05-09-2014

Objectives
• After studying the material in this module, student should
LIMITS, FITS AND be able to :
• Read and create limit dimensions.
TOLERANCES • Describe the nominal size, tolerance, limits and allowance
of two mating parts.
Dr. Dayananda Pai • Identify a clearance fit, interference fit and transition fit.
Aero & Auto Engg. Dept., • Describe the basic hole and basic shaft system.
M.I.T. Manipal • Dimension two mating parts using limit dimensions,
unilateral tolerances, and bilateral tolerances.
• Describe the classes of fit and give example of each.

Why study Limits & Fits?

• In manufacturing it is impossible to produce components
to an exact size, even though they may be classified as
identical.
• However, industry does demand that parts should be
produced between a given basic size.
• The difference between these sizes is called the
“tolerance” which can be defined as “the amount of
variation in size which is tolerated”. A broad, generous
tolerance is cheaper to produce and maintain than a
narrow precise one.

1
 05-09-2014

UNILATERAL SYSTEM OF
TOLERANCE
• The maximum and minimum permissible sizes within • This tolerance allows
which the actual size of a component lies are called Limits variation in only one
direction form the basic
size. This method is
• Upper limit advantageous when a
critical size is approached
• Lower Limit as material is removed
• Nominal Size during manufacturing, as
in case of close-fitting
holes and shafts. A
unilateral tolerance is
always all plus or all
minus; that is either the
plus or the minus value be
zero.

BILATERAL SYSTEM OF TOLERANCE

• This tolerance allows • If it is desired to specify an equal variation in both
directions, the combined plus or minus symbol (±) is used
variation in both
with a single value.
directions from the
basic size. Bilateral
tolerances are usually
given with location
dimensions or with any
dimensions that can be
allowed to vary in
either direction

2
 05-09-2014

Definitions
• BASIC SIZE This is the size about which the limits of
particular fit are fixed. It is the same for both “shaft” and
“hole”. It is also called the “nominal size”.
• TOLERANCE: Tolerance is defined as the difference
between maximum and minimum limits of size for a hole
or shaft. It is also the difference between the upper and
lower deviations

Clearance fit
• FIT • This fit results when the shaft size
is always less than the hole size
• A fit may be defined as the relative motion which can exist for all possible combinations
between a shaft and hole (as defined above) resulting within their tolerance ranges.
Relative motion between shaft
form the final sizes which achieved in their manufacture. and hole is always possible.
There are three classes of fit in common use : clearance, • The minimum clearance occurs
transition and interference. at the maximum shaft size and
the minimum hole size.
• The maximum clearance occurs
at the minimum shaft size and the
maximum hole size.

3

TRANSITION FIT
• A pure transition fit occurs
when the shaft and hole are
exactly the same size. This
fit is theoretically the
boundary between
clearance and interference
and is practically impossible
to achieve, but by selective
assembly or careful
machining methods, it can
be approached within very
fine limits.
INTERFERENCE FIT
• This is a fit which always
results in the minimum
shaft size being larger than
the maximum hole size for
all possible combinations
within their tolerance
ranges. Relative motion
between the shaft and hole
is impossible. The
minimum interference
occurs at the minimum
shaft size and maximum
hole size.
05-09-2014
• Practical transition fits result when the tolerance are such
that the largest hole is greater than the smallest shaft and
the largest shaft is greater than the smallest hole. Two
transition fits are given on the data sheet.
• This is a fit which always
results in the minimum shaft
size being larger than the
maximum hole size for all
possible combinations
within their tolerance
ranges. Relative motion
between the shaft and hole
is impossible. The minimum
interference occurs at the
minimum shaft size and
maximum hole size.
4

• Deviation: It is the algebraic difference between a limit of size
and the corresponding basic size.
• Upper Deviation: It is the algebraic difference between the
maximum limit of size and the corresponding basic size. It is
denoted by letters ‘ES’ for a hole and ‘es’ for a shaft.
• Lower Deviation: It is the algebraic difference between the
minimum limit of size and the corresponding basic size. It is
denoted by letters ‘EI’ for a hole and ‘ei’ for a shaft.
• Fundamental Deviation: It is the deviation, either upper or
lower deviation, which is nearest to the zero line for either a
hole or a shaft. It fixes the position of the tolerance zone in
relation to the zero line.
Hole Basis System
• In the hole basis system, the size of the hole is kept
constant and shaft sizes are varied to obtain various types
of fits.
• In this system, lower deviation of hole is zero, i.e. the low
limit of hole is same as basic size. The high limit of the
hole and the two limits of size for the shaft are then varied
to give desired type of fit.
• The hole basis system is commonly used because it is
more convenient to make correct holes of fixed sizes,
since the standard drills, taps, reamers and branches etc.
are available for producing holes and their sizes are not
adjustable. On the other hand, size of the shaft produced
by turning, grinding, etc. can be very easily varied.
05-09-2014
ALLOWANCE
• Allowance is the term given to the minimum clearance (
called positive allowance) or maximum interference
(called negative allowance) which exists between mating
parts. It may also be describe as the clearance or
interference which gives the tightest possible fit between
mating parts.
Shaft Basis System
• In the shaft basis system, the size of the shaft is kept
constant and different fits are obtained by varying the size
of the hole. Shaft basis system is used when the ground
bars or drawn bars are readily available. These bars do
not require further machining and fits are obtained by
varying the sizes of the hole.
• In this system, the upper deviation (fundamental
deviation) of shaft is zero, i.e. the high limit of the shaft is
same as basic size and the various fits are obtained by
varying the low limit of shaft and both the limits of the
hole.
5

