The condition of rotational equilibrium is achieved if the vector sum of all
the torques about the axis is zero, ie the sum of the clockwise moments has the same magnitude as the sum of the anticlockwise
moments.

8-2-4-1 Fig 3 Translational Equilibrium

8-2-4-1 Fig 4 Rotational Equilibrium

13. As an example consider the uniform bar shown in Fig 4 which is free to rotate about the axle. Three forces, F1 F2, and F3,
are shown acting on the bar at distances L1, L2, and L3 respectively from the axle. For rotational equilibrium the relationship:
F1 .L1 + F2 .L2 F3 .L3 = 0
must be satisfied.

Stability
14. Three conditions of stability with respect to equilibrium may be recognized as follows:
a. Stable Equilibrium. An object is in stable equilibrium if any small displacement caused by external forces or torques

Page 244 : Wed Jan 02 02:42:27 2002

Mechanics

8- 2- 4- 1

DO NOT DISTRIBUTE

AP 3456

tends to be self-correcting. An example would be a marble in the bottom of a round bottomed cup. If an external force
disturbs its equilibrium it will, after a few oscillations, settle back to its original position.
b. Unstable Equilibrium. An object is in unstable equilibrium if any small displacement tends to be escalating. A marble
balanced on the end of a finger would be an example. Any small displacement would cause the marble to move to a totally
different position.
c. Neutral Equilibrium. Neutral equilibrium is an intermediate state between stable and unstable equilibrium. Consider a
marble at rest on a flat horizontal surface. A small displacement will cause the marble to move but the resulting position is
unchanged from an equilibrium point of view from the original position. It will have no tendency either to be displaced
further, or to return to its original position.

Friction
15. When two solid surfaces which are in contact move, or tend to move, relative to each other, a force acts in the plane of
contact of the surfaces in a direction opposing the motion. This force is known as friction and is a result of interactions between
the molecules of the two surfaces.
16. Dynamic Friction. If an object is sliding over a surface at a constant velocity then the friction is an example of dynamic
friction and there will be some conversion of the kinetic energy of the object into heat. The force of dynamic friction, fd,
depends upon the material of the two surfaces, on their smoothness and on the component of the force, F, that presses the two
surfaces together. The frictional force is practically independent of the relative velocity of the two surfaces. It has been found
that:
fd = u d F
where ud is a constant for a given pair of surfaces and is known as the coefficient of dynamic friction. The value of ud varies
from about 0.06 for a smooth steel surface sliding on ice to 0.7 for rubber sliding over dry concrete.
17. Static Friction. If an object is placed on a plane whose angle of inclination can be increased it is found that the object
remains stationary until a certain angle of inclination is reached whereupon the object starts to move. The force preventing the
object from moving is known as static friction. As the plane's inclination is increased then the value of the component of the
weight parallel to the plane increases. As the friction force is equal and opposite to this force (otherwise the object would
move), then it too must increase as the inclination is increased until it reaches a critical angle. As with dynamic friction the
maximum force of static friction, fs max depends upon the materials, the smoothness of the surfaces in contact and on the
component of the force, F, pressing the two surfaces together. For any pair of surfaces it is found that:
fs max = us F
where us is constant for any two materials of specified smoothness and is called the coefficient of static friction. It varies from
about 0.1 for steel on ice to about 1 for rubber on dry concrete. The coefficient of static friction is always higher than that of
dynamic friction, which is why objects tend to start moving with a jerk.
18. Applicability. It should be noted that the relationships described above are derived from observations rather than from any
theoretical understanding of the mechanisms causing friction. Thus the equations do not have universal applicability; deviations
occur, for example, at extreme speeds, when the surfaces in contact are very small and when the force pressing the surfaces
together is very large.

Machines
19. A machine is any device which enables energy to be used in a convenient way to perform work. Typical examples of
simple machines are levers, winches, inclined planes, pulleys and screws. Machines do not save work; in general they allow a
smaller force to be applied in order to achieve a result, but the smaller force must be applied over a greater distance. As
examples of machines, the lever and a pulley system will be reviewed.
20. The Lever. A lever can be described as a rigid beam supported at a point or fulcrum that is fixed, and about which the beam
can turn. The arrangement is shown in Fig 5 where the purpose of the lever is to lift a load of weight W. A force F is applied at
the opposite end such that the lever is maintained in a horizontal position. In this situation the system is in rotational equilibrium

Page 245 : Wed Jan 02 02:42:27 2002

Mechanics

8- 2- 4- 1

DO NOT DISTRIBUTE

AP 3456

and so the moments about the pivot must be equal and opposite. The anticlockwise moment is W.11 whilst the clockwise
moment is F.11. Therefore if 12 is greater than 11 the force required to balance the load is less than the load. However, as
shown in Fig 6, in order to raise the load over a distance, h, the smaller force must be applied over a greater distance, h1.

8-2-4-1 Fig 5 Lever in Rotational Equilibrium

8-2-4-1 Fig 6 Lever - Smaller Force over Greater Distance

21. A Pulley System. A two pulley system is shown in Fig 7 in which a force, F, is applied through a distance, a, in order to
raise a weight, W, through a distance, b. From the principle of the conservation of energy, the potential energy gained by the
load will equal the work done by the effort, ignoring incidental energy losses (eg friction). Therefore:

Wb = Fa, or

W a
=
F
b

As with the lever a load can be lifted with a smaller force but that force must be applied over a greater distance. The ratio W/F is
known as the mechanical advantage; for the pulley system shown the mechanical advantage is 2.

8-2-4-1 Fig 7 A Two Pulley System

Page 246 : Wed Jan 02 02:42:27 2002

Mechanics

8- 2- 4- 1

DO NOT DISTRIBUTE

AP 3456

8-2-4-1 Fig 3a

8-2-4-1 Fig 3b

Page 247 : Wed Jan 02 02:42:28 2002

Mechanics

8- 2- 4- 1

DO NOT DISTRIBUTE

AP 3456

Mechanics
Chapter 2 - Kinematics
Introduction
1. Kinematics is the study of motion without reference to the forces involved. In this chapter linear and angular motion will be
examined.

LINEAR MOTION
Speed and Velocity
2. Speed is the ratio of the distance covered by a moving body, in a straight line or in a continuous curve, to the time taken.
The velocity of a body is defined as its rate of change of position with respect to time, the direction of motion being specified. If
the body is travelling in a straight line it is in linear motion, and if it covers equal distances in equal successive time intervals it is
in uniform linear motion.
3.

For uniform velocity, where s is the distance covered in time t, the velocity v is given by:
v=

4.

s
t

In the more general case, the instantaneous velocity vi is given by:

vi =

ds
dt

Acceleration
5. The acceleration of a body is its rate of change of velocity with respect to time. Any change of either speed or direction of
motion involves an acceleration; a retardation is merely a negative acceleration.

Page 248 : Wed Jan 02 02:42:28 2002

Mechanics

8- 2- 4- 2

DO NOT DISTRIBUTE

AP 3456

6. When the velocity of a body changes by equal amounts in equal intervals of time it is said to have a uniform acceleration,
measured by the change in velocity in unit time.
7. If the initial velocity u of a body in linear motion changes uniformly in time t to velocity v, its acceleration a is given
by:

a=

(v u)
t

Vector Representation
8. Velocity and acceleration are vector quantities and the laws of vector addition may be applied. Thus the resultant velocity
of a body having two separate velocities (eg an aircraft flying in wind) may be found by the parallelogram law. Also a single
velocity may be resolved into two or more components.

Relationship between Distance, Velocity, Acceleration and Time


9. Some useful formulae can be derived relating distance, velocity, acceleration and time. When dealing only with motion in a
straight line, the directional aspects of velocity and acceleration can be ignored, apart from the use of positive and negative signs
to indicate forward and backward motion.
10. Consider a body with initial velocity, u, which in time t attains, under uniform accleration a, a final velocity v. Suppose
that the distance covered during this time is s. It has been stated that:

a=

(v u)
t

. _ . at = v u or

v = u + at

(1)

The mean or average velocity of the body is (u + v)/2. Therefore the distance covered in time t will be given by:

s=

(u + v)
.t
2

(2)

Substituting for v from equation (1),

s=

1
(u + u + at).t or
2

s = ut +

1 2
at
2

(3)

From equation (1), t = (v u)/a. Substituting for t in equation (2),

s=

Page 249 : Wed Jan 02 02:42:28 2002

1 (u + v)(v u)
or
2
a

Mechanics

8- 2- 4- 2

DO NOT DISTRIBUTE

AP 3456

v2 = u2 +2as (4)
11. Note that if the distance travelled in time t is denoted by s, velocity can be obtained by differentiation:

v=

ds
dt

Similarly, the acceleration a is given by:

a=

dv d2 s
=
dt dt2

Conversely, given an expression for acceleration, integration will give an expression for velocity, and further integration an
expression for displacement.
12. Velocity-Time Graphs. The linear motion of a body can be illustrated by means of velocity-time graphs, examples of which
are shown in Fig 1. In each case the distance travelled by the body between times t1 and t2 is represented by the area under the
corresponding part of the graph (shaded in Fig 1). The instantaneous acceleration at any time is given by the slope of the curve
at that point.

8-2-4-2 Fig 1 Velocity - Time Graphs

Page 250 : Wed Jan 02 02:42:28 2002

Mechanics

8- 2- 4- 2

DO NOT DISTRIBUTE

AP 3456

Relative Velocity
13. It is sometimes necessary to determine the velocity with which one moving body appears to be moving with respect to
another. This is known as the relative velocity. For linear motion this type of problem may be solved graphically, as in Fig 2,
by drawing from an origin a vector representing the velocity of body A, and from the end of this vector drawing a vector to
represent the velocity of B reversed. In Fig 2 the third side of the triangle, ob, represents the velocity of A relative to B.

Page 251 : Wed Jan 02 02:42:28 2002

Mechanics

8- 2- 4- 2

DO NOT DISTRIBUTE

AP 3456

8-2-4-2 Fig 2 Relative Velocity

ROTARY MOTION
Angular Velocity
14. So far only motion in a straight line has been examined. Consider now a point P which moves in a circle of radius r at
constant speed v. Its angular velocity is given by d/dt where is the angular displacement in radians (Fig 3a). The speed v is
given by:

v = r.

d
= r!
dt

15. Although the velocity of the point is constant in magnitude, it is constantly changing in direction, and therefore, by
definition, P is subject to an acceleration. Consider the point P at a certain instant (position P 1 in Fig 3a), and also after a small
interval of time dt (position P2). It is clear that the vectors representing motion at these two instants are of the same length, but
are in different directions. They are represented by AB and AC in Fig 3b, BC representing the change in velocity in the time t.

Page 252 : Wed Jan 02 02:42:28 2002

Mechanics

8- 2- 4- 2

DO NOT DISTRIBUTE

AP 3456

The acceleration is given by BC/t.


16. Since t is small, the angular displacement, , is also small, and BC may be considered to be almost perpendicular to both
AB and AC, that is, directed towards the centre of the circle.
BC = v..

But =

P1 P2
r

If the time interval t is very small, the straight line P1P2 is very nearly equal to the distance P1P2 along the arc, which is v.t.

. _ . =

v.t
radians
r

. _ . BC =

v2 t
r

Acceleration =

BC
t

v2
or v!, or ! 2 r, all towards the centre.
r

2
17. To summarize, a body moving at constant speed v in a circle of radius r has a constant acceleration of v /r directed towards
the centre of the circle.
18. The following formulae for circular motion, similar to those for linear motion, may be derived:
For uniform angular velocity, = t.
For uniform angular acceleration,
! 2 = ! 1 + t (cf v = u + at)

! 1 + ! 2 )t
and
2

1
1
= ! 1 t + t2 (cf s = ut + at2 )
2
2
and
2

! 2 = ! 21 + 2! (cf v = u2 + 2as)

Page 253 : Wed Jan 02 02:42:29 2002

Mechanics

8- 2- 4- 2

DO NOT DISTRIBUTE

AP 3456

where
1 = initial velocity in radians per sec
2 = velocity in radians per sec after t sec
2
= angular acceleration in radians per sec
= angle through which turned, in radians

Relationship between Angular and Linear Velocity


19. Consider again the point P rotating with uniform angular velocity w and radius r (Fig 3a). The length of the arc P1P2 is the
radius multiplied by the angle in radians, ie:
arc P1 P2 = r
But for uniform angular velocity,
= !t, hence
arc P1P2 = rt and the linear velocity of P is rt, or r.
Similarly, the linear acceleration of P is equal to the angular acceleration times the radius; its direction is radically towards the
centre of rotation.

8-2-4-2 Fig 3a Angular Velocity

Page 254 : Wed Jan 02 02:42:29 2002

Mechanics

8- 2- 4- 2

DO NOT DISTRIBUTE

AP 3456

Summarizing, if
v = linear velocity

! = angular velocity

a = linear acceleration, and

= angular acceleration,

v = !r, or ! =

v
r

a = r, or =

a
r

8-2-4-2 Fig 3b Acceleration Towards the Centre

Page 255 : Wed Jan 02 02:42:29 2002

Mechanics

8- 2- 4- 2

DO NOT DISTRIBUTE

AP 3456

8-2-4-2 Fig 3 The Acceleration Experienced in Uniform Circular Motion

Vector Representation of Angular Quantities

Page 256 : Wed Jan 02 02:42:29 2002

Mechanics

8- 2- 4- 2

DO NOT DISTRIBUTE

AP 3456

20. Angular velocity and acceleration are vector quantitites. By convention they are represented by vectors; the vector length
represents the magnitude of the quantity and its direction is perpendicular to the plane of rotation, ie parallel to the axis of
rotation. The direction of the arrow is such that, on looking in its direction, the rotation is clockwise (Fig 4). This convention is
known as the right-handed screw law. Such vectors can be combined and resolved according to the normal principles of vectors.

8-2-4-2 Fig 4 Vector Representation of Angular Quantities

Mechanics
Chapter 3 - Dynamics
Introduction
1. Dynamics is the study of motion related to force. In Chap 2 it was shown how a body moves; in this chapter it will be
shown why the body moves in that way.

QUANTITIES
Mass
2. When an object is at rest a force is necessary to make it move; similarly a body in motion needs a force to be applied in
order to change its motion. This reluctance to any change in motion is called inertia and the property that gives rise to inertia is
mass. It can be shown that inertia is directly proportional to mass.
3. The mass of a body may be defined as the quantity of matter in the body. The unit of mass is the kilogram (kg) and the
standard kilogram is a cylinder of platinum-iridium alloy kept at the Bureau International des Poids et Mesures.

Density
4.

The density of a substance is the mass per unit volume of that substance. The unit of density is the kilogram per cubic metre
3
(kg m ). Relative density (or specific gravity) is the ratio of the density of a substance at a stated temperature to the density of
water at 4C.

