MACHINE DESIGN

Basics
Chapter 1 Introduction to Mechanical Engineering

© McGraw Hill 1
Current Mechanical Design focus (major research topics) : # 1

© McGraw Hill 2
 Scope:
 Design automation, including design representation, virtual
reality, geometric design, design evaluation, design
optimization, data-driven design, artificial intelligence in
design, simulation-based design under uncertainty, design of
complex systems, design of engineered materials systems,
shape and topology optimization, engineering design for
global development, ergonomic and aesthetic(人體工學和美
學)considerations in design methodology, and design for
market systems;
 Design of power transmission systems with an orientation
towards interdisciplinary research on design and optimization
of such systems and with an emphasis on new emerging
techniques and novel mechanisms;
 Design education;
© McGraw Hill 3
  Design of energy, fluid, and power handling systems;
 Design innovation and devices, including design of smart
products and materials;
 Design for manufacturing and the lifecycle, including design
for the environment, DFX, and sustainable design;
 Design of mechanisms and robotic systems, including design
of macro-, micro- and nano-scaled mechanical systems,
machine and robotic components, and machine system
design;
 Design theory and methodology, including creativity in
design, decision analysis, preference modeling, user-
centered design, design cognition, entrepreneurship and
teams in design, design prototyping, and design synthesis.

© McGraw Hill 4
 DFX也可以當成是「Design for X」的縮寫，而這個「X」則可
以代表產品在其生命週期的各個面向。

DFM: Design for Manufacturability（可製造性設計）

DFA: Design for Assembly（可裝配性設計）
DFT: Design for Testability（可測試性設計）
DFI: Design for Inspection（可檢驗性設計）
DFR: Design for Repair（可維修性設計）
DFA: Design for Automatic（可自動化設計）
DFC: Design for Cost（符合成本性設計）

© McGraw Hill 5
Current Mechanical Design focus (major research topics) : # 2

© McGraw Hill 6
 • Methodologically, this includes theoretical, experimental, and/or
historical approaches, along with their practical application,
including education.
• Substantially, the journal aims at covering all subjects related to
mechanisms and machines in general, such as: design theory and
methodology, kinematics of mechanisms, rotor dynamics,
computational kinematics, multibody dynamics, dynamics of
machinery, nonlinear vibrations, linkages and cams, gearing and
transmissions, transportation machinery, control and reliability of
mechanical systems, including hydraulic and pneumatic systems,
experimental test techniques, robots and robotic technologies,
mechatronics, micromachines, human-machine interfaces, haptics,
among others.
• Prospective fields of application of mechanism and machine
science to other scientific areas such as bioengineering and
molecular chemistry are also welcome.

© McGraw Hill 7
 Current Mechanical Design focus (major research topics) : # 3

• Machine design involves the conception, design and development of machines. It

requires the knowledge of basic sciences and multiple engineering topics, such as
physics, mathematics, mechanical drawing, mechanics, machine theory, materials,
metallurgy, thermodynamics, fluid mechanics, manufacturing processes,
computation, sensors, actuators, mechatronics, automation and control.

• As the industry evolves and becomes more demanding, it is worth focusing on the
new directions and the most promising achievements in machine design, able to
develop machines that meet the industry requirements.

• The conference is an opportunity for researchers and scientists in the area of

machine design to present their most recent results, to discuss together new
solutions, and to establish new contacts and find out common interests for further
cooperation. The conference will focus on original and new research results on
machine design.
© McGraw Hill 8
 Including topics:
• design theory and methodology Sustainability:可持續性
• design of machine elements 在不損害後代需求的情況
下滿足當代人的需求，同
• design of mechanisms 時確保經濟增長、環境保
• design optimization 護和社會福祉之間的平衡。
• design management
• CAD/CAM/CAE technology for design
• sustainability and renewable energy in design
• reliability of mechanical systems in design
• expert systems in design
• theoretical and experimental analyses in design engineering

© McGraw Hill 9
 • kinematics and dynamics of mechanisms and machines
• nonlinear phenomena in mechanisms, machines and mechanical
systems
• rapid prototyping
• modelling and Simulation
• experimental testing techniques
• intelligent sensors
• micro electro - mechanical systems (MEMS)
• electric, hydraulic and pneumatic systems
• automation, robotic systems, CNC machines, mechatronics
• micromachines

© McGraw Hill 10
• automatic and intelligent control systems
• machine learning algorithms
• human-machine interfaces, haptic systems
• digital twins
• internet of things

Digital twins
Digital Twin 在地點、流程與人員上的應用
https://youtu.be/iVS-AuSjpOQ https://www.youtube.com/watch?v
=RaOejcczPas

© McGraw Hill 11
 Design

To formulate a plan for the satisfaction of a specified need.