• Interchangeability:
Interchangeability occurs when one part in an assembly
can be substituted for a similar part which has been made
to the same drawing. Interchangeability is possible only
when certain standards are strictly followed.
• Universal interchangeability means the parts to be
assembled are from two different manufacturing sources.
• Local interchangeability means all the parts to be
assembled are made in the same manufacturing unit.
• A 50 mm diameter shaft is made to rotate in the bush. The
tolerances for both shaft and bush are 0.050 mm.
Determine the dimension of the shaft and bush to give a
maximum clearance of 0.075 mm with the hole basis
system.
• Solution:
• In the hole basis system, lower deviation of hole is zero,
therefore low limit of hole = 50 mm
High limit of hole = Low limit + Tolerance
= 50.00 + 0.050
= 50.050 mm = 50.050 X10– 3 m
05-09-2014
• Selective Assembly:
• In selective assembly, the parts are graded according to
the size and only matched grades of mating parts are
assembled. This technique is most suitable where close fit
of two components assembled is required.
• Suppose some parts (shafts & holes) are manufactured to
a tolerance of 0.01 mm, then an automatic gauge can
separate them into ten different groups of 0.001 mm limit
for selective assembly of the individual parts. Thus high
quality and low cost can be achieved.
• Selective assembly is used in aircraft, automobile
industries where tolerances are very narrow and not
possible to manufacture at reasonable costs.
High limit of shaft = Low limit of hole – Allowance
= 50.00 – 0.075
= 49.925 mm = 49.925 X10– 3 m
Low limit of the shaft = High limit – Tolerance
= 49.925 – 0.050
= 49.875 mm = 49.875X10– 3 m
6
 05-09-2014

• For each of the following hole and shaft assembly, find Hole : High limit of hole = If we choose high limit of
shaft-tolerance, hole tolerance and state whether the type 50.025 mm hole with high limit of
of fit is (a) clearance, (b) transition, and (c) interference. Low limit of hole = 50.00 shaft then
mm Allowance = 50.025 –
0.05 Hole tolerance = 50.025 – 50.05
• Hole 50 0.000.025 mm Shaft 50  0.005 mm 50.00 = 0.025 mm = – 0.025 (Interference)
0.02 Shaft : High limit of shaft But high limit of hole is
• Hole 50 00..05
00 mm Shaft 50 0.05 mm = 50.05 mm smaller than low limit of
Low limit of shaft = 50.005 shaft (clearance)
0.06 mm Thus, we conclude that
• Hole 50 00..04
00 mm Shaft 50  0.04 mm the type of fit is
Shaft tolerance = 50.05 –
50.005 = 0.045 mm Transition fit.

• Hole : High limit = 50.05
mm
• Low limit = 50.00 mm
• Tolerance = 0.05 mm
• Shaft : High limit = 50 –
0.02 = 49.98 mm
• Low limit = 50 – 0.05 =
49.95 mm
• Tolerance = 49.98 –
49.95 = 0.0 3 mm
• If we select high limit of • High limit = 50.04 mm • If we select high limit of
hole and high limit of • Low limit = 50.00 mm shaft and low limit of
shaft then • Tolerance = 50.04 –
hole, then
• Allowance = 50.05 – 50.00 • Allowance = 50.00 –
49.98 = 0.07 mm • = 0.04 mm
50.06
• If we choose low limit of • = – 0.06 mm
• Shaft : High limit = 50.06
hole and high limit of mm • It is clear that for any
shaft then combination of hole and
• Low limit = 50.04 mm
• Allowance = 50.00 – shaft the allowance will
49.98 = 0.02 mm • Tolerance = 50.06 – be negative.
50.04
• Thus, we conclude that • Thus, we conclude that
the type of fit is • = 0.02 mm the type of fit is
Clearance fit. Interference fit.

7
 05-09-2014

• In a limit system the following limits are specified for a • For the following hole and shaft assembly determine a)
hole and shaft assembly. hole and shaft tolerance and b) type of fit.

0.02 0.02 0.080

Hole= 30 0.00 mm Shaft= 30 0.05 mm • Hole= 20 00..025
00 mm
Shaft= 20  0.005 mm
Determine the tolerance and allowance.
Sol: Find min clearance and max. Clearance
Ans:

8
 05-09-2014

SYSTEM OF LIMITS AND FITS Tolerance grade on different applications

• Rapid growth of national & International Business Fundament Applications
• Economic success of Manufacturing industries. al Tolerance
• International Organisation for Standardisation (ISO) IT01-IT4 For production of gauges and measuring
specifies the internationally accepted system of limits and Instruments
fits.
• IS system of limits and fits comprises 18 grades of
fundamental tolerances to indicate the level of accuracy. IT5-IT7 Super finishing operations and grinding
• They are designated by ‘IT’ followed by a Number.
IT8-IT11 Turning, Boring, Rolling, Extrusion etc.
• 18 grades are IT01, IT0, and IT1 to IT16. (ISO system
comprise 20 grades) IT12-IT14 Sheet metal working
• Fine grades are referred to by the first few numbers. As
the numbers get larger, so the tolerance zone becomes IT15-IT16 Casting, stamping, flame cutting etc.
progressively wider

Derivation of standard tolerance Dimension steps for sizes up to 500mm

• The tolerance grades may be numerically determined in Millimeters Millimeters
terms of the standard tolerance unit ‘i’ where i in microns
is given by Above Upto Above Upto
i (microns)  0.453 D  0.001D - 3 80 120
• (for basic size upto and including 500 mm)
3 6 120 180
• and
I (microns)  0.004 D  2.1 6 10 180 250
• (for basic size above 500 mm upto and including 3150 10 18 250 315
mm),
18 30 315 400
• where D is in mm and it is the geometric mean of the
lower and upper diameters of a particular step in which 30 50 400 500
the component lies.
50 80

9
 05-09-2014

Fundamental deviation
• It is the deviation, either upper or lower deviation, which is
nearest to the zero line for either a hole or a shaft. It fixes
the position of the tolerance zone in relation to the zero
line. It determines the type of fit. The quality of fit is
determined by tolerance grades.
• IS:919 recommend 25 different deviations for both hole
and the shaft.