Page 257 : Wed Jan 02 02:42:29 2002

Mechanics

8- 2- 4- 3

DO NOT DISTRIBUTE

AP 3456

Momentum
5. The momentum of a body is defined as the product of its mass and its velocity. As the definition includes the velocity term,
momentum is a vector quantity and a change in either speed or direction constitutes a change in momentum. The unit of
1
momentum is the kilogram metre per second (kg ms ).

Force
6.

Sir Isaac Newton propounded three laws of motion:


a. First Law. A body remains in a state of rest or uniform motion (ie no acceleration) unless acted upon by an unbalanced
force.
b. Second Law. The rate of change of momentum is proportional to the applied force, and the change of momentum takes
place in the direction of the applied force (ie FRma).
c. Third Law. Whenever a body exerts a force on another body, the latter exerts a force of equal magnitude and opposite
direction on the former.

7.

By selecting a unit of force as that force which gives unit acceleration to unit mass the second law may be written as:
F = ma

2
The unit of force is the newton (N) which is the force necessary to induce an acceleration of 1ms in a mass of 1 kilogram.

Conservation of Momentum
8. Consider two bodies travelling in the same direction which collide, the duration of the collision being the short time t.
Throughout the collision each will experience a force equal and opposite to that experienced by the other (Newton's third law).
The impulse of the force is Ft, and is the same for each body. Thus the change of momemtum will be the same for each body. If
at the time of collision body A was overtaking body B, it is apparent that the effect of the impact will be to decrease the
momentum of A and increase that of B, and the total momentum of the system of two bodies will be unchanged.
9. The Law of Conservation of Momentum. The effects of the inter-action of parts of a closed system are summarized in the
Law of Conservation of Momentum, which states that the total momentum in any given direction before impact is equal to the
total momentum in that direction after impact.

Work
10. A force is said to do work when its point of application moves, and the amount of work done is the product of the force and
the distance moved in the direction of the force.
11. The unit of work is the joule (J) which is the work done when a force of 1 newton moves 1 metre in the direction of the
force.
12. The gravitational force acting on a body is the product of its mass and the acceleration of gravity, ie mg. Therefore the
work done against gravity in raising a body through a vertical distance, h, is mgh.

Energy
13. The energy possessed by a body is its capacity to do work; the unit of energy is the joule. A body may possess this capacity
by virtue of:
a. Its position, when the energy is called potential energy (PE).
b. Its motion, when the energy is called kinetic energy (KE).
14. Consider a mass m projected vertically upwards from the ground with initial velocity u. It is acted upon (downwards) by
2
2
gravity, and will attain a height h, which can be determined by substitution in the formula v = u + 2as, thus:

Page 258 : Wed Jan 02 02:42:30 2002

Mechanics

8- 2- 4- 3

DO NOT DISTRIBUTE

AP 3456

0 = u2 + 2(g)h
the velocity being zero at the highest point

h = u2 /2g
The work done reaching a height h is mgh. Substituting for h,
work done = mg.u 2 /2g

1
mu2 work done in
2

coming to rest from velocity u = KE.


15. At the highest point, where the velocity is zero, the kinetic energy is zero. If, however, the mass is allowed to fall, it will at
the moment of striking the ground have regained its original velocity, and thus its original kinetic energy. The energy which it
possessed at the highest point is potential energy and equals its original kinetic energy.
16. Consider also a body acted upon by a constant force F. Then F = ma

=m

v2 u 2
2s

where u, v and s have the conventional significance.


Thus
1
2
F.s = m(v u2 )
2
The left-hand side is the work done by the force, while the right-hand side is the resulting change in kinetic energy. The general
expression for the kinetic energy of a body moving with velocity v is:
1
KE = mv2
2
This illustrates the principal of the conservation of energy - energy can be neither created nor destroyed, though it may be
converted from one form to another. It should be noted that kinetic and potential energies are examples of one form of energy,
namely mechanical energy; a change may involve other forms of energy such as chemical, heat, light, sound, magnetic or
electrical energy.

Power
17. Power is the rate of doing work and has the unit, watt (W), equal to 1 joule per second.

CIRCULAR MOTION
Centripetal Force

Page 259 : Wed Jan 02 02:42:30 2002

Mechanics

8- 2- 4- 3

DO NOT DISTRIBUTE

AP 3456

2
18. It was shown in Chap 2 that a body travelling with uniform speed in a circle has an acceleration of v /r towards the centre
of the circle. The force producing this acceleration is termed centripetal force, and for a body of mass m the centripetal force is
2
mv /r towards the centre of the circle. It should be noted that in the case of a body travelling in a circular path at the end of a
string, while the mass is experiencing centripetal force towards the hand, there is an equal and opposite reaction on the hand
holding the string, known as centrifugal force. Centrifugal force exists only as an equal and opposite reaction to centripetal
force.

Moment of Inertia
19. Consider now a rigid body, such as a flywheel, free to rotate about an axis through its centre, at angular velocity w. When
the wheel is rotating all the particles of the wheel have the same angular velocity, but their linear velocities will depend on their
individual distances from the axis, those on the rim moving much faster than those near the axis.
20. A particle of mass m, distance r from the centre, and with linear velocity v, has kinetic energy 12 mv
1
2

or

mr2 ! 2 :The total kinetic energy of all the particles is 12 ! 2 . mr2 :

8-2-4-3 Fig 1 Simple Harmonic Motion

2
21. The sum of the products mr for all the particles in a rigid body rotating about a given axis is called the Moment of Inertia
(I).
I = mr2
22. Moment of inertia can be considered as the rotational equivalent of mass, and just as in linear motion, if a force is applied to
a body, force = mass x acceleration, so in circular motion if a torque is applied to a wheel,
torque = moment of inertia angular acceleration.

Page 260 : Wed Jan 02 02:42:30 2002

Mechanics

8- 2- 4- 3

DO NOT DISTRIBUTE

AP 3456

23. Radius of Gyration. The radius of gyration is a useful concept whereby the mass M (= m) of the wheel is considered to be
2
2
2
concentrated in a ring of radius k from the axis, such that mr = Mk . k is then known as the radius of gyration, and I = Mk .

SIMPLE HARMONIC MOTION


General
24. A type of motion which is frequently encountered is simple harmonic motion, which is defined as the motion of a body
which moves in a straight line so that its acceleration is directly proportional to the distance from a fixed point in the line, and
always directed towards that point. A mass suspended from a spiral spring which is given a small vertical displacement from its
equilibrium position and then released will oscillate with simple harmonic motion.
25. Simple harmonic motion can be illustrated by considering the projection of a point moving at uniform speed in a circular
path on to a diameter of that circle (Fig 1). Let the point P start at position A, and move with constant angular velocity w, in a
circle of radius r. Let PN be the perpendicular from the point to the diameter AOB. Then while the point P moves round the
circle from A through B back to A, N moves through O to B and back to A. The time for one complete cycle of oscillation is
2/ and this is called the periodic time or simply the period of the oscillation. The frequency of oscillation is the number of
complete cycles in unit time, and frequency = 1/periodic time = /2 The distance ON at any time, t, is given by x = r cos t.
The instantaneous velocity of N is given by:
dx
= r! sin !t, and the acceleration by:
dt

d2 x
= r! 2 cos !t
dt

= !2 x
Since the angular velocity w is constant, the acceleration is proportional to the distance ON, thus the movement of N accords
with the definition of simple harmonic motion. Note that the expression is negative, since the acceleration is measured in the
opposite direction to that of the displacement ON (ie towards O).

The Simple Pendulum


26. The simple pendulum provides an example of simple harmonic motion, provided that its amplitude of oscillation is small.
Consider a mass m suspended from a light inextensible cord of length l, displaced through a small angle (in radians) from the
vertical (Fig 2).
27. The forces acting on the mass are its weight mg acting vertically downwards, and the tension T in the cord. The weight can
be resolved into two components, mg cos equal and opposite to T, and the restoring force mg sin acting in the opposite
direction to the direction of displacement of the mass from its central position. Using the equation force = mass x acceleration:
mg sin = ma

a = g sin , or, for small angles

a = g (approximately)

Page 261 : Wed Jan 02 02:42:30 2002

Mechanics

8- 2- 4- 3

DO NOT DISTRIBUTE

AP 3456

But =

a=

x
1

gx
1

28. As g and l are constant for a particular location and a particular pendulum, the acceleration is proportional to the
displacement x, and acts in the direction of the force towards the mid point A, thus (for small displacements) satisfying the
conditions for simple harmonic motion.

8-2-4-3 Fig 2 The Simple Pendulum

29. By comparing the results of paras 25 and 27 it can be seen that the period of oscillation of a simple pendulum, for small
displacements, is given by:

period = 2

1
sec,
g

and the frequency by:

frequency =

1
2

g
cycles per second.
1

It will be noted that for small angles of swing, the period of oscillation of a simple pendulum is independent of the mass, and of
the amplitude of swing.

The Compound Pendulum


30. For a compound, or rigid, pendulum, the period is given by:

Page 262 : Wed Jan 02 02:42:30 2002

Mechanics

8- 2- 4- 3

DO NOT DISTRIBUTE

period = 2

I
mgh,

AP 3456

where I is the moment of inertia

about the axis of rotation, and h, is the distance from the centre of gravity to the pivot.

A Mass Supported by a Spring


31. The comparable equation for the period of oscillation of a mass m kilograms suspended from a spring of stiffness e
kilograms per metre is:

period = 2

m
sec.
e

Mechanics
Chapter 4 - Hydraulics
Introduction
1. Hydraulic systems provide a means of transmitting a force by the use of fluids. They are concerned with the generation,
modulation and control of pressure and flow of the fluid to provide a convenient means of transmitting power for the operation
of a wide range of aircraft services. A typical aircraft hydraulic system will be used for operating flying controls, flaps,
retractable undercarriages and wheelbrakes. Hydraulic systems can transmit high forces with rapid, accurate response to control
demands.

Definition of Terms
2.

The following definitions need to be understood:


a. Pressure. Pressure is the force per unit area exerted by a fluid on the surface of a container. Pressure is measured in
2
bars, Pascals (Pa), or Newtons per square metre (N/m ).
1 bar = 100,000 Pa = 100,000 N/m

b. Force. The force exerted on a particular surface by a pressure is calculated from the formula:
Force = Pressure Surface Area.
c. Fluid. A fluid is a liquid or gas which changes its shape to conform to the vessel that contains it.
d. Hydraulic Fluid. Hydraulic fluid is an incompressible oil. In aircraft systems, low flammability oils are used, the
boiling and freezing points of which fall outside operating parameters.

Transmission of Force and Motion by Fluids


3. When a force is applied to one end of a column of confined fluid a pressure is generated which is transmitted through the
column equally in every direction. Fig 1 illustrates a simple hydraulic system. Two cylinders are connected by a tube. The
2
2
cylinders contain pistons of surface areas 0.01m and 0.0001m giving a piston area ratio of 100:1. The system is filled with
hydraulic fluid and fitted with a pressure gauge. A force of 1000 Newtons (N) is applied to the small piston. The force will
Force (F)

produce a pressure (P) in the fluid Piston Area (A)


1000

ie P = 0.0001 = 10,000,000 Pa or 10 MPa, which is 100 bar.


The system pressure will appear on the gauge and will be felt on every surface within the system. Thus at the large piston a
6
force (F) will be exerted where F = P A = 10 10 0.01 = 100,000 N. The force applied at the small piston is therefore

Page 263 : Wed Jan 02 02:42:31 2002

Mechanics

8- 2- 4- 4

DO NOT DISTRIBUTE

AP 3456

increased on delivery by the large piston by a factor of 100, ie in the same ratio as that of piston area. This is sometimes referred
to as force multiplication.

8-2-4-4 Fig 1 A Simple Hydraulic System

8-2-4-4 Fig 2 Relative Movement in a Simple Hydraulic System

4. If the small piston is now moved down in its cylinder through a distance of 100 mm as illustrated in Fig 2 the large piston
will move upwards through a distance inversely proportional to the piston area ratio, ie through 1 mm. The work done by the
small force is transmitted hydraulically and equals the work expended in moving the greater force through a smaller distance, ie
force 1 distance 1 = force 2 distance 2.
or

1000 N 100 mm = 100,000 N 1mm

This is the principle of the hydraulic lever and is the operating principle of any hydraulic system.

Components of a Typical Hydraulic System


5. In a simple hydraulic system, force applied by hand or foot to a small master piston is transmitted to a larger slave
piston to operate the required service. In more complex, high performance systems, the master piston is replaced by hydraulic

Page 264 : Wed Jan 02 02:42:31 2002

Mechanics

8- 2- 4- 4

DO NOT DISTRIBUTE

AP 3456

pumps and the slave by actuators driving each of the powered services. A typical system illustrated in Fig 3 will operate at
between 200 and 300 bar, and it will include the following components:
a. Hydraulic Pump. The pump generates hydraulic pressure and delivers it to the pressure lines in the system. It will
usually be either engine driven or electrically powered.
b. Valves. Non-return valves control the direction of fluid flow, pressure relief valves the level of power produced, and
selector valves the amount of fluid flow to related actuators.
c. Actuators. Actuators convert the hydraulic power into usable mechanical power at the point required.
d. Hydraulic Fluid. The fluid provides the means of energy transmission as well as lubrication, and cooling of the system.
e. Connectors. The connectors link the various system components. They are usually rigid pipes, but flexible hoses are
also used.
f. Reservoir. Fluid is stored in a system reservoir in sufficient quantity and quality to satisfy system requirements. The
fluid becomes heated by operation of the system, and the reservoir performs the secondary functions of cooling the fluid and
of allowing any air absorbed in the fluid to separate out.
g. Filters. Hydraulic system components are readily damaged by solid particles carried in the fluid, and several stages of
filtration are included in a system to prevent debris passing from one component to the next. The filters perform as useful
tell-tales of fluid contamination. In addition, samples are taken periodically from the fluid and analysed to detect trends in
acid and other trace element levels.
h. Accumulator. An accumulator is a cylinder containing a floating piston. On one side of the piston is nitrogen at system
pressure, and on the other hydraulic fluid from the pressure line. When the hydraulic pressure is increased the nitrogen is
compressed. The compressed nitrogen then acts as a spring and can damp out system pressure ripples. It also acts as a
reserve of fluid and an emergency power source.

8-2-4-4 Fig 3 Typical Practical Hydraulic System

Mechanics
Chapter 5 - Introduction to Gyroscopes
INTRODUCTION
Description
1. A conventional gyroscope consists of a symmetrical rotor spinning rapidly about its axis and free to rotate about one or
more perpendicular axes. Freedom of movement about one axis is usually achieved by mounting the rotor in a gimbal, as in Fig
1 and complete freedom can be approached by using two gimbals, as illustrated in Fig 2.

Page 265 : Wed Jan 02 02:42:31 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

2. The physical laws which govern the behaviour of a gyroscope are identical to those which account for the behaviour of the
Earth itself. The two principal properties of a gyro are rigidity in inertial space (inertial space being a fixed spatial datum) and
precession. These properties are exploited in heading reference systems and inertial navigation systems, and some aircraft
instruments which are described in Vol 3.