Process requires innovation, iteration, and decision-making.
Communication-intensive.
Products should be.
• Functional.
• Safe.
• Reliable.  Functional = Does it work?
 Usable = Can users actually use it without extensive
• Competitive. training or handholding?
 Delightful = Does this give users pleasure? Enough
• Usable. pleasure that they use it…a lot. Enough pleasure that
• Manufacturable. they actually smile the first time they try it. Enough
pleasure that they will tell others about it.
• Marketable.
https://dskaletsky.medium.com/functional-usable-delightful-a883ee2b4388
© McGraw Hill 12
 Mechanical Engineering Design

Mechanical engineering design involves all the disciplines of

mechanical engineering.
Example.
• Journal bearing(軸頸軸承): fluid flow, heat transfer, friction,
energy transport, material selection, thermomechanical
treatments, statistical descriptions, etc.

© McGraw Hill 13
 The Design Process

Iterative in nature.
Requires initial estimation,
followed by continued
refinement.

Access the text alternative for slide images.

Fig. 1–1
© McGraw Hill 14
 Design Considerations
Some characteristics that influence the design
1 Functionality. 14 Noise.
2 Strength/stress. 15 Styling.
3 Distortion/deflection/stiffness. 16 Shape.
4 Wear. 17 Size.
5 Corrosion. 18 Control.
6 Safety. 19 Thermal properties.
7 Reliability. 20 Surface.
8 Manufacturability. 21 Lubrication.
9 Utility. 22 Marketability.
10 Cost. 23 Maintenance.
11 Friction. 24 Volume.
12 Weight. 25 Liability.
13 Life. 26 Remanufacturing/resource recovery.

© McGraw Hill 15
 Computational Tools

Computer-Aided Engineering (CAE).

• Any use of the computer and software to aid in the engineering
process.
• Includes.
• Computer-Aided Design (CAD). https://adsknews.autodesk.com/
• Drafting, 3-D solid modeling, etc. en/views/ai-lab-cvpr-2022/
• Computer-Aided Manufacturing (CAM).
• CNC toolpath, rapid prototyping, etc.
• Engineering analysis and simulation.
• Finite element, fluid flow, dynamic analysis, motion, etc.
• Math solvers.
• Spreadsheet, procedural programming language, equation solver, etc.

© McGraw Hill 16
 CLIP-Forge allows users to describe 3D objects via spoken words, and the system then
generates 3D voxelized (Minecraft-style) models of those objects. This is an early step in
using words for 3D geometry generation, which can help users build entire 3D scenes for
games, movies, and more.

© McGraw Hill 17
 Have you ever wondered how physical objects with hundreds of parts are assembled in CAD? It’s
often a tedious process that JoinABLe automates by learning how a pair of parts connects to form
joints. This work, done with peers at MIT, could be applicable to everything from furniture assembly
to robotic assembly lines.
© McGraw Hill 18
 Point2Cyl, in a joint effort with Stanford University and KAIST, approached this problem by
decomposing point clouds into extrusion cylinders that are fully editable in CAD, and since nearly
any object can be scanned and translated into a point cloud, this method is quite versatile.
© McGraw Hill 19
 Acquiring Technical Information
Libraries.
• Engineering handbooks, textbooks, journals, patents, etc.
Government sources.
• Government agencies, U.S. Patent and Trademark, National
Institute for Standards and Technology, etc.
Professional Societies (conferences, publications, etc.)
• American Society of Mechanical Engineers, Society of
Manufacturing Engineers, Society of Automotive Engineers, etc.
Commercial vendors.
• Catalogs, technical literature, test data, etc.
Internet.
Access to much of the above information.

© McGraw Hill 20
 A Few Useful Internet Sites

www.globalspec.com
www.engnetglobal.com
www.efunda.com
www.thomasnet.com
www.uspto.gov

© McGraw Hill 21
 Tolerances （公差）

Close tolerances
generally increase cost.
• Require additional
processing steps.
• Require additional
inspection.
• Require machines
with lower production
rates.

Access the text alternative for slide images. Fig. 1–2

© McGraw Hill Source: From Ullman, David G., The Mechanical Design Process, 3rd ed., McGraw-Hill, New York, 2003. 22
 Breakeven Points

A cost comparison between two possible production methods.

Often there is a breakeven point on quantity of production.
EXAMPLE
Automatic screw machine.
• 25 parts/hr.
• 3 hr setup.
• $20/hr labor cost.
Hand screw machine.
• 10 parts/hr.
• Minimal setup.
• $20/hr labor cost.
Breakeven at 50 units.
Access the text alternative for slide images. Fig. 1–3
© McGraw Hill 23
 Stress and Strength

Strength.
• An inherent property of a material or of a mechanical element.
• Depends on treatment and processing.
• May or may not be uniform throughout the part.
• Examples: Ultimate strength, yield strength.
Stress.
• A state property at a specific point within a body.
• Primarily a function of load and geometry.
• Sometimes also a function of temperature and processing.