Designation of holes, shafts & Fit

• Fundamental deviation for shafts for sizes upto 500mm

c  52 D 0.2 for D  40
 (95  0.8 D) for D  40
d  16 D 0.44
e  11D 0.41
f  5.5 D 0.41
g  2.5 D 0.34 ……And upto zc

10
 05-09-2014

Dowel Pins

Values of tolerance grades

Tolerance value Tolerance value
grade grade
IT6 10i IT12 160i
IT7 16i IT13 250i
IT8 25i IT14 400i
IT9 40i IT15 640i
IT10 64i IT16 1000i
IT11 100i

11
 05-09-2014

Tolerance grades IT01 to IT05 Calculation of tolerance and limit of size

• IT01=0.3+0.008D • 60H8f7
• IT0=0.5+0.012D • Step 1: (Calculation of basic size)
• IT1=0.8+0.020D Given size falls in the diameter steps 50-80mm Hence
• IT5=0.7*IT6=0.7*10i=7i and so on.
D= 50  80  63.25mm
• Step 2: (fundamental deviation for hole)
For H hole FD=0
• Step 3:(Fundamental deviation for shaft)
For f shaft es= -5.5D0.41
= -30.113 microns~ -30 microns= -0.030mm

• Step 4: (Tolerance unit) • (ii) Shaft: High limit =Basic size - FD for Shaft=
60.00-0.030=59.970mm
i (microns)  0.453 D  0.001D  1.856microns
Low Limit=High limit – tolerance =
• Step 5: Tolerance grade IT8 for hole =25i=46.4 59.970 – 0.030 = 59.940mm
microns=0.046mm
• Similarly Tolerance grade for IT7=0.030mm
• Step 5: Calculation of limits
• (i) Hole: Low limit
Basic size +FD for hole=60+0.000=60.000 mm
High Limit
Low limit + Tolerance=60.046mm

12
 Mining Machinery Design (MMC 508)
Chapter 2
Manufacturing Considerations in Design

Ajit Kumar
Assistant Professor
Mining Machinery Engineering
 Why study Limits & Fits?
• In manufacturing it is impossible to produce components to an exact
size, even though they may be classified as identical.
• However, industry does demand that parts should be produced between
a given basic size.
• The difference between these sizes is called the “Tolerance” which can
be defined as “the amount of variation in size which is tolerated”. A
broad, generous tolerance is cheaper to produce and maintain than a
narrow precise one.
• The maximum and minimum permissible sizes within which the actual
size of a component lies are called Limits
 Upper limit
 Lower Limit
 Nominal Size
 LIMITS, FITS & TOLERANCE
Terminology: The terms related to limit system as per BIS are as
below
 Size: It is a number expressed in a particular unit
in the measurement of length.
 Basic Size: It is the size based on which the
dimensional deviations are given. It is the size of a
Hole part to which all limits of variation (i.e., tolerances)

Basic Size
are applied. to arrive at final dimensioning of the
Shaft mating parts. The nominal or basic size of a part is
often the same.
 Actual Size: It is the size of the component by actual measurement after it is
manufactured. It should lie between the two limits of size. The difference
between the basic size and the actual size should not exceed a certain limit;
otherwise it will interfere with the interchangeability of the mating parts.
Limits of Size: These are the extreme permissible sizes within
which the operator is expected to make the component. Maximum
limit of size is the greater of the two limit size, whereas the
Minimum limit of size is the smaller of the two limit of size.
Hole: In the B.I.S. system of limits and fits, all
internal features of a component including those
Hole
Basic size
which are not cylindrical are designated as
‘Hole’.
Shaft

Shaft: In the B.I.S. system of limits and fits, all external

features of a component including those which are not
cylindrical are designated as ‘Shaft’.
It may be noted that when an assembly is made of two parts, the
part which enters into the other, is known as enveloped surface (or,
shaft for cylindrical part) and the other in which one enters is called
enveloping surface (or hole for cylindrical part). The term shaft
refers not only to the diameter of a circular shaft, but it is also used
to designate any external dimension of a part. The term hole refers
not only to the diameter of a circular hole, but it is also used to
designate any internal dimension of a part.
  Deviation: It is the algebraic
difference between a size, to its
corresponding basic size. It may
be positive, negative or zero.
 Upper Deviation: It is the
algebraic difference between
maximum limit of size and its
corresponding basic size.
 Lower Deviation: It is the algebraic difference between
minimum limit of size and its corresponding basic size.

 Actual Deviation: It is the algebraic difference between

actual size and its corresponding basic size.
 Lower Deviation

Upper Deviation
Tolerance: It is the difference
Tolerance
between the maximum limit of size
and the minimum limit of size. It is
always positive and is expressed only
Hole
as a number without a sign.
Max.dia.

Min.dia.

Shaft

• Zero line: In graphical representation of the above terms,

the zero line represents the basic size. This line is also
called as the line of zero deviation.
 Lower Deviation

Upper Deviation
Fundamental Deviation: There are 25
Tolerance
fundamental deviations in the B.I.S.

Tolerance
system represented by letter, symbols
Hole
(Capital letters for Holes and small
letters for Shaft).
Max.dia.

Min.dia.

Max.dia.
Min.dia.
Shaft
 Fundamental Tolerance: This is also
called as ‘Grade of Tolerance’. In the
Lower Deviation

Upper Deviation
Tolerance
Indian Standard System, there are 18
grades represented by number
symbols, both for hole and shaft
denoted as IT01, IT0, IT1,
Hole
IT2.....IT16. A high number number
Max.dia.

Min.dia.

denotes greater tolerance zone.

Max.dia.
Min.dia.