Definition of Terms
3.

The following fundamental mechanical definitions provide the basis of the laws of gyrodynamics:
a. Momentum. Momentum is the product of mass and velocity (mv).

8-2-4-5 Fig 1 Gyroscope With One Degree of Freedom

b. Angular velocity. Angular velocity () is the instantaneous velocity (v) at the periphery of a circle of radius r, divided
by r (! = vr). Angular velocity is normally measured in radians per second.
c. Angular Momentum. If the mass (m) of a body is concentrated at a radius (r) from the axis of rotation, the angular
momentum is the product of the instantaneous linear momentum (mv) and the radius.
2
Angular momentum = mvr or mr
2
d. Moment of Inertia. The moment of inertia (I) of a body is the summation of mr for every particle of mass m located at
radius r from the centre of the mass, which goes to make up the mass of the body. In the case of a disc or wheel of uniform
mr 2

mass distribution throughout its radius r, the moment of inertia I about its axle is 2 . For a cylinder of outside radius r1
and inner radius r2:
2

I=

m(r1 + r22 )
2

(An alternative expression for Angular Momentum is the product of Moment of Inertia I, and Angular Velocity .)

8-2-4-5 Fig 2 Gyroscope with Two Degrees of Freedom

Page 266 : Wed Jan 02 02:42:32 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

Table 1 - Classification of Gyros


Type of Gyro
Rate Gyroscope

Rate Integrating
Gyroscope
Displacement
Gyroscope

Uses in Guidance and Control


Gyro Characteristics
1. Aircraft Instruments
1. Modified single degree of
freedom gyro.
2. Illustrates the principle of
secondary precession.
1. Inertial Navigation
1. Modified single degree of
freedom gyro.
2. Can also be a two degree of
freedom gyro.
1. Heading Reference
1. Two degrees of freedom.
2. Inertial Navigation System 2. Defines direction with respect
to space, thus it is also called a
space gyro, or free gyro.
3. Aircraft Instruments

e. Radius of Gyration. The radius of gyration of a body is that distance from the axis of rotation at which all the mass of
that body can be considered to act. It is normally denoted by k. Consequently, in calculations of moments of inertia of
bodies of irregular shapes, I = mk2, where m is the mass of the body.
f. Gyro Axes. In gyrodynamics it is convenient to refer to the axis about which the torque is applied as the input axis and
that axis about which the precession takes place as the output axis. The third axis, the spin axis, is self evident. The XX,
YY and ZZ axes shown in the diagrams are not intended to represent the x, y and z axes of an aircraft in manoeuvre.
However, if the XX (rotational) axis of the gyro is aligned with the direction of flight, the effects of flight manoeuvre on
the gyro may be readily demonstrated in similar fashion to the instrument descriptions in Vol 3.

Classification of Gyroscopes

Page 267 : Wed Jan 02 02:42:32 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

4.

AP 3456

Gyroscopes are classified in Table 1 in terms of the quantity they measure, namely:
a. Angular displacement from a known datum.
b. Rate of angular displacement of a vehicle.
c. The integral of an input with respect to time.

5. It should be realized however that the above classification is one of a number of ways in which gyroscopes can be classified.
Referring to Table 1 it will be seen that a displacement gyroscope could be classified as a two degree of freedom gyro or a space
gyro. Note also that the classification of Table 1 does not consider the spin axis of a gyroscope as a degree of freedom. In this
chapter a degree of freedom is defined as the ability to measure rotation about a chosen axis.

LAWS OF GYRODYNAMICS
Rigidity in Space
6. If the rotor of a perfect displacement gyroscope is spinning at constant angular velocity, and therefore constant angular
momentum, no matter how the frame is turned no torque is transmitted to the rotor axis. The law of conservation of angular
momentum states that the angular momentum of a body is unchanged unless a torque is applied to that body. It follows from this
that the angular momentum of the rotor must remain constant in magnitude and direction. This is simply another way of saying
that the rotor spin axis continues to point in the same direction in inertial space. This property of a gyro is defined in the First
Law of Gyrodynamics.

The First Law of Gyrodynamics


7. If a rotating body is so mounted as to be completely free to move about any axis through the centre of mass, then its spin
axis remains fixed in inertial space however much the frame may be displaced.
8. A space gyroscope loses its property of rigidity in space if the spin axis is subjected to random torques, some causes of
which will be examined later.

8-2-4-5 Fig 3 Precession

Precession

Page 268 : Wed Jan 02 02:42:32 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

1
Consider the free gyroscope in Fig 3, spinning with constant angular momentum about the XX axis. If a small mass M is
1
placed on the inner gimbal ring it exerts a downward force F so producing a torque T about the YY axis. By the laws of
1
rotating bodies this torque should produce an angular acceleration about the YY axis, but this is not the case:
9.

a. Initially the gyro spin axis will tilt through a small angle ( in Fig 3), after which no further movement takes place
1
about the YY axis. The angle is proportional to T and is a measure of the work done. Its value is almost negligible and
will not be discussed further.
1
1
b. The spin axis then commences to turn at a constant angular velocity about the axis perpendicular to both XX and YY ,
1
1
ie the ZZ axis. This motion about the ZZ axis is known as precession, and is the subject of the Second Law of
Gyrodynamics.

The Second Law of Gyrodynamics


10. If a constant torque T is applied about an axis perpendicular to the spin axis of an unconstrained, symmetrical spinning
body, then the spin axis will precess steadily about an axis mutually perpendicular to the spin axis and the torque axis. The
angular velocity of precession is given by:

8-2-4-5 Fig 4 Determining Precession

T
I!

11. Precession ceases as soon as the torque is withdrawn, but if the torque application is continued, precession will continue
until the direction of spin is the same as the direction of the applied torque. If, however, the direction of the torque applied about
the inner gimbal axis moves as the rotor precesses, the direction of spin will never coincide with the direction of the applied
torque.

Direction of Precession
12. Fig 4 shows a simple rule of thumb to determine the direction of precession:

Page 269 : Wed Jan 02 02:42:32 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

a. Consider the torque as being due to a force acting at right angles to the plane of spin at a point on the rotor rim.
b. Carry this force around the rim through 90 in the direction of rotor spin.
c. The torque will apparently act through this point and the rotor will precess in the direction shown.

CONSERVATION OF ANGULAR MOMENTUM


Explanation
13. In linear motion, if the mass is constant, changes in momentum caused by external forces will be indicated by changes in
velocity. Similarly in rotary motion if the moment of inertia is constant, then the action of an external torque will be to change
the angular velocity in speed or direction and in this way change the angular momentum. If, however, internal forces (as distinct
from external torques) act to change the moment of inertia of a rotating system, then the angular momentum is unaffected.
Angular momentum is the product of the moment of inertia and angular velocity, and if one is decreased so the other must
increase to conserve angular momentum. This is the Principle of Conservation of Angular Momentum.
14. Consider the ice-skater starting her pirouette with arms extended. If she now retracts her arms she will be transferring mass
closer to the axis of the pirouette, so reducing the radius of gyration. If the angular momentum is to be maintained then, because
of the reduction of moment of inertia, the rate of her pirouette must increase, therefore:
a. If the radius of gyration of a rotating body is increased a force is considered to act in opposition to the rotation caused by
the torque, decreasing the angular velocity.
b. If the radius of gyration is decreased a force is considered to act assisting the original rotation caused by the torque, so
increasing the angular velocity.

Cause of Precession
1
1
1
15. Consider the gyroscope rotor in Fig 5a spinning about the XX axis and free to move about the YY and ZZ axes. Let the
quadrants (1, 2, 3 and 4) represent the position of the rotor in spin at one instant during the application of an external force to the
1
1
spin axis, producing a torque about the YY axis. This torque is tending to produce a rotation about the YY axis while at the
1
same instant the rotor spin is causing particles in quadrants 1 and 3 to recede from the YY axis, increasing their moment of
1
inertia about this axis, and particles in quadrants 2 and 4 to approach the YY axis decreasing their moment of inertia about this
1
axis. Particles in quadrants 1, 2, 3 and 4 tend to conserve angular momentum about YY , therefore:
1
a. Particles in quadrants 1 and 3 exert forces opposing their movement about YY .
1
b. Particles in quadrants 2 and 4 exert forces assisting their movement about YY .

8-2-4-5 Fig 5 Instant of Spin and Precession

Page 270 : Wed Jan 02 02:42:32 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

16. Hence 1 and 4 exert forces on the rotor downwards, whilst 2 and 3 exert forces upwards. These forces can be seen to form
1
a couple about ZZ , (Fig 5b), causing the rotor to precess in the direction shown in Fig 5c.

Gyroscopic Resistance
17. In demonstrating precession it was stated that after a small deflection about the torque axis movement about this axis
ceased, despite the continued application of the external torque. This state of equilibrium means that the sum of all torques
acting about this axis is zero. There must, therefore, be a resultant torque L acting about this axis which is equal and opposite to
the external torque, as shown in Fig 6. This resistance is known as Gyroscopic Resistance and is created by internal couples in a
precessing gyroscope.
1
18. Formation of Gyroscopic Resistance. Consider now the gyroscope in Fig 5c spinning about an axis XX and precessing
1
1
about the ZZ axis under the influence of a torque T, about the YY axis. The rotor quadrants represent an instant during the
1
precession and spin. Using the argument of para 15, the particles in quadrants 1 and 3 are approaching the ZZ axis and
1
exerting forces acting in the direction of precession, while in quadrants 2 and 4 the particles are receding from the ZZ axis and
1
exerting forces in opposition to the precession. The resultant couple is therefore acting about the YY axis in opposition to the
external torque. This couple is the Gyroscopic Resistance. It has a value equal to the external torque thus preventing movement
1
about the YY axis.
19. Caging. Gyroscopic Resistance is always accompanied by precession, and it is of interest to note that if precession is
prevented, gyroscopic torque cannot form and it is as easy to move the spin axis when it is spinning as when it is at rest. This
1
can be demonstrated by applying a torque to the inner gimbal of a gyroscope with one degree of freedom. With the ZZ axis
locked the slightest touch on the inner gimbal will set the gimbal ring (and the rotor) moving. This behaviour is exploited in
caging devices.

Secondary Precession
20. If a sudden torque is applied about one of the degrees of freedom of a perfect displacement gyroscope the following
phenomena should be observed:
a. Nodding or nutation occurs. Here it is sufficient to note that nutation occurs only for a limited period of time and
eventually will cease completely. Additionally nutation can only occur with a two degree of freedom gyro and to a large

Page 271 : Wed Jan 02 02:42:32 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

extent it can be damped out by gyro manufacturers.


b. A deflection takes place about the torque axis, (dip), which remains constant provided that the gyro is perfect and the
applied torque is also constant.
1
c. The gyro precesses, or rotates, about the ZZ axis.
21. If, however, an attempt is made to demonstrate this behaviour, it will be seen that the angle of dip will increase with time,
apparently contradicting sub-para 20b..
1
22. To explain this discrepancy, consider Fig 7. If the gyro is precessing about the ZZ axis, some resistance to this precession
must take place due to the friction of the outer gimbal bearings. If this torque T is resolved using the rule of thumb given in para
12, it will be seen that the torque T causes the spin axis to dip through a larger angle. This precession is known as secondary
precession.

8-2-4-5 Fig 6 Gyroscopic Resistance

8-2-4-5 Fig 7 Precession Opposed by Secondary Precession

Page 272 : Wed Jan 02 02:42:33 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

23. Secondary precession can only take place when the gyro is already precessing, thus its name. Note also that secondary
precession acts in the same direction as the originally applied torque.

THE RATE GYROSCOPE


Principle of Operation
1
1
24. Fig 8 shows a gyroscope with freedom about one axis YY . If the frame of the gyro is turned about an axis ZZ at right
1
1
1
angles to both YY and XX , then the spin axis will precess about the YY axis. The precession will continue until the
1
direction of rotor spin is coincident with the direction of the turning about ZZ .
25. Suppose the freedom of this gyroscope about the gimbal axis is restrained by the springs connecting the gimbal ring to the
1
1
frame as in Fig 9. If the gyroscope is now turned about the ZZ axis, precession about the YY axis is immediately opposed by
a torque applied by the springs. It has been shown that any torque opposing precession produces a secondary precession in the
same direction as the original torque (see para 23). If the turning of the frame is continued at a steady rate, the precession angle
1
about the YY axis will persist, distending one spring and compressing the other, thereby increasing the spring torque.
1
Eventually the spring torque will reach a value where it is producing secondary precession about ZZ equal to, and in the same
direction as, the original turning. When this state is reached the gyroscope will be precessing at the same rate as it is being
1
turned and no further torque will be applied by the turning. Any change in the rate of turning about the ZZ axis will require a
different spring torque to produce equilibrium, thus the deflection of the spin axis ( in Fig 9) is a measure of the rate of
turning. Such an arrangement is known as a Rate Gyroscope, and its function is to measure a rate of turn, as in the Rate of Turn
Indicator.

8-2-4-5 Fig 8 Gyro with One Degree of Freedom - Precession

Page 273 : Wed Jan 02 02:42:33 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

26. The relationship between the deflection angle and rate of turn is derived as follows:
Spring Torque is proportional to or
Spring Torque = K (where K is a constant)
At equilibrium:
Rate of Secondary Precession = Rate of Turn
ie

K
I!

= Rate of Turn

is proportional to Rate of Turn I


(I is the angular momentum of the rotor and is therefore constant).
The angle of deflection can be measured by an arrangement shown at Fig 10 and the scale calibrated accordingly.

8-2-4-5 Fig 9 Rate Gyroscope

Page 274 : Wed Jan 02 02:42:33 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

THE INTEGRATING GYROSCOPE


Principle of Operation
27. An integrating gyroscope is a single degree of freedom gyro using viscous restraint to damp the precessional rotation about
the output axis. The integrating gyro is similar to the rate gyro except that the restraining springs are omitted and the only factor
opposing gimbal rotation about the output axis is the viscosity of the fluid. Its main function is to detect turning about the input
1
1
axis (YY in Fig 11), by precessing about its output axis (ZZ in Fig 11).
28. The integrating gyro was designed for use on inertial navigation stable platforms, where the requirement was for immediate
and accurate detection of movement about three mutually perpendicular axes. Three integrating gyros are used, each performing
its functions about one of the required axes. These functions could be carried out by displacement gyros, but the integrating
gyro has certain advantages over the displacement type. These are:
a. A small input rate causes a large gimbal deflection (gimbal gain).
b. The gyro does not suffer from nutation.
29. Fig 11 shows a simple integrating gyro. It is basically a can within which another can (the inner gimbal) is pivoted about
1
its vertical (ZZ ) axis. The outer can (frame) is filled with a viscous fluid which supports the weight of the inner gimbal so
1
reducing bearing torques. The rotor is supported with its spin (XX ) axis across the inner gimbal. In a conventional non-floated
gyro, ball bearings support the entire gimbal weight and define the output axis. In the floated integrating gyro the entire weight
of the rotor and inner gimbal assembly is supported by the viscous liquid, thereby minimizing frictional forces at the output
1
(ZZ ) axis pivot points. The gimbal output must however be defined and this is done by means of a pivot and jewel
arrangement. By utilizing this system for gimbal axis alignment, with fluid to provide support, the bearing friction is reduced to
a very low figure.