© McGraw Hill 24
 Uncertainty 1
Common sources of uncertainty in stress or strength.
• Composition of material and the effect of variation on properties.
• Variations in properties from place to place within a bar of stock.
• Effect of processing locally, or nearby, on properties.
• Effect of nearby assemblies such as weldments and shrink fits on stress
conditions.
• Effect of thermomechanical treatment on properties.
• Intensity and distribution of loading.
• Validity of mathematical models used to represent reality.
• Intensity of stress concentrations.
• Influence of time on strength and geometry.
• Effect of corrosion.
• Effect of wear.
• Uncertainty as to the length of any list of uncertainties.

© McGraw Hill 25
 Uncertainty 2

Stochastic method.
• Based on statistical nature of the design parameters.
• Focus on the probability of survival of the design’s function
(reliability).
• Often limited by availability of statistical data.

© McGraw Hill 26
 Uncertainty 3

Deterministic method.
• Establishes a design factor, nd.
• Based on absolute uncertainties of a loss-of-function parameter
and a maximum allowable parameter.

loss-of-function parameter
nd = (1 - 1)
maximum allowable parameter

• If, for example, the parameter is load, then

loss-of-function load
Maximum allowable load = (1 - 2)
nd

© McGraw Hill 27
 Example 1–1

Consider that the maximum load on a structure is known with an uncertainty of

±20 percent, and the load causing failure is known within ±15 percent. If the load
causing failure is nominally 2000 lbf, determine the design factor and the
maximum allowable load that will offset the absolute uncertainties.

Solution
To account for its uncertainty, the loss-of-function load must increase to 1 ∕ 0.85,
whereas the maximum allowable load must decrease to 1 ∕ 1.2. Thus to offset the
absolute uncertainties the design factor, from Equation (1–1), should be
1 0.85
Answer =nd = 1.4
1 1.2
From Equation (1–2), the maximum allowable load is found to be
2000
Answer Maximum allowable =
load = 1400 lbf
1.4
© McGraw Hill 28
 Prof. Feng’s Comment on Example 1-1
Percentage Uncertainty
The uncertainty of a measured value can also be presented as a
percent or as a simple ratio (the relative uncertainty). It is computed
as:

e.g. 1000 ± 1 meter (or 1000 meters ± 0.1%)

1 0.85 𝑛𝑛𝑑𝑑 =
1 + 0.15 1.15
= = 1.4375
=nd = 1.4 1 − 0.2 0.8
1 1.2
This could be better for usage in our class.
Good to match the following textbook content;
especially for the design (safety) factor with
the relation between strength and stress

© McGraw Hill 29
 Design Factor Method

Often used when statistical data is not available.

Since stress may not vary linearly with load, it is more common to
express the design factor in terms of strength and stress.
loss-of-function strength S
nd = (1 - 3)
allowable stress σ (or τ )
All loss-of-function modes must be analyzed, and the mode with
the smallest design factor governs.
Stress and strength terms must be of the same type and units.
Stress and strength must apply to the same critical location in the
part.
The factor of safety is the realized design factor of the final design,
including rounding up to standard size or available components.
© McGraw Hill 30
 Example 1–2

A rod with a cross-sectional area of A and loaded in tension with an axial force of P =
2000 lbf undergoes a stress of σ = P∕A. Using a material strength of 24 kpsi and a design
factor of 3.0, determine the minimum diameter of a solid circular rod. Using Table A–17,
select a preferred fractional diameter and determine the rod’s factor of safety.

Solution
Since A = πd2∕4, σ = P∕A, and from Equation (1–3), σ = S ∕ nd, then
P P S
σ= = =
A π d 2 4 nd
Solving for d yields
12
 4 Pnd 
12
 4 ( 2000) 3 
Answer =d =
 π S   π ( 24= 0.564 in
000) 
From Table A–17, the next higher preferred size is 85 in = 0.625 in. Thus, when nd is
replaced with n in the equation developed above, the factor of safety n is
π Sd 2 π ( 24 000) 0.6252
Answer =n = = 3.68
4P 4 ( 2000)
Thus, rounding the diameter has increased the actual design factor.
© McGraw Hill 31
 Renard series are a
system of preferred
numbers dividing an
interval from 1 to 10 into
5, 10, 20, or 40 steps.
This set of preferred
numbers was proposed in
1877 by French army
engineer Colonel Charles
Renard.