Shaft
Fit: It is the relationship that exists between the mating parts, a
hole and shaft with respect to their dimensional difference
before assembly. Three types of fit are given hereunder;

Clearance Fit: It is a fit which always

provides clearance. Here the tolerance
Clearance fit
zone of the hole will be above the tolerance
zone of the shaft. Maximum clearance is
Hole Shaft
the difference between the maximum hole
and minimum shaft. Minimum clearance is
the difference between the minimum hole
and maximum shaft.
 Interference Fit: It is a fit which
Tolerance Zone of Tolerance Zone of
always provides interference. Here
Hole Shaft the tolerance zone of the hole will
be below the tolerance zone of the
Shaft shaft. Maximum interference is the
Hole algebraic difference between the
minimum hole and maximum shaft.
Minimum Interference is the
algebraic difference between the
maximum hole and minimum shaft.
 Transition Fit: It is a fit which may
Shaft
sometimes provides clearance and
Hole Hole
sometimes interference. When this class
Hole
of fit is represented graphically, the
Shaft
Shaft tolerance zone of the hole and shaft will
overlap each other.

Mass Production: Mass production means production of a unit,

component or part in large numbers.
Advantages:
1. Time for the manufacturing of the components is reduced.
2. The cost of pieces is reduced.
3. Spare parts can be quickly made available.

Disadvantages:
1. Special purpose machines are necessary.
2. Jigs and Fixtures are needed.
3. Gauges are to be used instead of conventional precision instruments.
4. Initial expenditure will be very high.
Interchangeability: When the components are mass produced,
unless they are interchangeable, the purpose of mass production
is not fulfilled. By interchangeability, we mean that identical
components, manufactured by different personnel under
different environments, can be assembled and replaced without
any further rectification during the assembly stage, without
affecting the functioning of the component when assembled.

Hole Basis System: Where the size of

the hole is kept constant and the size of
the shaft is varied to get the different
class of fits, then it is known as the
Hole Basis System.
Shaft Basis System: Where the size of the shaft is kept constant
and the variations given to the hole to get the different class of
fits, then it is known as the shaft basis system.
Basic Shaft System

Shaft
Zero line
In the next Chapter, we will be dealing with the
“Statistical Considerations in Design”.
 Mining Machinery Design
MMC 508
Units – 3
Statistical Considerations in Design
Instructor: Dr. Ajit Kumar

Department of Mining Machinery Engineering

IIT (ISM) Dhanbad
17-02-2021 1
Introduction
• Statistics deals with drawing conclusions from a given or
observed data.
• Statistical techniques are used for collection, processing,
analysis and interpretation of numerical data.
• On the basis of statistical analysis, valid conclusions are drawn
and reasonable decisions are taken.
• Statistics enables the engineers to understand the phenomena
of variations and how to effectively predict and control them.
 Contd…
Population: It is defined as a collection of all elements that are studied and about
which the conclusions are being drawn. On the other hand, a sample is defined as a
collection of some, but not all, of the elements of the population.
• The basic data consists of observations, such as the diameters of shafts manufactured
in one shift. In this case, the group of all shafts is called ‘Population’.

• When the group is large, it is not possible to take observations of the entire population.
In such cases, only a small portion of the population is examined and this portion is
called a ‘Sample’. A sample is a part of the population but the converse is not true. It is
easier to study a sample than studying the whole population. It costs less and takes
less time.

• A representative sample has the characteristics of the population in the same

proportions, as they are included in the entire population. Therefore, many times a
sample is analysed instead of the entire population.
 Contd…
• E.g.: Let us consider an example of 100 shafts of hydrodynamic bearing, with recommended
tolerance of 40e7. The shafts are manufactured on lathe and finished on grinding machine.
Their diameters are measured by micrometer and the readings are tabulated in Table 1. The
readings in Table1 are called ‘raw data’. A data is defined as the collection of numbers
belonging to observations of one or more variables. In this case, the diameter of shaft is a
variable and one hundred numerical readings taken by micrometer are a data.
• Raw data is a data before it is arranged or analysed by any statistical method. Raw data does
not show any pattern or trend of population and does not lead to any conclusions. The
objective of statistical techniques is to rearrange and present the data in a useful way so that
the decisions can be taken.
• Now, the data in Table 1 are arranged on the basis of diameter of shaft against the number of
shafts. As shown in Table 2, a particular diameter such as 39.926 mm is written in the first
column and each shaft with this diameter is shown by a mark X against it. Finally, the number
of marks are counted and written in the last column. For example, the total number of shafts
with a diameter of 39.940 mm is 8. The data in Table 2 is further rearranged in the form of
frequency distribution table and presented in Table 3. Frequency distribution is defined as an
organised display of data that shows the number of observations that fall into different
classes.
Table 1. Reading of diameter of 100 Table 2. Diameter against
shafts (mm) number of shafts
Contd…

Table 3. Frequency distribution table

 Contd…
• It is observed from Table 3 that the first ‘class’ consists of shafts with
diameters ranging from 39.926 mm to 39.930 mm. The number of
observations belonging to each class is called ‘class frequency’. Thus, the
class frequency of the first class is 5.
• The range 39.926 to 39.930 mm, which defines the class, is called the ‘class
interval’ and the limits 39.926 mm and 39.930 mm are called lower and
upper class limits.
• The difference between the limits, i.e., (39.930 – 39.925) = 0.005 mm, is
called ‘class width’.
• In Table 3, the classes are equal, that is, the width of the interval from
beginning of one class to the beginning of the next class is same for every
class.
• Equal classes are always preferred in statistical analysis. When classes
are unequal, the distribution is more difficult to interpret.
 Contd…
• Representing Frequency Distribution: There are two methods of
representing the frequency distribution namely, histogram and frequency
polygon as shown in Fig. 1.