8-2-4-5 Fig 10 Rate of Turn Indicator

Page 275 : Wed Jan 02 02:42:33 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

30. The gyroscope action may now be considered. If the whole gyro in Fig 12 is turned at a steady rate about the input axis
1
1
(YY ), a torque is applied to the spin axis causing precession about the output axis (ZZ ). The gimbal initially accelerates
(precesses) to a turning rate such that the viscous restraint equals the applied torque. The gimbal then rotates at a steady rate
1
about ZZ proportional to the applied torque. The gyro output (an angle or voltage) is the summation of the amount of input
turn derived from the rate and duration of turn and is therefore the integral of the rate input. (Note that the rate gyro discussed in
paras 24 - 26 puts out a rate of turn only). The movement about the output axis may be made equal to, less than, or greater than
movements about the input axis by varying the viscosity of the damping fluid. By design the ratio between the output angle ()
and the input angle () can be arranged to be of the order of 10 to 1. This increase in sensitivity is called gimbal gain.

8-2-4-5 Fig 11 Simple Integrating Gyroscope

Page 276 : Wed Jan 02 02:42:33 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

8-2-4-5 Fig 12 Function of Integrating Gyroscope

Page 277 : Wed Jan 02 02:42:33 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

31. A gyro mounted so that it senses rotations about a horizontal input axis is known as a levelling gyro. Two levelling gyros
are required to define a level plane. Most inertial platforms using conventional gyros align the input axis of their levelling gyros
with True North and East.
32. Motion around the third axis, the vertical axis, is measured by an azimuth gyro, ie one in which the input axis is aligned
with the vertical, as in Fig 13.

8-2-4-5 Fig 13 Azimuth Gyroscope

THE DISPLACEMENT GYROSCOPE


Definition

Page 278 : Wed Jan 02 02:42:33 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

33. A displacement gyro is a two degree of freedom gyro. It can be modified for a particular task, but it always provides a
fixed artificial datum about which angular displacement is measured.

Wander
34. Wander is defined as any movement of the spin axis away from the reference frame in which it is set.
35. Causes of Wander. Movement away from the required datum can be caused in two ways:
a. Imperfections in the gyro can cause the spin axis to move physically. These imperfections include such things as
friction and unbalance. This type of wander is referred to as real wander since the spin axis is actually moving. Real
wander is minimized by better engineering techniques.
b. A gyro defines direction with respect to inertial space, whilst the navigator requires earth directions. In order to use a
gyro to determine directions on earth, it must be corrected for apparent wander due to the fact that the earth rotates or that
the gyro may be moving from one point on earth to another (transport wander).
36. Drift and Topple. It is more convenient to study wander by resolving it into two components:
a. Drift, which is defined as any movement of the spin axis in the horizontal plane around the vertical axis.
b. Topple, which is defined as any movement of the spin axis in the vertical plane around a horizontal axis.
37. Summary. Table 2 summarizes the types of wander. From para 35 it should be apparent that the main concern when using
a gyro must be to understand the effects of earth rotation and transport wander on a gyro.

Table 2 - Types of Wander

Earth Rotation
38. In order to explain the effects of earth rotation on a gyro it is easier to consider a single degree of freedom gyro, since it has
only one input and one output axis. The following explanation is based on a knowledge of rotational vector notation.
39. Consider a gyro positioned at a point A in Fig 14. It would be affected by earth rotation according to how its input axis was
aligned, namely:
a. If its input axis was aligned with the earth's spin axis, it would detect earth rate e 15.04 degrees per hour.

8-2-4-5 Fig 14 Components of Earth Rate

Page 279 : Wed Jan 02 02:42:33 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

b. Azimuth Gyro. If its input axis was aligned with the local vertical it would detect 15.04 x sin Latitude (sin )/hr. Note
that by definition this is drift.
c. North Sensitive Levelling Gyro. If its input axis were aligned with local North, it would detect 15.04 x cos Latitude (cos
) /hr. Note that by definition this is topple.
d. East Sensitive Levelling Gyro. Finally if the input axis were aligned with local East, that is, at right angles to the earth
rotation vector, it would not detect any component of earth rotation.

Transport Wander
40. If an azimuth gyro spin axis is aligned with local North (ie the true meridian) at A in Fig 15 and the gyro is then transported
to B, convergence of the meridians will make it appear that the gyro spin axis has drifted. This apparent drift is in addition to
that caused by Earth rotation. The gyro has not in fact drifted; it is the direction of the True North which has changed.
However, if the gyro is transported North-South, there is no change in the local meridian and therefore, no apparent drift.
Similarly, as all meridians are parallel at the Equator, an East-West movement there produces no apparent drift. Transport rate
drift thus depends on the convergence of the meridians and the rate of crossing them; ie the East-West component of ground
speed (U). The amount of convergence between two meridians is ch long x sin lat. Any given value of U thus produces an
increase in apparent gyro drift as latitude increases. The amount of drift due to transport rate may be found as follows:
Meridian conv. C (/hr) = [ch long/hr] sin .
(nm/hr)
sec and
Now ch long/hr = ch Eastings
60

since 1 = 60 nm and ch Eastings (nm/hr) = U

C=

Page 280 : Wed Jan 02 02:42:34 2002

U
sec sin ( /hr)
60

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

But sec sin =

1
sin
cos

sin
tan
cos

8-2-4-5 Fig 15 Apparent Drift

::_ C =

U
tan ( /hr)
60

This can be converted to radians/hour by multiplying

by 180
ie ::_ C =

60
180

= U tan

60 180

Now an arc of length 60 nm on the earth's surface subtends an angle of 1 ( / 180) at the centre of the earth.
::_ R

= 60 where R = earth's radius


180

or

1
=
R 60 180

Page 281 : Wed Jan 02 02:42:34 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

Substituting into the above equation for Meridian Convergence (radians/hour)

C = U tan

1
R

or
C=

U
(radians/hour)
R

41. Consider now two levelling gyros, whose input axes are North and East respectively, and whose output axes are vertical.
U

a. The East component of aircraft velocity in Fig 16 will be sensed by the North gyro as a torque of R about its input axis.
If the gyro is not corrected for this transport wander, it is said, by definition, to topple.
V

b. Similarly, due to the effect of aircraft velocity North, the East gyro will topple at the rate of R .

8-2-4-5 Fig 16 Transport Wander

Apparent Wander Table


42. All the equations derived in the study of earth rate and transport wander rate are summarized in Table 3.

Table 3 - Components of Drift and Topple - Earth Rate and Transport Wander Rate
Local North
Earth Rate degrees or
radians per hour
Transport Wander
radians per hour

e cos

Local East
Nil

U
R

V
R

Local Vertical Correction Sign


+
e sin
U
R

Topple
e

= Angular Velocity of the Earth

Page 282 : Wed Jan 02 02:42:34 2002

+E
W

tan

Drift
R

= Earth's radius

Mechanics

Latitude

8- 2- 4- 5

DO NOT DISTRIBUTE

= East/West component of
groundspeed

AP 3456

= North/South component of
groundspeed

The units for earth rate can be degrees or radians, whilst for transport wander they are radians.
43. Correction Signs. The correction signs of Table 3 apply only to the drift equations, and they should be applied to the
gyro readings to obtain true directions. These correction signs will be reversed for the Southern Hemisphere.

Practical Corrections for Topple and Drift


44. If all the corrections of Table 3 were applied to three gyros with their input axes aligned to true North, true East and the
local vertical, true directions would be defined continuously, and in effect the gyros would have been corrected for all apparent
wander. However, these corrections make no allowance for the real wander of a gyro and consequently an error growth
proportional to the magnitude of the real drift and topple will exist. As a rough rule of thumb, an inertial platform employing
gyros with real drift rates in the order of 0.01/hr will have a system error growth of 1-2 nm/hr CEP.
45. Twin gyro platforms (TGPs), on the other hand, employ cheaper lower quality gyros whose drift rates may be in the order
of 0.1/hr. If these real drift rates were not compensated for, system inaccuracies would be unacceptably large. For this reason,
most TGP systems make use of the local gravity vector to define the level plane, thus compensating for both real and apparent
drifts.
46. Specifically, gyro wander may be corrected in the following ways:
a. Topple. Topple is normally corrected for in gyros by the use of either gravity switches (see Figs 17a, b and c), or by case
levelling devices (see Figs 18a, b and c). These devices sense movement away from the vertical, and send appropriate
signals to a torque motor until the vertical is re-established. The levelling accuracy of these methods is a approximately 1.

8-2-4-5 Fig 17a Gravity Sensitive Switch

8-2-4-5 Fig 17b Gravity Levelling

Page 283 : Wed Jan 02 02:42:34 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

8-2-4-5 Fig 17c Schematic Electrical Erection System

b. Drift. Drift corrections can be achieved by:


(1) Calculating corrections using Table 3 and applying them to the gyro reading.
(2) Applying a fixed torque to the gyro so that it precesses at a rate equal to the earth rate for a particular latitude. Drift
will be fully corrected for only at the selected latitude.
(3) Applying variable torques, using the same approach as in (2) above, but being able to vary the torque according to
the latitude. These azimuth drift corrections make no allowance for real drift, which can only be limited by coupling the
azimuth gyro to a flux valve.
47. To complete this study of the displacement gyro, it remains to mention a limitation and an error peculiar to this type of
gyro, gimbal lock and gimbal error.

Page 284 : Wed Jan 02 02:42:34 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

8-2-4-5 Fig 18a Electrical Case Levelling

Gimbal Lock
48. Gimbal lock occurs when the gimbal orientation is such that the spin axis becomes coincident with an axis of freedom.
Effectively the gyro has lost one of its degrees of freedom, and any attempted movement about the lost axis will result in real
wander. This is often referred to as toppling, although drift is also present.

8-2-4-5 Fig 18b Commutator Sensing Device

Page 285 : Wed Jan 02 02:42:34 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

8-2-4-5 Fig 18c Syncro Sensing Device

Page 286 : Wed Jan 02 02:42:35 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

Gimbal Error
49. Gimbal error occurs when there is a misalignment between the aircrafts aerodynamic axes and the navigation system axes.
Most modern systems automatically compensate for these errors.

THE RING LASER GYRO


Introduction
50. The ring laser gyro is now a serious competitor to conventional gyros for a number of applications including aircraft inertial
navigation systems and attitude - heading reference systems. The ring laser gyro has no moving parts and is therefore
particularly suitable for strap down inertial navigation systems.

Principle of Operation
51. The laser gyro makes use of the high sensitivity of the laser's oscillating frequency to variations of the dimensions of its
resonant structure. The resonant structure consists of an optical system of three or more mirrors in which a light wave can travel
continuously along a closed path in both clockwise and anti-clockwise directions. The closed path is filled with ionized
helium-neon gas which provides gain over a very narrow bandwidth. The assembly can oscillate at any frequency within this
bandwidth. By careful control of both gain and the circumference of the closed path a condition can be achieved where the gain
is only high enough to sustain oscillation at one resonant frequency.
52. If the ring laser is stationary the resonant frequency for both clockwise and anti-clockwise waves are identical. If the ring
laser is rotated about an axis perpendicular to the plane of the closed path, the resonant frequencies of the clockwise and
anti-clockwise waves are different. This is because the light travelling in the direction of the rotation must travel a slightly
longer path to complete one revolution, while the opposite wave will travel a shorter path. Thus the frequencies of oscillation are
determined by the rate of rotation of the assembly.

8-2-4-5 Fig 19 Ring Laser Gyro

Page 287 : Wed Jan 02 02:42:35 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

The frequency difference f of the two waves is:

f =

4A
L

Where
A is the area enclosed by the path
is the oscillating wavelength
L is the length of the closed path
is the rate of rotation
If the value of f is measured by the mixing of an output signal for each wave on a photodetector, a signal is obtained which is
proportional to rotation rate.

Advantages of the Ring Laser Gyro


53. The advantages of the ring laser gyro are:
a. Performance is unaffected by high "g".
b. It has no moving parts and therefore high reliability and low maintenance requirements.
c. It has a rapid turn-on time.

8-2-4-5 Fig 5a

Page 288 : Wed Jan 02 02:42:35 2002

Mechanics

8- 2- 4- 5

DO NOT DISTRIBUTE

AP 3456

8-2-4-5 Fig 5b

8-2-4-5 Fig 5c

Weapons Applications
Chapter 1 - Conventional Damage Mechanisms
Introduction
1. Weapons may be classified in a number of ways; this chapter sets out to classify conventional (ie neither nuclear, chemical
nor biological) weapons by the manner in which damage is caused to the target. In this way it will be seen that certain weapons
may be more suitable for attack against certain targets than others. Inevitably, as with most classification systems, there will
occasionally be some degree of overlap and many practical weapons will exhibit more than one mechanism.
2.

Four main categories of conventional weapon damage mechanisms may be recognized as follows:
a. Penetration. Penetration weapons may be either solid or may contain a small quantity of high explosive. Solid
projectiles are designed to cause damage to the target by virtue of their kinetic energy at impact. Those with an explosive
either use the explosive to increase the damage after penetration - or in the form of a shaped charge to effect penetration.
Clearly with this type of weapon a near miss causes no damage at all.
b. Blast. Blast weapons have a relatively large content of high explosive and are designed to damage the target by the
effects of the shock wave. Although a hit will normally cause the maximum damage, a near miss will often be acceptable
depending upon the nature of the target and the miss distance.
c. Fragmentation. Fragmentation weapons are designed to damage the target by subjecting it to a dense pattern of high
velocity fragments. Normally fragmentation is used jointly with blast in one weapon. Clearly there is some overlap in
effect with penetration weapons.
d. Incendiary. Incendiary weapons are normally in the form of bombs and are designed to start a major fire in the target
using the target materials as fuel.

Penetration Weapons
3. Kinetic Energy Projectiles. Kinetic energy projectiles rely on their high kinetic energy on impact to achieve penetration.

Page 289 : Wed Jan 02 02:42:35 2002

Weapons Applications

8- 2- 5- 1

DO NOT DISTRIBUTE

AP 3456

Some may have a small high explosive filling fused to detonate after penetration to enhance the damage caused inside the target.
Perhaps the most familiar KE weapon is the ball ammunition used with small arms and some aircraft cannons. As KE is
determined by the equation 12 mv2 , where m - mass and v = velocity, it is clearly advantageous to have high values of both,
2
although v is usually the dominant factor. Accordingly, sharply pointed penetration rounds are frequently manufactured from
hard, high density, tungsten based alloys or depleted uranium. Similarly this type of warhead is most commonly associated with
forward firing weapons so as to achieve high impact velocities. The current trend is for the projectiles to be sub-calibre and
fitted with a disposable sabot to fit the bore of the gun or rocket tube.