 The histogram consists of a set of

rectangles. The base of the rectangle on
the x-axis is equal to the class width, and
the area of the rectangle is proportional to
the class frequency. When the class
widths are of equal size, the heights of
rectangles are proportional to the class
frequencies.

Fig. 1. Frequency distribution

 Contd…
 The frequency polygon is a line graph of class frequency plotted
against class marks or midpoints of class intervals. When a large
number of observations are taken and very small class widths are selected,
the frequency polygon becomes an approximate curve.
• One such frequency curve, which is widely used in the statistical analysis, is
the normal curve as shown in Fig. 2.

Fig. 2. Frequency curve

CHARACTERISTICS OF FREQUENCY CURVES
• Central Tendency: It is the middle point of
distribution. It is sometimes, referred as the
‘Measure of Location’. In Fig. 3, the central
location of curve-2 lies to the right of those of
curve-1 and curve-3. It is also observed that the
central location of curve-1 is equal to that of
curve-3.

• Dispersion: There are certain observations Fig. 3. Comparison of central tendencies

that tend to spread about an average value
called ‘variation’ or ‘dispersion’ of a
population. Dispersion is defined as the
spread of the data in a distribution, that is,
the extent to which the observations are
scattered. In Fig. 4, curve-1 has a wider
spread or dispersion than curve-2. The
central tendency and dispersion are the two
important characteristics of frequency Fig. 4. Comparison of dispersions
distribution.
CHARACTERISTICS OF FREQUENCY CURVES
• Skewness & Kurtosis: There are two more characteristics of
population, namely, skewness and kurtosis as shown in Fig. 5
and Fig.6, respectively. Curves representing data in statistical
analysis are of two types—symmetrical and skewed.
 The normal curve illustrated in Fig. 2 is an example of a
symmetrical curve. It rises and falls smoothly with a bell shape. A
symmetrical curve is widely used in statistical analysis of
tolerances.
 Curves 1 and 2 in Fig. 5 are skewed. In skewed curves, the values
in frequency distribution are concentrated at either the low end or Fig. 5. Comparison of two
the high end of the measuring scale on the horizontal axis. The skewed curves
values are not equally distributed. Curve-1 is skewed to the right
because it tails off toward the high end of the scale. It is also called
a positively skewed curve. Curve-2 is skewed to the left because
it tails off toward the low end of the scale. It is called a negatively
skewed curve. The Weibull distribution used in reliability
analysis of rolling contact bearing is an example of a skewed
curve.
 Kurtosis is the measure of sharp peaks. In Fig. 6, curves 1 and 2
differ in the shape of their peaks. Curve-2 has a sharper peak than
curve-1. They have the same central tendency and both are
symmetrical. However, they have different degrees of kurtosis. Fig. 6. Comparison of
kurtosis
Measure of Central Tendency & Dispersion
• There are different measures of central tendency, such as the mean, the median or the mode.
The most popular unit to measure the central tendency is the arithmetic mean denoted by
the letter μ. Suppose the population consists of N observations 𝑋1 , 𝑋2 ,…, 𝑋𝑁 . The mean is given
by

Or,

If observations 𝑋1 , 𝑋2 ,…, 𝑋𝑘 occur 𝑓1 , 𝑓2 ,…, 𝑓𝑘 times respectively (i.e., occur with frequencies 𝑓1 ,
𝑋2 ,…, 𝑓𝑘 ), the arithmetic mean is given by

Or,
 Contd…
Since, σ 𝑓𝑖 = 𝑁 = total number of observations, the mean is given by

 The dispersion is measured in number of units like the range, the mean deviation or the
standard deviation. The most popular unit for dispersion is the standard deviation denoted by
𝜎.
ො The standard deviation is defined as the root mean square deviation from the mean.
Therefore,

When observations 𝑋1 , 𝑋2 ,…, 𝑋𝑘 occur at frequencies 𝑓1 , 𝑓2 ,…, 𝑓𝑘 the standard deviation is given
by
 Contd…
Since, σ 𝑓𝑖 = 𝑁 = total number of observations, the mean is given by

Therefore,

Squaring both sides,

Using the relationship,

 Contd…
the expression for 𝜎ො 2 is written as

Since,

Hence,
and,

From (a), (b) & (c);

 Contd…
• When observations belong to a sample of a population, it has been observed that by replacing N
by (N – 1) in the denominator of Eq. 5, a better estimate of the standard deviation is obtained.
For large values of N (N > 30), there is practically no difference between these two expressions.
Therefore, the standard deviation in such cases is given by

where S is the standard deviation of observations belonging to the sample of the population.
Variance is defined as the square of the standard deviation. A standard variable Z is defined as

 The standard variable measures the deviation from the mean in the units of the standard
deviation.
Tutorial- 2
P1. One hundred test specimens made of grey cast iron FG 300 are tested on a universal
testing machine to determine the ultimate tensile strength 𝑆𝑢𝑡 of the material. The results are
tabulated as follows:

Calculate: (i) the mean; (ii) the variance; and (iii) the standard deviation for this sample.

17-02-2021 16
Tutorial- 2
Solution:

From Eq. 2

From Eq. 6,

17-02-2021 17
Probability
 Probability is defined as the chance or likelihood that a particular event will occur. In
mathematical terms, it is a number that varies from 0 to 1.
• The probability is 0 when the event is impossible to occur. The probability is 1 when
the event is certain to occur. If there are equal chances that the event will occur or
will not occur, the probability is 0.5.
• In more precise terms, the probability is defined as a number p, which varies from 0 to 1
and which indicates the chance that a particular event E will occur, given that it can
happen f ways out of n equally likely ways. Therefore,
𝒇
𝒑=𝑷 𝑬 =
𝒏
• ෨ The probability that the
If the event does not occur, it is called ‘not E’ and written as 𝐸.
෨ Therefore,
event will not occur is written as 𝑃(𝐸).
𝒒 = 𝑷(𝑬 ෩ ) = 𝟏 – 𝑷(𝑬)
where q is the probability of non occurrence. Also, 𝒑 + 𝒒 = 𝟏
 Contd…
E.g.- Suppose, we toss a coin twenty times (n = 20), and that we get twelve heads (f = 12).
Then the probability of getting heads is given by,
𝒇 𝟏𝟐
𝒑=𝑷 𝑬 = = = 𝟎. 𝟔 𝐨𝐫 𝟐𝟎%
𝒏 𝟐𝟎
Although the answer is 0.6, if the coin is tossed again and again for a large number of times,
the probability that it will come up with heads is 0.5.
P2. Five bolts with internal cracks are accidentally mixed with 95 bolts without any defects.
What is the probability that the assembly shop will use a defective bolt? Also, find out the
possibility of not using the defective bolts.
Answer: Given, Number of bolts without defect = 95, Number of bolts with defect = 5
Step I: Probability of using defective bolts
In this example, the event (E) is to use a defective bolt. Out of 100 bolts, five are defective.
Therefore, the event can happen in five (f = 5) ways out of one hundred (n = 100) equally
likely ways.
𝒇 𝟓
𝒑=𝑷 𝑬 = = = 𝟎. 𝟎𝟓
𝒏 𝟏𝟎𝟎
 Contd…
Step II: Probability of not using defective bolts
෨ namely, not using the defective bolt, is given by,
The probability of not event 𝑃(𝐸),

𝑞 = 𝑃 𝐸෨ = 1 – 𝑃 𝐸 = 1 − 0.05 = 0.95

Also, 𝑝 + 𝑞 = 0.05 + 0.95 = 1

The physical significance of the number 1 is that there is a certainty of using either a defective
or non-defective bolt in the assembly shop.
Probability Distribution
• There is an important concept of random variables in statistical considerations.
• Let us consider an example of one hundred test specimens made of grey cast iron FG
300 which are tested on a universal testing machine to find out the ultimate tensile
strength of the material. The different values of UTS are obtained in each test.
• There are a number of factors that vary from test to test. There are variations in size of
the test specimen because of tolerances. The chemical composition of each specimen
also varies, although within limits.
• The testing on a universal testing machine is called ‘random experiment’ because the
specimens are selected at random.
• The values of UTS obtained in such testing are called ‘random variables’. A random
variable is defined as a variable that takes different values in random
experiments.
• Dimensions of component, weight, strength, forces acting on component, stress or
properties of materials are all random variables. In other words, a random variable is a
variable whose values depend upon the outcome of a random experiment.
 Contd…
• Let us take an example of a random experiment to understand
the above concept. There are two six-sided cubes with
numbers 1 to 6 marked on their surfaces. Every time, the two
cubes are tossed and the sum of numbers that will appear is
denoted as x. Each cube can display any number from 1 to 6. It
depends upon the outcome of tossing. Therefore, tossing the
two cubes is a ‘random experiment’. The variable x, which
is the addition of two numbers, is a ‘random variable’.
• Figure 7 shows the all possible outcomes of these experiments.
There are a total of 36 possible outcomes. The random variable
x has a specific value for each possible outcome. For example,
in the first event shown at the first row and first column, the
variable 𝑥 = 1 + 1 = 2.
• It is seen from Fig. 7, that x is 2 in only one out of 36 likely
outcomes. Therefore, the probability of getting x = 2 is 1 out of
36 or 1/36.
Fig. 7. Outcome of toss of
• The values of x for each of the remaining outcomes and their two dice
probabilities are calculated in a similar manner. The results are
given in Table 4 and shown schematically in Fig. 8.
 Contd…
• Table 4 that gives a list of all possible values of a random variable and the corresponding
probabilities is called a probability distribution table. Sometimes, the designer is interested
in finding out the probability that x is less than a particular value. For example, in previous
example, the designer may be interested in finding out the probability that the ultimate tensile
strength is less than 300 𝑁/𝑚𝑚2 . This is because he has assumed 300 𝑁/𝑚𝑚2 as UTS for
calculating the dimensions of the cast iron component. Referring to Table 4, the probability
that 𝒙 is less than 𝒙𝒊 is obtained by adding the probabilities of all values of 𝒙 up to and
including 𝒙𝒊 . For example, the probability that x is less than or equal to 4 is given by,

• Table 5 shows the values calculated by the above method. The probability that 𝑥 is less than
𝑥𝑖 , is called cumulative probability and is denoted by the cumulative probability function
𝐹(𝑥). Table 5 is called the cumulative probability distribution table. This distribution is also
shown in Fig. 9.
Table 4. Results of random experiment
Contd…

Fig. 8. Outcome of toss of two

dice
Table 5. Cumulative probability
distribution Contd…

Fig. 9. Cumulative probability

distribution
Probability Distributions
• Some standard discrete and continuous
probability distributions
− Gaussian, or normal
− Lognormal
− Uniform
− Weibull
• In statistical analysis, the most popular
probability distribution curve is the normal
curve. Typical plots of the Normal/Gaussian
distribution, with small or large standard
deviation, look like the following as shown in
Fig. 10. The distribution is called normal or
Gaussian. The equation of the normal curve in Fig. 10. Typical plots of the
terms of the standard variable Z is given by Normal/Gaussian distribution
𝟏 𝟐
𝒇 𝒁 = 𝒆−𝒁 Τ𝟐 (8)
𝟐𝝅
 Contd…
• An important characteristic of the normal curve is that the total area below the curve shown in
Fig. 11 from 𝑍 = – ∞ to 𝑍 = +∞ is one or unity. The areas included between different values of
𝑍 are as follows:

Fig. 11. Normal distribution

 Contd…
• In many design problems, it is required to find out the area below the normal curve from 𝑍 =
0 to a particular value of 𝑍 as shown by the shaded area in Fig. 12. The values of this area
are given in Table 6.

Fig. 12. Area below normal curve

Table 6. Area under normal curve from 0 to Z Contd…
Population Combinations
• There are many problems in machine design where it is required to combine two or more
populations in a specific manner to obtain the resultant population.
• As an example, there are two populations in journal bearing—a population consisting of inner
diameter of bearings and a population consisting of outer diameter of shafts. Statistically,
both populations are random variables. The system is interchangeable and a shaft should
match with any bearing selected at random. Further, they are fitted in such a way that there
is a proper clearance between the bearing and the shaft. In this case, subtracting the shaft
population from the bearing-population can form a third population consisting of clearances.
The clearance population is a random variable.
• Consider a simple case of three bearings with diameters 𝐷1 , 𝐷2 and 𝐷3 and two shafts of
diameters 𝑑1 and 𝑑2 . The means of populations for bearing and shaft are given by,
𝐷1 +𝐷2 +𝐷3
𝜇𝐷 = (a)
3
𝑑1 +𝑑2
𝜇𝑑 = (b)
2
 Contd…
Since there are six possible combinations of the shaft and bearing, the population of clearance consists of
the following six elements:

𝐶1 = 𝐷1 − 𝑑1
𝐶2 = 𝐷1 − 𝑑2
𝐶3 = 𝐷2 − 𝑑1
𝐶4 = 𝐷2 − 𝑑2
𝐶5 = 𝐷3 − 𝑑1
𝐶6 = 𝐷3 − 𝑑2

The mean of clearance population 𝜇𝐶 is given by

𝐶1 + 𝐶2 + 𝐶3 + 𝐶4 + 𝐶5 + 𝐶6
𝜇𝐶 =
6
𝐷1 − 𝑑1 + 𝐷1 − 𝑑2 + 𝐷2 − 𝑑1 + 𝐷2 − 𝑑2 + 𝐷3 − 𝑑1 + 𝐷3 − 𝑑2
𝜇𝐶 =
6
2 𝐷1 + 𝐷2 + 𝐷3 − 3 𝑑1 + 𝑑2
𝜇𝐶 =
6
𝐷1 +𝐷2 +𝐷3 𝑑1 +𝑑2
𝜇𝐶 = − 2 (c)
3
From (a), (b) and (c), Contd…
𝜇𝐶 = 𝜇𝐷 − 𝜇𝑑

Therefore, when two populations are subtracted, the mean of the resultant population is obtained by a
subtraction of their individual means. In general, when a population 𝑌 is subtracted from a population 𝑋, the
mean of the resultant population is given by

𝜇 = 𝜇𝑋 − 𝜇𝑌 (9)

Similarly, it can be proved that when two populations are added, the mean of the resultant population is
obtained by an addition of their individual means. Or,

𝜇 = 𝜇𝑋 + 𝜇𝑌 (10)

The standard deviations for populations of bearing and shaft are given by

2 𝐷1 −𝜇𝐷 2 + 𝐷2 −𝜇𝐷 2 + 𝐷3 −𝜇𝐷 2

𝜎ො𝐷 = (d)
3
2 𝑑1 −𝜇𝐷 + 𝑑2 −𝜇𝐷 2
2
𝜎ො𝑑 = (e)
2
Substituting, Contd…
𝐴1 = 𝐷1 − 𝜇𝐷 , 𝐴2 = 𝐷2 − 𝜇𝐷 , 𝐴3 = 𝐷3 − 𝜇𝐷 , and 𝐵1 = 𝑑1 − 𝜇𝑑 , 𝐵2 = 𝑑2 − 𝜇𝑑

The above equations become;

2 𝐴1 2 + 𝐴2 2 + 𝐴3 2
𝜎ො𝐷 = (f)
3
2 𝐵1 2 + 𝐵2 2
𝜎ො𝑑 = (g)
2

The standard deviation of the clearance population is given by,

2 2 2 2
2
𝐷1 − 𝑑1 − 𝜇𝐶 + 𝐷1 − 𝑑2 − 𝜇𝐶 + 𝐷2 − 𝑑1 − 𝜇𝐶 + 𝐷2 − 𝑑2 − 𝜇𝐶 +⋯
𝜎ො𝐶 =
6

Since, 𝜇𝐶 = 𝜇𝐷 − 𝜇𝑑

The above expression is written as;

2 2 2 2 2 2…
2
𝐴1 − 𝐵1 + 𝐴1 − 𝐵2 + 𝐴2 − 𝐵1 + 𝐴2 − 𝐵2 + 𝐴3 − 𝐵1 + 𝐴3 − 𝐵2
𝜎ො𝐶 =
6
 Contd…
2 2 2 2 2 2
2
𝐴1 − 𝐵1 + 𝐴1 − 𝐵2 + 𝐴2 − 𝐵1 + 𝐴2 − 𝐵2 + 𝐴3 − 𝐵1 + 𝐴3 − 𝐵2 …
𝜎ො𝐶 =
6
2 2 𝐴21 +𝐴22 +𝐴23 +3 𝐵12 +𝐵22 −2 𝐴1 +𝐴2 +𝐴3 𝐵1 +𝐵1
𝜎ො𝐶 = (h)
6

Since, 𝐴1 + 𝐴2 + 𝐴3 = 𝐷1 + 𝐷2 + 𝐷3 − 3𝜇𝐷 = 0 and 𝐵1 + 𝐵2 = 𝑑1 + 𝑑2 − 2𝜇𝐷 = 0

The equation (h) is written as;

2
𝐴12 + 𝐴22 + 𝐴23 𝐵12 + 𝐵22 2 2
𝜎ො𝐶 = + = 𝜎ො𝐷 + 𝜎ො𝑑
3 2

Or, 𝜎ො𝐶 = 𝜎ො𝐷 2 + 𝜎ො𝑑 2

Therefore, the standard deviation follows the Pythagorean rule. It can be proved that the above result is
also valid for an addition of two populations. In general, when there are two populations 𝑋 and 𝑌, the
standard deviation of the resultant population, both for addition and subtraction, is given by