8-2-5-1 Fig 1 Shaped Charge

4. Shaped Charge Warhead. The shaped or hollow charge effect was discovered in 1888 by an American named Munroe. He
found that by forming shapes in the surface of a piece of explosive and then detonating it with the shape's surface in contact with
a metal sheet the shape was reproduced on the sheet. Later experiments showed that the deeper the shaping the greater the
depression in the metal sheet. In 1939 experiments with lined cavities showed a new phenomenon. Conical cavities lined with a
thin layer of metal collapsed following detonation and formed a high speed jet of liquid metal. The process is illustrated in Fig
1
1. Jet tip speeds of about 7,000 ms were achieved whilst at the rear of the jet where a slug is formed speeds were significantly
1
lower, approximately 2,000 ms . This jet has a high penetrative capability against armour and has been used for many years in
anti-tank warhead design. Factors which govern the depth of penetration are:
a. The type and mass of explosive charge.
b. The shape and dimensions of the cavity.
c. The type and thickness of the liner.
d. The stand-off distance.
The deepest penetration is achieved when the impact is normal to the surface attacked, and in order to achieve maximum
penetration a stand-off distance is required to enable the jet to build up to its full velocity. This stand-off distance depends upon
the target and the charge design; for a 45 cone angle between 3 and 6 cone diameters of stand-off would generate the optimum
jet penetration. Rotation of the missile as in spin stabilized weapons diminishes the effectiveness of the shaped charge. In
general, cones with small apex angles give greater penetration but smaller hole diameters, whereas lower aspect ratio cones give
larger holes but with reduced penetration. A bonus gained with this damage mechanism is the vaporific effect. Jet particles
penetrating tanks and aircraft fuselages give up energy and heat gases within such compartments and this heating can be enough
to cause an internal explosion. If the liner is made of aluminium and if oxygen is present the burning of the aluminium increases
the vaporific effect.

Page 290 : Wed Jan 02 02:42:35 2002

Weapons Applications

8- 2- 5- 1

DO NOT DISTRIBUTE

AP 3456

5. Explosively Formed Projectile (EFP) Warhead. A low aspect ratio development of the conical lined shaped charge was
produced during the Second World War by Miznay and Schardin. The result of increasing the cone angle is that the jet
1
production is minimized and a single large slug fragment tends to be produced with a velocity of about 2000 ms . Although it
is less penetrative than a jet, the slug produces a very large diameter hole in the target. By careful design of the liner this single
slug can take up an aerodynamic shape and achieve its maximum penetration at a distance of over 500 cone diameters range. It
is possible to design a number of Miznay-Schardin segments into a single warhead, initiated by one charge. Thus a number of
large self-forging sub-projectiles can be generated. As the sub-projectiles are in themselves very directional it is possible for the
warhead to be designed with high lethality in only one particular plane or sector.

8-2-5-1 Fig 2 Pressure Variations in the Vicinity of an Explosion

Blast Weapons
6. General Description. The outrush of high pressure gases following the detonation of a high explosive warhead, violently
displaces the surrounding layers of air which in turn displace further layers of air so that the disturbance spreads rapidly
outwards far beyond the region directly affected by the explosive gases. After the flash of the explosion there is a brief interval
before the shock front of the disturbed air arrives and causes a large and sudden rise in pressure. For a short time the pressure
remains above atmospheric and this constitutes the positive phase of the disturbance. The positive phase is followed by a
somewhat longer negative phase during which the pressure is below ambient and thereafter minor pressure oscillations may
continue for some time (see Fig 2). Damage may be caused primarily either as a result of the pressure difference across the
target (diffraction loading) as the blast wave passes, or by the drag resistance to the air moving rapidly over the target. Drag
sensitive targets are usually considered soft while diffraction sensitive targets are regarded as hard. Clearly a weapon
delivering a high peak over-pressure will be better suited to a hard target, while a long period of over-pressure, even if at a lower
level, will be better matched to soft targets. The term blast is commonly applied both to the expanding explosive gases and to
the pressure waves set up in the surrounding atmosphere. Blast damage can be closely related to the positive impulse. Since
blast effect falls off rapidly with distance from the point of detonation, blast warheads must be placed in very close proximity to
the target or the disruptive effect will be low. The blast warhead is most effective against underwater targets because of the
greater density of water compared with air. For a given warhead size the impulse of the shock wave in water can be as much as
30 times that in air. Blast effect also falls off with altitude such that at 50,000 feet approximately 50% more explosive would be
required to generate the same over-pressure as that achieved at sea level. Generally the blast effect is not employed in isolation
but is associated with fragmentation effects (exceptions are torpedoes and fuel/air explosives (FAE)).
7. Fuel/Air Explosive (FAE). Fuel/air explosive comprises an aerosol cloud of combustible fuel which is ignited to produce a
blast over a significant area rather than at a point. Since only the fuel, and not the oxidizing agent, needs to be carried in the
weapon, it is relatively weight efficient. FAE weapons tend to have relatively long impulses and relatively low over-pressures
and so are generally best suited to soft targets such as aircraft, unreinforced buildings, missiles, unarmoured vehicles and
personnel. They have also been used with some success for mine clearance.

Page 291 : Wed Jan 02 02:42:36 2002

Weapons Applications

8- 2- 5- 1

DO NOT DISTRIBUTE

AP 3456

Fragmentation Weapons
8. General Description. This term is used to refer to the breaking down of the walls of a projectile caused by the detonation of
high explosive contained within it. Much of the detonation energy is transferred to the fragments as kinetic energy. If the
explosion occurs inside a smooth homogenous case the resulting fragments are of irregular shape and variable size, (natural
fragmentation). They are projected outwards with an initial velocity of several thousand feet per second, which owing to their
irregular shape is poorly sustained, particularly by the smaller fragments. The size and velocity of the fragments is determined
by the strength of the projectile walls, by the charge/weight ratio and by the nature of the explosive filling. Greater efficiency
can be achieved by predetermining in some way the nature of the fragmentation rather than relying on natural fragmentation.
9. Fragment Production. Several methods of fragment production have been developed:
a. Pre-formed Fragment (PFF) Warhead. Pre-formed metal fragments are secured to the outside of the thin metal casing
which contains the explosive. Alternatively the fragments may be embedded in the explosive filling. Such fragments are
normally of mild steel or tungsten and are produced in various weights and shapes.
b. Explosively Formed Fragmentation (EFF) Warhead. The warhead casing is notched, grooved or scored in some
predetermined pattern which ensures break up in a manner giving predictable fragment size. Alternatively a simple and
economical way of producing fragments of a known size is to line the warhead casing with a thin notched membrane of
metal or plastic, thereby controlling the distribution of the initial shock wave onto the warhead casing.
c. Continuous Rod Casing. Steel rods or short lengths of wire are assembled into the form of a hollow cylinder which is
packed with explosive. By welding the rod ends alternately the entire casing expands into a large bracelet hoop when the
warhead explodes. This is known as a continuous rod warhead and produces a cutting effect. Although it is a most
effective damage mechanism against aircraft targets it has the disadvantage of needing very accurate initiation with respect
to target position since the damage mechanism is strictly directional in the rolling plane of the warhead. Its effect is thus
critically dependent upon accurate fusing.
With a lateral belt distribution of fragments a large proportion of the energy will normally be directed away from the target thus
reducing the overall efficiency of the weapon. Much design effort has therefore now turned towards concentrating the fragments
into annular or directional patterns thus increasing the fragment density.

Incendiary Weapons
10. General Description. Incendiary weapons differ from other conventional weapons in that they are devices which utilize the
chemical energy of the target, ie the target is destroyed by the burning of its own combustible parts. Such weapons are
economical damage mechanisms because most of the energy required for destruction comes from the target itself and does not
have to be stored in the weapon. They are however generally ineffective against hard targets. The incendiary filling requires to
be easily ignitable, to burn at a high temperature and to continue burning long enough to give a good chance of starting a fire.
Some fillings, for example those containing phosphorus, are spontaneously inflammable so that no problem of ignition arises.
Most incendiary weapons, however, are manufactured from a hydrocarbon fuel suitably brought to the required viscosity by
chemical thickeners such as naphtha or aluminium soap (NAPALM).

Weapons Applications
Chapter 2 - Guidance
The Need for Guidance
1. Introduction. Installing a guidance system into a weapon is expensive - an expense which is compounded by the ancillary
ground or airborne equipment which is necessary for its control, and the possible need to include a propulsion system. It is
important therefore that the decision to include guidance produces a cost-effective result which overcomes the errors and
shortcomings of unguided weapons.
2. Shortcomings of Unguided Weapons. The shortcomings of unguided weapons may be examined under the headings of
range and accuracy. As an example the free-fall bomb will be considered.
a. Limited Range. The range to which a free-fall bomb can be delivered depends mainly upon the release conditions, ie
speed, height, attitude. The maximum range of air launched bombs is about 5 nm, which is typical both of a toss or loft
delivery from low level and of the forward throw of a high level release.

Page 292 : Wed Jan 02 02:42:36 2002

Weapons Applications

8- 2- 5- 2

DO NOT DISTRIBUTE

AP 3456

b. Reduced Accuracy. An unguided bomb will incur delivery errors at launch and during its trajectory. These errors arise
from inaccurate knowledge at release of parameters such as aircraft speed, heading, angle of attack, sideslip, g forces,
altitude, range to target, and release disturbance. After release the weapon will experience ballistic dispersion and unknown
wind effects, and the target may also be moving - possibly taking evasive action. Although many of these potential
inaccuracies can be minimized by the weapon aiming computer, there will always be a degree of residual inaccuracy due to
imperfect data and prediction. The effect tends to be magnified as the range from release to target is increased.
3. Advantages of Guided Weapons. A guidance system will supply corrections to the weapons flight right up to the point of
impact. The advantage is most apparent when the target is such that a near miss will have little effect. Perhaps one of the best
examples was seen in the Vietnam war when laser guided bombs were introduced. One task was to destroy bridges, but many
hundreds of sorties with unguided bombs often failed to inflict any damage whereas a single operation with guided bombs was
usually sufficient to bring down a span. Thus fewer sorties had to be flown to achieve the desired result and there were
consequently fewer aircraft losses. Although the guided weapons were considerably more expensive than the unguided versions,
they were cost-effective overall.

8-2-5-2 Fig 1 Guidance and Control Loop

Initial Considerations for Guidance


4. In order to deliver a warhead accurately, the guidance system must be able to calculate the relative positions and velocities
of the weapon and the target, determine errors in the weapon flight path with respect to the required flight path, and then provide
inputs to a control system to steer it. A simple guidance and control loop is illustrated in Fig 1.
5. As the guidance system must determine a position and velocity relationship between the weapon and the target it requires
information about the target. If the target is stationary then position information is in essence all that is required although in
practice other data are normally used to refine relative target position. If the target is moving then the weapon needs a
continuous flow of relative position and velocity information, and a variety of sensors and techniques may be used.

Guidance Agencies
6. In general electromagnetic or acoustic information is used to determine the position or velocity of targets. In medium and
long range air-to-surface and surface-to-surface systems, inertial or mixed inertial/map-matching techniques may be employed
either with or without some target sensing system for the terminal phase.
7. Visible Light. Visible light provides an obvious means of setting up a line of sight. The human eye can usually discriminate
between closely spaced targets and between a target and its background but although target direction is readily determined,
range-finding is rather more difficult. The major disadvantage of visible light is that it is absorbed and scattered by clouds, haze
and smoke, and so a visible light system can be severely hampered by these phenomena. Furthermore camouflage and very high
relative velocities can also degrade the system. Nevertheless optical systems using visible light are often employed in short
range guidance applications.
8. Infra-Red Radiation. All bodies at temperatures above absolute zero radiate infra-red energy, and at the temperatures
associated with engines there is a considerable amount of radiation emitted in the infra-red band. As it is close to visible light in
the electromagnetic spectrum, it is subject to similar absorption and scattering effects. Countermeasures against infra-red
detection are difficult to devise, although decoys can be effective. An infra-red detector must have a narrow bandwidth so that it
can discriminate between the target emissions and other radiation. Photo-conductive cells are used in which the absorption of IR
energy causes a change in the cells electrical resistance or conductivity. The first cells to be produced used lead sulphide but
later cells use lead selenide, lead telluride or indium antimonide. Detectors are generally cooled to around 77o K to reduce

Page 293 : Wed Jan 02 02:42:36 2002

Weapons Applications

8- 2- 5- 2

DO NOT DISTRIBUTE

AP 3456

internal noise and thus improve detectivity. Although lead sulphide cells are the most sensitive they operate at wavelengths
between 2 and 4 microns and are limited to stern attacks only. Other detectors can operate in the range 4.5 to 5 microns and can
be fitted with filters to cut out lower wavelengths, thus reducing the effect of bright skies and solar radiation. These cells permit
all round attacks against targets heated by skin friction.
9. Radio/Radar. The radio/radar band of the electromagnetic spectrum provides perhaps the most widely used method of
obtaining target position and velocity. Radar applications are generally limited in range to the optical horizon and are subject to
some scattering and attenuation by the atmosphere and by precipitation, although to a much lesser extent than visible or infra-red
wavelengths. Due to the longer wavelength, target resolution is generally not as high as with visible and infra-red systems;
however radar has the advantage that it readily provides range information.
10. Acoustic Propagation. All moving targets with the exception of space craft generate noise. However, it is only underwater
that this target characteristic can be exploited, since the velocity of sound in water is about five times that in air and sound
attenuation is considerably less. Sound waves are, however, refracted as they pass through different temperature layers in the
sea, and target noise may be masked by other man-made and natural noises.
11. Inertial/Map-Matching Techniques. Inertial systems are normally used in medium and long range applications, particularly
where Earth referencing is necessary. On their own, inertial systems cannot be used to locate moving targets, and will rarely
have the accuracy to locate small targets without augmentation by map-matching or electromagnetic systems in the terminal
phase. Inertial and digital map-matching systems have the advantage of being totally self-contained and resistant to
countermeasures.

Control
12. The control system of a guided weapon has three main functions:
a. To accept manoeuvre demands from the guidance system and convert these into the required control movements.
b. To provide roll or roll-rate stabilization.
c. To provide dynamic stability.
There are two main types of control system; aerodynamic control and thrust control.
13. Aerodynamic Control. Aerodynamic controls may be provided by nose (or canard) controls, tail controls, moving wings or
by conventional ailerons or elevons. These may be configured to provide either cartesian or polar control. In cartesian control
the two sets of control surfaces are at right angles to each other and the guidance demands are supplied as Up/Down,
Left/Right signals. These cause the weapon to change direction in a series of flat skidding turns without banking. Polar
control is sometimes known as twist and steer and is similar to that employed in aircraft. The weapon has only one set of
wings, and rolls or banks until the target lies in the pitch plane, when direction is then changed by altering the pitch angle.
14. Thrust Control. Thrust control may be used to vary either direction or speed. Directional control may be achieved by
swivelling the main motor on gimbals, or by deflecting the rocket exhaust. Speed control is achieved by cutting off the main
motor thrust when the desired velocity is reached and then using small vernier motors to enable minor adjustments to be made.