𝜎ො𝐶 = 𝜎ො𝑋 2 + 𝜎ො𝑌 2 (11)

 Contd…
In this article, we have considered only
addition and subtraction of two populations. Table 7. Mean and Standard deviations of the function Z
Sometimes, it is required to divide or multiply
one population by another. In some cases, it is
required to square a population or to take an
inverse of the population. Table 7 shows all
such combinations and how to obtain mean
and standard deviation of combinations. The
values given in this table are approximate and
standard textbooks in statistics can be
referred for more exhaustive treatment. In
Table 7,

𝑍 = Function of two independent random

variables 𝑋 and 𝑌

𝑎 = Constant
Design and Natural Tolerances
• The variations in the dimensions of a component occur due to two reasons:
− Because of a large number of chance causes and,
− Due to assignable causes.
• The variations due to chance causes occur at random. They are the characteristics of the
manufacturing method and measurement technique. The variations due to assignable causes can be
located and corrected. When they are corrected, the system is said to be under ‘statistical control’.
• In a statistically controlled system, the dimensions of the component are normally distributed with a
particular value of standard deviation. The natural tolerance is defined as the actual capabilities
of the process, and can be considered as limits within which all but a given allowable fraction
of items will fall. In general, the natural tolerance of a process is the spread of the normal
curve that includes 99.73% of the total population. Referring to the normal curve, shown in Fig. 11,
the values of the standard variables 𝑍1 and 𝑍2 corresponding to this population are –3 and +3. From
Eq. 7, a standard variable 𝑍 is defined as

Now, 𝑋 = 𝜇 + 𝑍. 𝜎ො
 Contd…
Therefore, 𝑋1 = 𝜇 + 3. 𝜎ො and 𝑋2 = 𝜇 − 3. 𝜎ො
Thus , the Natural Tolerances are ±3. 𝜎ො .
On the contrary, the design tolerances are specification limits, set somewhat arbitrarily by the designer from
considerations of the proper matching of the two components and functioning of the assembly. The design
tolerances can be achieved only when the manufacturing process is so selected that the natural
tolerances are within the design tolerances. The percentage of rejected components depends upon the
relationship between these two tolerances. Based on this relationship, the following observations are made:
• When the design tolerance is less than ±𝟑. 𝝈
ෝ , the percentage of rejected components is
inevitable.
• When the design tolerance is equal to ±𝟑. 𝝈 ෝ , there is virtually no rejection provided that the
manufacturing process is centred. For an off-centre process, some components are rejected.
• When the design tolerance is slightly greater than ±𝟑. 𝝈
ෝ , there is no rejection even if the
manufacturing process is slightly off-centre.
It is necessary for the designer to select a manufacturing process for a component in such a way that the
natural tolerance of the process is slightly less than design tolerance. The design tolerance should be
about ±𝟒. 𝝈 ෝ .
Tutorial- 3
P2. It has been observed from a sample of 200 bearing bushes that the internal diameters are
normally distributed with a mean of 30.010 mm and a standard deviation of 0.008 mm. The
upper and lower limits for the internal diameter, as specified by the designer, are 30.02 and
30.00 mm respectively. Calculate the percentage of rejected bushes.

Answer:

17-02-2021 38
Tutorial- 3
P3. The recommended class of transition fit between the recess and the spigot of a rigid
coupling is 60H6-j5. Assuming that the dimensions of the two components are normally
distributed and that the specified tolerance is equal to the natural tolerance, determine the
probability of interference fit between the two components (For corresponding tables and
equation, refer the textbook “Design of Machine Elements by Bhandari”)

Answer:

17-02-2021 39
Assignment
A1. The tolerance specified by the designer for the diameters of a transmission shaft is
25.000 ± 0.025 𝑚𝑚. The shafts are machined on three different machines. It was observed from
the sample of shafts that the diameters are normally distributed with a standard deviation of
0.015 mm for each of the three machines. However, the mean diameter of shafts fabricated on
the three machines is found to be 24.99, 25.00 and 25.01 mm respectively. Determine the
percentage of rejected shafts in each case and comment on the result.
(Answer: Percentage of rejected shafts for Machine A, Machine B and, Machine C are 16.86,
9.5 and 16.86)

A2. The recommended class of fit for the journal and the bush of a hydrodynamic bearing is
40H6-e7. The dimensions of the journal and the bush are normally distributed and the natural
tolerance is equal to the design tolerance. From the considerations of hydrodynamic action and
bearing stability, the maximum and minimum clearances are limited to 0.08 and 0.06 mm
respectively. Determine the percentage of rejected assemblies.
(Answer: Percentage of rejected assemblies is 4.44%)

17-02-2021 40
Assignment
A3. A tension rod is subjected to axial stress within elastic limit. According to Hooke’s law,
𝜎 = 𝐸. 𝜀
It has been observed that the strain (e) in the tension rod is a normally distributed variable with
a mean of 0.001 𝑚𝑚/𝑚𝑚 and a standard deviation of 0.000 07 𝑚𝑚/𝑚𝑚. The modulus of
elasticity (E) is also a normally distributed random variable with a mean of 207 000 𝑁/𝑚𝑚2 and
a standard deviation of 6000 𝑁/𝑚𝑚2 . Determine the mean and standard deviation of the
corresponding stress variable (s). Comment on the analysis.
(Answer: Mean and Standard Deviation of the Stress Population are 207 𝑁/𝑚𝑚2 and 15.69
𝑁/𝑚𝑚2 )

17-02-2021 41
 In the next Chapter, we will be dealing with the
Wire Rope Design

17-02-2021 42