Homing Guidance Systems


15. In homing systems the weapon has a homing head or seeker which looks forward towards the target.
16. Passive Homing. In passive homing the seeker head locks on to energy radiated directly by the target as illustrated in Fig 2.
This energy gets more powerful as the target is approached and so discrimination improves with decreasing range; accuracy is
therefore high. After launch the delivery vehicle plays no further part in the attack and so can manoeuvre freely to avoid
defences. As energy is not transmitted by the guidance system the target has no direct warning of the attack, although it may be
able to detect the presence of the launch vehicle and may possibly be able to detect the weapons motor emissions. In general
passive homing systems are used in short range applications, although they may be used in the terminal phase of longer range
systems. Passive systems are prone to defeat by fairly simple countermeasures and decoy devices. As the target is controlling
the emissions it can often stop or reduce these if it suspects it is under attack. As the homing head is a broad-band device decoys
can be effective; these may include infra-red flares, active radar decoys and underwater noise decoys. A formation or group of
targets within the seeker heads field of view may cause the weapon to be confused and home to the centroid of the total return
thus missing any individual target. Examples of passive homing systems are infra-red air-to-air missiles, anti-radar missiles and
passive torpedoes.

8-2-5-2 Fig 2 Passive Homing

Page 294 : Wed Jan 02 02:42:36 2002

Weapons Applications

8- 2- 5- 2

DO NOT DISTRIBUTE

AP 3456

8-2-5-2 Fig 3 Semi-Active Homing

8-2-5-2 Fig 4 Active Homing

Page 295 : Wed Jan 02 02:42:37 2002

Weapons Applications

8- 2- 5- 2

DO NOT DISTRIBUTE

AP 3456

17. Semi-Active Homing. In a semi-active system the weapon homes onto energy which is reflected from a target which has
been designated by a target illuminator. The concept is illustrated in Fig 3. Because the illumination can be tuned to a specific
frequency, or narrow band of frequencies to cater for Doppler shift, decoying is more difficult to achieve. However there are
disadvantages. The illuminator is in use for the whole of the engagement and so cannot leave the target area or be used to
illuminate further targets; more equipment in the form of an illuminator and target tracker is needed; and the target is warned by
the illumination transmissions that it is under attack. Examples of semi-active systems are found in many surface-to-air guided
missiles, long range air-to-air missiles and in laser guided bombs.
18. Active Homing. In active homing systems the illuminating transmitter is carried in the weapon (Fig 4). Thus after launch
the launch vehicle again plays no further part in the attack. The disadvantages of active homing systems mostly stem from the
need to carry extra equipment in a relatively small weapon. The need for a transmitter and receiver leads to increased weight
and complexity, reduced reliability, and a modest operating range. It may also be difficult to accommodate effective EPM (See
Vol 4, Pt 1, Sect 2, Chap 4).

Command Guidance Systems


19. In command systems the weapon looks back for commands. The weapon is ignorant of target position and depends upon a
commanding agency to direct its trajectory.
20. Beam Riding. In this system a beam is projected onto the target and the weapon in flight looks back and senses when it is
off the beam centre line. Error signals are sent to the controls to direct the missile back to the centre of the beam. The system
has the advantage of simplicity and since the weapon sensor is looking away from the target it can be difficult to jam the weapon
itself, although the employment of countermeasures against the target tracking system are likely to be successful due to the high
tracking accuracy required. However there are a number of disadvantages; the beam has to be accurately aligned with the target
during the whole engagement and extra equipment is required both to produce the beam and to track the target. Because the
beam will be divergent accuracy will decrease with range although this may be partly overcome if virtually non-divergent laser
beams are used. In most systems the beam is simply pointed at the target and the missile follows it to impact. This involves a
longer flight path and more g than if a predicted system were employed. Finally the target is warned by the beam transmission
that it is under attack.
21. Command Link. In the command link system the weapon is given commands which are relayed to its controls to alter its
trajectory. Both the weapon and the target must be tracked accurately in order that the system computer can compare their
trajectories and determine the commands to be relayed to the weapon. As with the beam rider the weapon is relatively simple,
requiring no on board guidance system, merely rearward facing sensors to receive the command signals. With a sophisticated
tracking and computing system the weapon can be commanded onto an intercept course, ie a full lead trajectory, which is a
significant improvement over any homing system. The system has a number of disadvantages; both the weapon and target must
be continually tracked during the engagement and accuracy decreases with range. There is a considerable amount of equipment
needed; a target tracker, a weapon tracker, a computer and a command link system. As with the beam rider, the target may be
warned by an illuminating tracker, but many SAMS have a reversionary optical mode.

Variations of Command Trajectories

Page 296 : Wed Jan 02 02:42:37 2002

Weapons Applications

8- 2- 5- 2

DO NOT DISTRIBUTE

AP 3456

22. Command Full-Lead. In command full-lead the computer has knowledge of the target track and of the weapon
characteristics and calculates an intercept point ahead of the target such that the weapon follows a virtually straight line to
impact. This represents the best weapon trajectory to achieve the earliest possible intercept and for this reason it is used in anti
ballistic missile systems. However the system can be expensive as two trackers may be required, one for the target and one for
the weapon.
23. Command to Line-of-Sight (CLOS). In the CLOS trajectory the computer generates no lead angle at all but simply
commands the weapon directly to the target as the target is tracked. The target, weapon and tracker are all in line until intercept
and for this reason the system is sometimes known as Three point guidance. The tracker may be referred to as a differential
tracker as its function is to detect small differences between weapon and target sight lines so that the weapon can be commanded
to reduce the difference to zero. CLOS systems may be manual, semi-automatic (SACLOS), or automatic (ACLOS).
24. CLOS Comparison. In manual CLOS all the operations are done by man. The target and weapon are tracked by eye
through a sight and the weapon is commanded by using a hand controller. Short range anti tank missiles often use this type of
guidance. In a SACLOS system the operator tracks the target, but weapon tracking, computing and command generation is done
automatically. This increase in complexity is usually very cost effective as it reduces the operators task to just tracking the
target; the weapon can be ignored. An example of a SACLOS system is optical Rapier. In an ACLOS implementation the target
is tracked automatically in addition to the weapon, typically using radar; Blindfire Rapier is an example.
25. Partial-Lead Solutions. A computer in a command system can generate any trajectory between the extremes of full-lead
and no-lead (CLOS). Typically such a system can produce a better solution than CLOS but without the cost penalty of full-lead
providing that the tracker can scan both target and weapon simultaneously (ie a track-while-scan radar). The term COLOS,
Command Off-the-Line-of-Sight, is sometimes used for these types of system.
26. Example. Fig 5 shows the trajectories followed by full-lead, half-lead and CLOS.

Hybrid Guidance
27. Both homing and command guidance systems have advantages and disadvantages. Hybrid guidance systems aim to have
the best features of both, together with an increased resistance to electronic countermeasures and the ability to engage multiple
targets. The techniques are most appropriate in surface-to-air missile systems but could be used in air-to-air systems. The
system relies on a phased array track-while-scan fire control radar backed by a real-time digital computer. Both the missiles and
the targets are tracked by the fire control radar. After launch the missile is directed towards the intercept area under command
from the launch agency or by using a simple on board navigation system. Although the fire control radar tracks the missile it
does not routinely command it - eventually the missile will reach a point where the target is within the field of view of its own
receiver or transmitter/receiver and final homing is completed in either active or semi-active mode. Target information is
downlinked from the missile radar and can be correlated with data from the fire control radar thus enabling EPM action to be
taken if there is a mismatch in information from the two sources. An additional advantage is that the missile sensor or radar is
switched on only for the last few seconds before impact thus reducing the opportunity to jam the missile.

8-2-5-2 Fig 5 Typical Command Trajectories

Page 297 : Wed Jan 02 02:42:37 2002

Weapons Applications

8- 2- 5- 2

DO NOT DISTRIBUTE

AP 3456

Weapons Applications
Chapter 3 - Ballistics
Ballistics and Dispersion
1. The trajectory which a missile follows when released from an aircraft is determined by the motion of the aircraft and of the
missile at the moment of release, and by the forces acting upon the missile subsequent to release. Each type of missile exhibits
different flight characteristics, and moreover, individual missiles of a type behave in a slightly different manner to others of the
same type. This variation is primarily due to manufacturing tolerances, stability of the missile, and variations in release
velocities due, for example, to difference in ERU charge. In the case of guns, vibration, heating, barrel wear and the
aerodynamic movement of externally mounted gun pods, contribute to this effect. The spread of these trajectories around a
mean trajectory is called ballistic dispersion. This chapter will investigate the principles involved in the ballistics of bombs
released in level, dive, and loft/toss attacks, and in the ballistics of gun projectiles.

BOMBS
Release Disturbance
2. When a bomb is released from an aircraft it passes initially through an area of significantly disturbed airflow under the
wings, fuselage or in a bomb bay. In general this disturbance imparts an oscillatory motion to the bomb and the extent of the
disturbance will depend upon release conditions, upon the station from which the bomb is released, and whether the bomb is
released singly or as part of a multiple release. The result is further complicated by whether a single release is in the presence or
not of other bombs on the aircraft, and - in multiple releases - by the order in which the bombs are released and by the interval
between releases.
3. The effect of release disturbance can be minimized by keeping to a minimum that time which the bomb spends in the
disturbed airflow. For this purpose ejector release units (ERUs) are used to impart an additional velocity to the bomb on release.
Release disturbance is increased in dive attacks when the effect of gravity on the separation of the bomb from the aircraft is less
than in level flight, and is at a maximum in conditions of decreasing g as in a bunted dive attack. Conversely, release
disturbance will be at a minimum in conditions of increased g such as in a toss attack.

Level Releases
4. Time of Flight. If a body is released anywhere within the Earths gravitational influence then it will tend to accelerate
towards the centre of the Earth. The acceleration imparted by the gravitational force varies slightly with location and altitude but

Page 298 : Wed Jan 02 02:42:37 2002

Weapons Applications

8- 2- 5- 3

DO NOT DISTRIBUTE

AP 3456

2
for practical purposes can be assumed to be constant at 32.174 ft s . If a bomb is released from an aircraft in level flight, and
assuming that no ejector release unit is used, then the initial vertical velocity of the bomb is zero, and its horizontal velocity is
the same as that of the aircrafts true air speed (TAS). The subsequent trajectory of the bomb is determined by the acceleration
towards the ground, g, and by the effects of drag. Assuming an ideal bomb, ie one with no drag, then the time taken for it to hit
the ground can be found using the equations of motion. If the bomb is released from a height h, then the time of flight, t, is
found as follows (where u = initial velocity and g = acceleration due to gravity):
1
h = ut + gt2
2
However u = 0, therefore h = 12 gt2
ie
t=

2h
g

Notice that the horizontal velocity of the aircraft has no influence over the time of flight of the bomb.
5. Forward Throw. The forward throw, which is the horizontal distance travelled by the bomb prior to ground impact, is found
by multiplying the time of flight by the aircrafts horizontal velocity (TAS). This assumes the ideal bomb and zero wind. In
practice every bomb has some drag and its time of flight is longer than the ideal bomb, especially if the bomb is retarded.
However non-retarded bombs delivered at low altitude in level or dive attacks have times of flight and forward throws very
similar to the ideal bomb. Actual values are found from tables for each type of bomb, both retarded and non-retarded, aircraft,
and release conditions. The effect of wind will be covered in Chapter 4.

Loft/Toss and Dive Attacks


6. Bombs released in a steady climb are termed lofted bombs whilst those released under g in a climb are termed tossed
bombs. In a dive attack, the flight path may be under conditions of constant or reducing g (bunted dive). In each case the bomb
follows a parabolic flight path with an initial velocity dictated by the aircraft velocity at release modified slightly by the ERU
ejection velocity.
7. The position of the ideal bomb throughout its flight can be determined, with reference to Figs 1 and 2, by the following
equations, using an up is positive convention, where V = aircraft TAS at release (positive for loft/toss, negative for dive), =
climb/dive angle at release (positive for climb, negative for dive), h = height at release and t = time of flight:
x = Vt cos (1)

y = Vt sin

gt2
+ h for loft toss (2)
2

and y = Vt sin

gt2
+ h for drive (3)
2

8. In equation (3), both V and are negative, as is the sine of the negative angle. The first term therefore becomes positive, ie
in practice the equation becomes identical to equation (2) and positive values of V and can be used in either case. The time of
flight can be determined from equation (2) where, at impact, y = 0, thus:

0 = Vt sin

Page 299 : Wed Jan 02 02:42:37 2002

gt2
+h
2

Weapons Applications

8- 2- 5- 3

DO NOT DISTRIBUTE

AP 3456

ie

gt2
Vt sin h = 0
2

which is a quadratic equation in standard form and can be solved by the equation. So:

t=

V sin +

(V sin ) + 2gh
g

The alternative negative sign before the square root term in the equation has been omitted. Mathematically there are two
occasions when the parabola intersects the x axis, however in practice only the positive solution is relevant.

8-2-5-3 Fig 1 Bomb Trajectory in Loft/Toss Delivery

8-2-5-3 Fig 2 Bomb Trajectory in Dive Attack

9.

The forward throw can then be determined from equation (1).

Page 300 : Wed Jan 02 02:42:38 2002

Weapons Applications

8- 2- 5- 3

DO NOT DISTRIBUTE

AP 3456

10. In Fig 2 the angle ABD is known as the angular gravity drop and is the major component in determining bombing sight
settings. In practice ballistic tables give time of flight and forward throw for the required height of release and TAS. The effect
of ERU velocity is normally included in the result.

GUN PROJECTILES
Introduction
11. Gun projectiles differ from bombs in that they are dispatched from the aircraft at high velocity and, as they are primarily
employed at short ranges to the target, their time of flight is short. In flight however, gun projectiles are subject to the same
aerodynamic and gravitational forces as bombs.

The Firing Sequence


12. When the propellent charge is ignited, gases are evolved which build up pressure in the chamber, accelerating the projectile
along the bore. Eventually a point is reached where the pressure rise ceases as the bore volume behind the projectile increases.
The gases however continue to expand and are finally dispersed at the muzzle. The muzzle velocity of the projectile depends
upon:
a. The average pressure behind the projectile during its travel along the barrel.
b. The cross-sectional area of the barrel.
c. The length of the barrel.
d. The mass of the projectile.
13. Changes in these parameters will induce different muzzle velocities between projectiles or between the mean velocities of
groups of projectiles. Such changes may be caused by:
a. Variations in the charge that lead to fluctuations in the average pressure. These variations may occur because of
differences in charge weight, grain size or ambient temperature.
b. Temperature changes in the barrel that lead to changes in the cross-sectional area of the bore.
c. Minor deviations from standard in the mass of the projectile.
14. Rotation is imparted to the projectile by the driving band which engages with the rifling in the barrel of the gun. Spin rates
of approximately 10,000 rpm are attained and this rotation effectively stabilizes the projectile during its short flight. The
centrifugal forces generated may be used for arming purposes.

Velocity Jump
15. When the projectile leaves the muzzle it has a component velocity along the gunline (VM in Fig 3). The gun itself is
moving with the same velocity as the aircraft (VG), and the projectile therefore also possesses this component of velocity. The
resultant initial path of the projectile (VO) lies between the gunline and the flight path and is inclined to the gunline at an angle
which is termed the velocity jump ().
16. For most attack conditions the angle between the gunline and the aircraft flight path, , is small (typically less than 3) and
since must be less than then must be very small. It can thus be seen that VO is very nearly equal to the arithmetic sum of
VM and VG.

Drag and Gravity Drop


17. A well-established spinning projectile proceeds with its major axis nearly coincident with its trajectory.

8-2-5-3 Fig 3 The Origin of Velocity Jump

Page 301 : Wed Jan 02 02:42:38 2002

Weapons Applications

8- 2- 5- 3

DO NOT DISTRIBUTE

AP 3456

Except for very long ranges this is so very nearly true that the motion of the projectile may be considered to be that of a particle
acted on only by gravity and opposed only by air resistance (drag). If the velocity and air density are known this drag can be
calculated.
18. If air resistance were neglected, the gravity drop for a projectile fired horizontally would be 12 gt2 , where t is the time of
flight of the projectile. Air resistance increases the time of flight over a given range and this increased time of flight may be
taken to be approximately R/Va, where R is the range and Va is the average speed of the projectile over the range R. The gravity
drop, GD, is therefore given by:
1
GD = g
2

R
Va

Weapons Applications
Chapter 4 - Sighting
AIR-TO-GROUND
Introduction
1. The essential problem in air-to-ground weapon aiming is to position the aircraft such that the range and azimuth of the target
from the aircraft are matched by the instantaneous fall of shot of the weapon. At the same time it is necessary to ensure that the
aircraft is not in a position where it is susceptible to damage from the effects of its own weapons or from striking the ground.
2. In modern systems the relative position of the target and the instantaneous fall of shot of the weapon may be computed more
or less continuously - limited by the iteration rate of the computer. The accuracy to which this can be achieved will depend
largely upon the degree of accuracy with which the various parameters, eg range and azimuth to target, TAS, height, wind
velocity, weapon ballistics, are known or can be measured. In less sophisticated systems, or in degraded systems, some of the
parameters may need to be assumed or achieved manually, for example it may be necessary for the pilot to fly at a
predetermined height or airspeed. It is not the intention of this chapter to describe the operation of any particular weapon aiming
system, but rather to describe some of the basic principles of weapon aiming which are common to all systems.
3. In its simplest form an aiming mark is displayed in the weapon aiming sight indicating the current ground position to which
the weapon will fall. The task of the pilot is to fly the aircraft so that this aiming mark coincides with the target. The way in
which this aiming mark position is determined will be described, initially in the zero wind (or zero target movement) situation.

Bombing
4.

Consider an aircraft flying straight and level, at a height H, releasing a bomb at point P and then continuing with the same

Page 302 : Wed Jan 02 02:42:38 2002

Weapons Applications

8- 2- 5- 4

DO NOT DISTRIBUTE

AP 3456

velocity (Fig 1). The ideal (ie dragless) bomb would maintain the same forward speed as the aircraft and would remain
vertically below the aircraft until bomb impact. At this time the bomb would be at Y and the aircraft at Q. In practice all bombs
have some element of drag, and in the case of retarded bombs the drag effect is considerable. Thus the real bomb will impact at
the point T having followed the parabolic path shown; the distance between T and Y is known as the trail distance. The line PT
joining the aircraft position at release to the impact point is known as the line of sight or as the mean trajectory line. The
equivalent diagram for a dive attack is shown in Fig 2 although clearly in this case the aircraft cannot maintain its velocity
indefinitely after release without hitting the ground.

8-2-5-4 Fig 1 Bomb Release from Level Flight

8-2-5-4 Fig 2 Bomb Release from Dive Attack

5. For a given set of flight parameters, ie speed and height, and for any particular weapon, the forward throw can be
determined from tables or by the weapon aiming computer. Similarly the aircraft height above target can be predetermined and
flown or can be found by the aircraft systems, typically by using radar altimeter, laser rangefinder, or radar.

Page 303 : Wed Jan 02 02:42:38 2002

Weapons Applications

8- 2- 5- 4

DO NOT DISTRIBUTE

6.

AP 3456

In the PZT (Fig 1) the angle can be calculated from:

tan =

H
FT

is equal to the angle between the flight path and the sight line and is known as the angular gravity drop. Therefore by
positioning the aiming mark at an angle below the flight path, the correct release range will be achieved when the aiming mark
and target are coincident. In practice the actual release point may need to be slightly earlier than the theoretical release point to
account for pilot reaction time and delays in the weapon release system.

Angle of Attack
7. Although angular gravity drop is strictly measured in relation to the aircraft flight path, in practice an airframe related
datum line is used. This line is determined by the aircraft manufacturer for each aircraft type, and is variously known as fuselage
datum line (FDL), horizontal fuselage datum (HFD), or longitudinal fuselage datum (LFD). The term LFD will be used in this
text. As angular gravity drop is defined with reference to flight path, it is necessary to establish flight path with reference to
LFD.
8. Angle of attack is defined as the angle between the mean chord line and the relative airflow. The angle of incidence is
defined as the angle between the mean chord line of the wing and the LFD. The angle of relevance in weapon aiming is that
between the flight path and the LFD and is known as the fuselage Angle of Attack (AOA).
9.

There are several factors which affect AOA. As fuselage AOA is tied to AOA it is affected in the same way as follows:
a. AUW. As AUW increases, fuselage AOA for a fixed airspeed also increases.
b. Load Factor. Load factor is a measure of lift/weight. As manoeuvre affects the required lift, so load factor increases
with applied g. As g increases, fuselage AOA also increases. In level flight the load factor is one, in a steady dive the load
factor is equal to the cosine of the dive angle.
c. Airspeed. Lift for a given AOA varies with airspeed. If other parameters remain constant, an increase in speed results in
a decrease in AOA. Fuselage AOA follows this variation.
d. Aircraft Configuration. Wing sweep, flaps and other lift variables affect AOA for fixed flight parameters. Also the
fitting of external stores increases the AOA over and above the increase due to the additional weight.
e. Inertia. Rapid pitch changes are not immediately followed by changes in flight path because of aircraft inertia. It is
therefore important to avoid rapid changes in AOA when relying on predetermined values of fuselage AOA.

10. Fuselage AOA is usually determined by formula, by graphical means, or from tables. Alternatively it may be computed on
board by the air data system. It is normally a positive figure, ie LFD is above flight path.

Dip and Parallax


11. The foregoing discussion has assumed that the weapon aiming sight and the weapon are coincident, whereas in virtually all
situations the sight is above and forward of the weapon stations. The vertical error is known as dip; the longitudinal error is
known as parallax. In practice the two errors are usually combined and a joint correction is applied.

8-2-5-4 Fig 3 Relationship of Sight Components

Page 304 : Wed Jan 02 02:42:38 2002

Weapons Applications

8- 2- 5- 4

DO NOT DISTRIBUTE

AP 3456

Depression Angle
12. The final depression angle (D) below LFD that needs to be applied to the aiming mark is the algebraic sum of angular
gravity drop (), fuselage AOA (), and dip and parallax (). These are shown diagrammatically in Fig 3; Fig 3a shows the case
of postive , Fig 3b the case of negative .

Strafing
13. The principle of sight depression is equally applicable to strafe attacks. There are two additional factors to be considered;
velocity jump, which was described in Chapter 3, and gun harmonization or incidence, which is a measure of the depression or
elevation of the gun line with respect to the LFD. All of the components of the strafing sight line depression angle are shown in
Fig 4. Because of the reduced time of flight and consequent smaller gravity drop of gun projectiles compared to bombs, strafe
depression angles will normally be smaller than bombing depression angles.

8-2-5-4 Fig 4 Components of Strafe Depression Angle

Page 305 : Wed Jan 02 02:42:39 2002

Weapons Applications

8- 2- 5- 4

DO NOT DISTRIBUTE

AP 3456

Wind Velocity and Target Movement


14. In order to counter the effect of wind on the weapon or the effect of target movement, the aiming point must be moved into
wind or ahead of the target by an amount equal to the product of wind velocity, or target velocity, and time of flight of the
weapon. For brevity, only wind corrections will be considered; corrections for target movement are essentially the same.
15. It is normally convenient to consider the along and across components of wind velocity separately. The components may
be calculated but in practice they are usually extracted from a diagram such as that shown in Fig 5. The wind vector relative to
the attack heading is back plotted from the origin, the circles being graduated in knots. The along and across components can
then be read from the y and x coordinates respectively of the wind vector tail. Thus in the example shown, a relative wind
velocity of 050/20 is plotted, giving a head wind component of approximately 13 knots and a cross wind component of
approximately 15 knots from the right.

Corrections for Wind Velocity


16. Along Component. The effect of the head or tail component of wind velocity is compensated for by adjusting the weapon
release point, ie reducing the range for head wind and increasing the range for tail wind. In practice the sight depression is
increased or decreased by a set amount for values of head and tail components.
17. Across Component. The ideal bomb remains vertically below the aircraft until bomb impact. Therefore, provided that the
wind velocity does not vary between release height and the ground, if the aircrafts projected track at release lies over the target
the bomb will experience no cross track error. Non-retarded bombs released from low altitude very nearly satisfy these criteria.
However, with retarded bombs and non-retarded bombs released from higher altitudes, the time of flight of the bomb is
significantly greater than the time of flight of the aircraft to the target position. The bomb will therefore experience a greater
wind effect than the aircraft. Conversely weapons with shorter times of flight, gun projectiles, will suffer less wind effect.

8-2-5-4 Fig 5 Wind Velocity Resolving Diagram

18. Ground Distance. The ground distance by which a weapon would miss the target if no cross wind correction were to be
applied can be calculated as the product of the wind velocity and the time of flight of the weapon (both in compatible units).
This distance may be translated into an angular allowance by dividing by the slant range at release, ie:
angular allowance (mils) =

Page 306 : Wed Jan 02 02:42:39 2002

Weapons Applications

8- 2- 5- 4

DO NOT DISTRIBUTE

AP 3456

WVa 1.69 tw 1000


release slant range
where WVa = cross wind component (kts)
tw = weapon time of flight (secs)
The aiming mark may be slewed left or right in the sight by this value.

Aircraft Self-damage
19. Other than enemy action the major sources of potential damage to an aircraft delivering a weapon are hitting the ground
after failure to recover from a dive, and being struck by fragments generated by the exploding weapon.
20. To avoid hitting the ground it is necessary to have a method for determining the height loss involved in recovering from a
dive attack.
21. The fragments from a weapon detonation may be regarded as lying on a hemispherical surface, centred on the explosion,
whose radius first increases and then decreases as gravity becomes the major force acting on the fragments. The determination
of the size of this hemisphere is beyond the scope of this chapter, but given a hemisphere radius it is necessary to calculate a
minimum release range which will ensure that the aircraft avoids the debris hemisphere; this range is known as the minimum
firing range.

Height Loss in a Dive Recovery


22. Fig 6 illustrates a dive recovery profile from which it will be seen that there are two components to the height loss. After
weapon release the aircraft will continue in the dive as a result of pilot reaction time and aircraft inertia which leads to the
element of height loss hr. Thereafter the flight path may be considered to be the arc of a circle of radius R leading to the second
element of height loss, hv.
23. Height Loss in Reaction Time. If the aircraft velocity is V, the reaction time t, and the dive angle, , it will be seen from the
diagram that hr = Vt sin .

8-2-5-4 Fig 6 Height Loss in a Dive Recovery

Page 307 : Wed Jan 02 02:42:39 2002

Weapons Applications

8- 2- 5- 4

DO NOT DISTRIBUTE

AP 3456

8-2-5-4 Fig 7 Acceleration in Dive Recovery

24. Height Loss in Manoeuvre. During its circular flight path the inward acceleration experienced by the aircraft at any point is
the applied g (n), minus the component of the Earths gravitational force as shown in Fig 7. At the initial recovery point this
equates to ng g cos; at the bottom of the dive the inward acceleration is g(n 1). It is however a reasonable approximation
for this purpose to use g(n1) throughout for dive angles less than 25. From considerations of circular motion inward
2
acceleration is also equal to V /R. Therefore:

g(n1) =

V2
R

ie

R=

V2
g(n 1)

From Fig 6
hv = R (R cos ) = R(1 cos )

But R =

V2
g(n 1)

Therefore

Page 308 : Wed Jan 02 02:42:39 2002

Weapons Applications

8- 2- 5- 4

DO NOT DISTRIBUTE

AP 3456

hv =

V2 (1 cos )
g(n 1)

25. Total Height Loss. The total height loss is the sum of hv and hr, ie
V2 (1 cos)
+ Vt sin
g(n 1)

Total height loss =

26. It should be noted that this calculation assumes that V remains constant.

Minimum Firing Range (FR(MIN))


27. FR(MIN) is normally calculated to provide safe clearance from a debris hemisphere of known radius. As with the height
loss problem, FR(MIN) comprises two elements, the first of which is attributable to reaction time. Fig 8 illustrates the situation
where FR(MIN) is the sum of the reaction distance FP and the manoeuvre portion PT. The debris hemisphere has a radius h.
The aircraft flies at a velocity V and during the recovery pulls n g to follow a circular flight path of radius R. The reaction time
is t seconds.
28. Reaction Distance. The reaction distance is simply the product of velocity and reaction time; ie FP = Vt.
29. Manoeuvre Portion. In Fig 8, OPT is a right-angled triangle with the side OP = R, OT (the hypotenuse) = R + h and PT
the required distance. By Pythagoras:
2

(R + h) = R2 + P2

ie PT2 = (R + h) R2
2

= (R + 2Rh + h 2 ) R2
= h2 + 2Rh

ie PT =

h2 + 2Rh

However, from circular motion considerations:

R=

V2
g(n 1)

Therefore, PT =

h2 +

2V2 h
g(n 1)

So FR(MIN) = FP + PT

Page 309 : Wed Jan 02 02:42:39 2002

Weapons Applications

8- 2- 5- 4

DO NOT DISTRIBUTE

AP 3456

= Vt +

h2 +

2V2 h
g(n 1)

As with the height loss calculation, a constant value of V is assumed.

8-2-5-4 Fig 8 Minimum Firing Range

30. Height Loss Consideration. FR(MIN) should not be considered in isolation. In shallow angle attacks clearance of the
debris hemisphere will normally ensure safe ground clearance, however this is not necessarily the case in steep dive attacks.
Thus both FR(MIN) and height loss should be calculated for any proposed attack profile.

AIR-TO-AIR
Introduction
31. The essential requirement for successful air-to-air gunnery is for the gun projectile and the target to arrive at the same place
simultaneously. It is therefore necessary to predict the paths of both the projectile and the target. In practice neither of these
predictions can be achieved with absolute accuracy, and in all aiming systems some assumptions have to be accepted. Even if
the target velocity could be measured and predicted with high accuracy, once the gun has been fired any changes in target
velocity, during the time of flight of the projectile, cannot be acted upon.
32. The purpose of a gyro-gunsight is to produce a depressed aiming mark with which the pilot can track the target and which
will then ensure that the projectiles are fired in the correct direction. A head-up display driven by a digital computer may show
similar information or may be used to generate a hot line, which is a representation of the projectile path. The depressed
aiming mark implementation will be discussed in this chapter.
33. The angle that should be generated by the gunsight is the resultant of the following components:
a. Lead for target motion - in the plane of target motion.
b. Gravity drop - in the vertical plane.
c. Velocity jump - in the plane of symmetry of the firing aircraft.
d. Gun harmonization - in the plane of symmetry of the firing aircraft.
e. Dip - in the plane of symmetry of the firing aircraft.
34. The effects of gravity drop, velocity jump, gun harmonization and dip have been covered in the air-to-ground case. It is not

Page 310 : Wed Jan 02 02:42:39 2002

Weapons Applications

8- 2- 5- 4

DO NOT DISTRIBUTE

AP 3456

normal to correct for parallax in the case of forward firing weapons as any correction would be negligible.

Lead for Target Motion


35. The lead angle required for target motion should be computed using the following variables:
a. Present range.
b. Target velocity.
c. Firing velocity.
d. Air density.
e. Projectile ballistics.
In practice only some of these variables are known, constants are normally used for the unknowns.

Sighting Triangle
36. The sighting triangle is a geometrical planar simplification of the air-to-air sighting problem, but nevertheless is useful for
explaining the principles involved and the errors engendered by this flat model are small. The general triangle is shown in Fig 9
where the various parameters are tabulated.
37. Fig 10 shows the case where the target is flying at 90 to the line of sight. The required deflection angle () may be found
from:

sin =

Vt
Va

target speed
mean bullet speed

38. If the targets angle off () is less than 90, the speed at which it appears to cross the line of sight at 90 will be less than its
TAS, and will continue to decrease as the angle off reduces. Only the component of TAS at 90 to the line of sight, known as
the crossing speed, is relevant in the determination of the deflection angle. This situation is illustrated in Fig 11 from where it
will be seen that:
crossing speed = V t sin
and, therefore

sin =

vt sin
va

39. It is assumed and has been found, that although the mean bullet velocity, Va, will vary with range, altitude, and firing
aircraft speed, a mean value for the particular type of ammunition in use, calculated for each practical firing range and for
various altitudes will be reasonably accurate. Also, in the above equation, an average value for velocity jump has been included
in ; any error induced by this approximation is small.

8-2-5-4 Fig 9 The Sighting Triangle

Page 311 : Wed Jan 02 02:42:40 2002

Weapons Applications

8- 2- 5- 4

DO NOT DISTRIBUTE

Fighter
Target
Point of Impact
Sightline
Sight Range
Bullet Path
Air Range
Vf

AP 3456

The attacking aircraft at F.


The aircraft at T.
The point I along the target's flight path where the
bullet strikes the target.
The line joining fighter and target, FT.
The distance between target and figher.
The line joining the fighter and point of impact, FI.
The range from the fighter to the point of impact.
Fighter TAS.

Vt
Vm

Target TAS.

Va

Average bullet speed over the air range

The deflection angle; between the sightline and


gunline.
The angle off; between sightline and the target flight
path.
The angle between the fighter and flight path and
gunline.
Velocity jump.

Muzzle velocity

8-2-5-4 Fig 10 Sighting Triangle - Angle Off = 90 degrees

Page 312 : Wed Jan 02 02:42:40 2002

Weapons Applications

8- 2- 5- 4

DO NOT DISTRIBUTE

AP 3456

8-2-5-4 Fig 11 Sighting Triangle - Angle Off < 90 degrees

40. It will be evident that the major unknown in the solution of the air-to-air sighting problem is a value for the target crossing
speed. Achieving the correct firing range is relatively easy using radar ranging or stadiametric ranging techniques. The manner
in which the gyro-gunsight solves the deflection angle problem will be covered in the annex to this chapter. Although digital
computing techniques are beginning to supplant the gyro-gunsight, it is still in use.

Weapons Applications

Page 313 : Wed Jan 02 02:42:40 2002

Weapons Applications

8- 2- 5- 4A

DO NOT DISTRIBUTE

AP 3456

Annex to Chapter 4 - Gyro-Gunsight


Introduction
1. From the discussion on air-to-air sighting it will be apparent that the principal unknown in the determination of the required
deflection angle is the targets crossing speed. The gyro-gunsight (GGS) attempts to measure this parameter in terms of rate of
turn when the target is followed with the sight.
2. The pilot is presented with two superimposed displays: a fixed display representing the gunline, and a moving aiming mark,
or pipper, the position of which is determined by the gyro part of the system. The angle between the two sight components
equates to the required deflection angle.

The Gyro System


3. A completely unrestrained, or free gyro, three axis gimballed, maintains its spin axis in a constant direction in space and
therefore changes its alignment relative to the aircraft axes as these rotate in a turn. Conversely, a fully-constrained, or caged
gyro, is forced to remain in a fixed alignment relative to the aircraft axes and cannot change this alignment if and when the
aircraft turns. Axes fixed in relation to a spinning gyro rotate relative to the airframe in an equal and opposite manner to the
rotation of the airframe in space.
4. The gyroscopic performance of the GGS lies between these two extremes. In the roll plane it behaves as a free gyro but in
the pitch and yaw planes there are partially restraining forces: whenever a turn is made a balance is gradually established
between the restraining drag force, which tends to decrease any deflection between gyro spin axis and aircraft axis and the
inertial force which tends to hold the gyro to its initial direction in space, and thus tends to increase this same deflection angle.
When the forces are in balance then both gyro axis and the aircraft axis rotate at the same angular rate w, and have an angular
deflection between them.
5. The gyro is situated in a magnetic field established by electromagnets. The gyros magnetic field centre is permanently
aligned with the common gunline/heading and, when caged, the gyros rotor axis is locked to this line. When uncaged, the
spinning gyros axis remains aligned to the gyros field centre unless the aircraft turns. If the aircraft turns in space, then the
fundamental feature of the GGS is that the rate of turn in either the pitch or yaw planes enforces a proportional deflection of the
gyros spin axis away from the gyros field centre, which itself moves with the aircraft. Thus as the aircrafts heading turns in
space, the gyro rotor axis turns in space in the same sense, but lags through an angle which is forced to be proportional both to
the gyro units rate of turn and to the gyros set sensitivity. This relationship defines the term sensitivity which may be
summarized in the equation:
Gyro deection = Sensitivity Rate of turn

8-2-5-4A Fig 1 Gyro and Magnet System

The sensitivity setting, which has the dimension of time, is made by adjustments to the degree of constraint in the GGS design.

Control of Gyro Deflection

Page 314 : Wed Jan 02 02:42:40 2002

Weapons Applications

8- 2- 5- 4A

DO NOT DISTRIBUTE

AP 3456

6. In order to describe the deflection control system, three laws of physics must be invoked:
a. Faradays Law. Any conductor moving in a magnetic field will have a current induced in it, the strength of which
depends upon the strength of the magnetic field and the rate of cutting the lines of force.
b. Lenzs Law. The induced current in the conductor sets up a mechanical force which opposes the motion of the
conductor.
c. The Law of Precession. Any force applied to a gyro will be precessed through 90 in the direction of rotation.
7. Fig 1 illustrates the gyro and magnet system. The assembly has a plane mirror attached to one end of the spindle while a
copper or aluminium dome is attached to the other; the Hookes joint allows freedom of movement in all planes. An electric
motor rotates the gyro assembly at about 3,000 rpm. The metal dome rotates between four pairs of electromagnets the field
strength of which are governed by current flowing through the surrounding wire coils, known as range coils. The magnet system
is rigid with its axis coincident with the gunline. In essence, the magnet axis is responsible for the fixed part of the display; the
gyro dome centre can be regarded as the pipper position.
8. Fig 2 shows the situation when the aircraft is flying straight and level; the axes of the aircraft, magnets and dome are all
coincident. As the spinning dome cuts the magnetic field, a force arises to oppose its motion (Lenzs Law). However, all the
magnets are equidistant from the centre of the dome and therefore all the lines of force are being cut at an even rate; reactions
are therefore all equal. In Fig 2 these reactions are shown as vector arrows from each magnet. The overall effect of the
resultant drag is a tendency to rotate the dome in the direction opposite to that in which it is spinning. This effect is overcome
by the governed motor.

8-2-5-4A Fig 2 Magnet System - Straight and Level Flight

9. When the aircraft turns the relative position of the magnets and the dome changes. The magnet axis is fixed to the aircraft
axis whilst the gyro dome has a tendency, due to gyroscopic rigidity, to remain pointing to its original direction in space. A
point on the dome near the circumference will cut the lines of force at a faster rate than a point near the centre; the induced
current, and therefore the forces, near the circumference are consequently stronger than those near the centre. Fig 3 illustrates
the situation during a climbing turn to starboard. The magnets have been numbered for ease of reference. Magnets 2 and 3 have
moved closer to the circumference of the dome and therefore induce a greater force than that due to magnets 1 and 2 which are
near the dome centre. The resultant force on the dome can be found by vector addition as shown, but this resultant is then
precessed through 90 in the direction of dome rotation. The overall effect is, therefore, to try to move the dome back towards
the magnet centre.
10. While an acceleration continues, the dome will lag behind the magnet system at such a distance that the resultant force
which acts on the dome is equal to the acceleration. Therefore the higher the acceleration, the more the dome will lag before the
forces are in equilibrium.

Page 315 : Wed Jan 02 02:42:41 2002

Weapons Applications

8- 2- 5- 4A

DO NOT DISTRIBUTE

AP 3456

Tracking
11. The tracking task is to fly the aircraft in such a way that the optical line of sight can be swung into coincidence with the
target and that the rotating line of sight can be held on the target. If and when the sightline intercepts a target at range R s, then
the sightline rate of rotation, _ , should be matched to the target's angular rate. Thus for the sightline to remain on target, Rs _
must equal the target's crossing speed.

8-2-5-4A Fig 3 Magnet System - Climbing Turn to Starboard

12. The requirement is that _ = V sin /Rs where V is the targets scalar speed and is the angle between the line of sight and
the targets velocity vector. The attacking pilot must therefore adjust his own rate of turn in an attempt to establish a tracking
line of sight, whereupon he will have implicitly measured the targets crossing speed by establishing the equality between these
angular rates.

Range and Sensitivity


13. The sensitivity of a gyro gunsight is determined by the strength of the current applied to the range coils; variations in the
magnetic field produced by these coils permit an appropriate sight deflection to be produced.
14. Consider the situation where it is necessary to fire at a target from two different ranges, and assume that the bullet velocity
is constant and equal over both ranges. Fig 4 shows the situation where the target crosses the line of sight at 90 at a constant
speed. The two firing positions are A and B and the respective target positions are A' and B'. During the bullet time of flight
from A to A' the target flies from T to A'. Similarly during the longer bullet time of flight from B to B' the target flies from T to
B'. The two triangles AA'T and BB'T are similar and therefore the required lead for target motion is the same in both cases.

8-2-5-4A Fig 4 Target Movement and Bullet Time of Flight

Page 316 : Wed Jan 02 02:42:41 2002

Weapons Applications

8- 2- 5- 4A

DO NOT DISTRIBUTE

AP 3456

15. It will be apparent though that in order to track the target, a higher rate of turn will be necessary for the case where the
target range is smaller. It can be shown that:

rate of turn, ! =

crossing speed
present range

16. The basic gyro gunsight relationship is:


Gyro deflection = sensitivity x rate of turn
Therefore as range decreases the gyro gunsight would produce an increase in deflection angle which was not required. In order
to produce the correct deflection the sensitivity must be adjusted and this is achieved by varying the current in the range coils as
a function of range. The required sensitivity can be shown to be given by:

Sensitivity =

Sight Range
Muzzle Velocity

Gravity Drop and Dip


17. A bullet fired from a gun drops away from the gunline in a curved trajectory under the influence of gravity. If no allowance
was made for this fact the line of sight would not coincide with the cone of fire except at two widely spaced points within the
effective range of the gun (Fig 5). In order to follow the bullet trajectory the sightline would need to be raised initially and then
depressed. This is achieved by unbalancing the magnet system with gravity coils wound round the two vertical poles but
connected in series opposition, so that one pole can be increased in strength while the other is weakened, without affecting the
total field strength. The gyro axis would however shift vertically. The arrangement is shown in Fig 6. The controlling currents
are passed through a sine/cosine resolver, fed with bank information from an attitude indicator, so that gravity drop can be
applied in the vertical plane regardless of bank angle. Dip is calculated as a function of range and allowed for in association
with gravity drop.

8-2-5-4A Fig 5 Gravity Drop

Page 317 : Wed Jan 02 02:42:41 2002

Weapons Applications

8- 2- 5- 4A

DO NOT DISTRIBUTE

AP 3456

8-2-5-4A Fig 6 Gravity Drop Windings

18. As gravity drop depends upon time of flight of the bullet, provision must be made for ballistics and altitude. The gravity
drop circuits are therefore routed through a ballistics box and an altitude compensation unit.

Optical System
19. Fig 7 illustrates a typical GGS optical system. The lamp illuminates two reticules. The light path through the fixed reticle,
which has the fixed display engraved upon it, is reflected by a fixed mirror and passed through a collimating lens before being
reflected by the reflector glass towards the pilot. The light path through the gyro reticule is reflected by the gyro mirror and the
semi-relecting mirror before passing through the collimating lens. The purpose of the collimating lens is to focus the two images
at infinity so that they appear superimposed regardless of pilot head position.

Limitations
20. The GGS is designed to predict the correct deflection for a quarter attack on a straight and level target. The sight
underdeflects against a manoeuvring target by an amount depending upon the target rate of turn and the firing range. Minor
inaccuracies arise when firing at off-design air speeds.

8-2-5-4A Fig 7 Typical GGS Optical Arrangement

Page 318 : Wed Jan 02 02:42:41 2002

Weapons Applications

8- 2- 5- 4A

DO NOT DISTRIBUTE

Page 319 : Wed Jan 02 02:42:41 2002

AP 3456

Weapons Applications

8- 2- 5- 4A

You might also like