© McGraw Hill 32
 Example 1–3 (1)

A vertical round rod is to be used to support a hanging weight. A person will place the
weight on the end without dropping it. The diameter of the rod can be manufactured
within ±1 percent of its nominal dimension. The support ends can be centered within
±1.5 percent of the nominal diameter dimension. The weight is known within ±2
percent of the nominal weight. The strength of the material is known within ±3.5 percent
of the nominal strength value. If the designer is using nominal values and the nominal
stress equation, σnom = P∕A (as in the previous example), determine what design factor
should be used so that the stress does not exceed the strength.
Solution
There are two hidden factors to consider here. The first, due to the possibility of eccentric
loading, the maximum stress is not σ = P∕A (review Chapter 3). Second, the person may
not be placing the weight onto the rod support end gradually, and the load application
would then be considered dynamic.
Consider the eccentricity first. With eccentricity, a bending moment will exist giving
an additional stress of σ = 32M ∕ (πd 3) (see Section 3–10). The bending moment is given by
M = Pe, where e is the eccentricity. Thus, the maximum stress in the rod is given by
P 32 Pe P 32 Pe
σ =+ = + (1)
A πd 3 πd 2 4 πd 3
© McGraw Hill 33
 Example 1–3 (2)

Since the eccentricity tolerance is expressed as a function of the diameter, we will write
the eccentricity as a percentage of d. Let e = ked, where ke is a constant. Thus, Equation
(1) is rewritten as
4P 32 Pk d 4P
σ =2 + e
= 2 (1 + 8ke ) (2)
πd πd 3
πd
Applying the tolerances to achieve the maximum the stress can reach gives
4 P (1 + 0.02)  4P 
σ max
= 1 + 8 ( 0.015
2 
= )  1.166  2  (3)
π  d (1 − 0.01)  πd

= 1.166σ nom
Suddenly applied loading is covered in Section 4–17. If a weight is dropped from a
height, h, from the support end, the maximum load in the rod is given by Equation (4–59)
which is
12
 hk 
W + W 1 + 
F=
 W

© McGraw Hill 34
 Shock and Impact 1

Impact – collision of two masses with initial relative velocity.

Shock – a suddenly applied force or disturbance.

© McGraw Hill 35
 Shock and Impact 2

Example of automobile collision.

• m1 is mass of engine.
• m2 is mass of rest of vehicle.
• Springs represent stiffnesses of various structural elements.
Equations of motion, assuming linear springs.
mx1 + k1 x1 + k2 ( x1 − x2 ) =
0
(4 - 57)
mx1 + k3 x2 − k2 ( x1 − x2 ) =
0
Equations can be solved for any impact velocity

Fig. 4–26
Access the text alternative for slide images.

© McGraw Hill 36
Suddenly Applied Loading 1
Weight falls distance h and suddenly
applies a load to a cantilever beam.
Find deflection and force applied to the
beam due to impact.

Abstract model considering beam as simple spring

From Table A–9, beam 1,
k= F/y =3EI/l 3
Assume the beam to be massless, so no momentum
transfer, just energy transfer
Loss of potential energy from change of elevation is
W(h + δ)
Increase in potential energy from compressing spring is
kδ 2/2
Conservation of energy
W(h + δ) = kδ 2/2 Access the text alternative for slide images.
© McGraw Hill 37
 Suddenly Applied Loading 3

Rearranging
W W
δ −2 δ −2 h=
2
0
k k
Solving for δ
12
W W  2hk 
δ = ± 1 + 
k k  W 
Maximum deflection
12
W W  2hk 
δ = + 1 +  (4 - 58)
k k  W 
Maximum force
12
 2 hk 
F= kδ =
W + W 1 +  (4 - 59)
 W 
© McGraw Hill 38
 Example 1–3 (3)

where F is the force in the rod, W is the weight, and k is the rod’s spring constant. Since
the person is not dropping the weight, h = 0, and with W = P, then F = 2P. This assumes
the person is not gradually placing the weight on, and there is no damping in the rod.
Thus, Equation (3) is modified by substituting 2P for P and the maximum stress is

= (1.166) σ nom 2.332 σ nom

σ max 2=
nd=Smin/σmax
The minimum strength is =0.965/2.332
=0.414
(1 − 0.035) Snom =
S min = 0.965 S nom
Equating the maximum stress to the minimum strength gives

2.332 σ nom = 0.965 S nom

From Equation (1–3), the design factor using nominal values should be

S nom 2.332
Answer nd =
= = 2.42
σ nom 0.965
© McGraw Hill 39
 Example 1–3 (4)
Obviously, if the designer takes into account all of the uncertainties in this example and
accounts for all of the tolerances in the stress and strength in the calculations, a design factor
of one would suffice. However, in practice, the designer would probably use the nominal
geometric and strength values with the simple σ = P∕A calculation. The designer would
probably not go through the calculations given in the example and would assign a design
factor. This is where the experience factor comes in. The designer should make a list of the
loss-of-function modes and estimate a factor, ni, for each. For this example, the list would be

Loss-of-Function Estimated Accuracy ni

Geometry dimensions Good tolerances 1.05
Stress calculation
Dynamic load Not gradual loading 2.0*
Bending Slight possibility 1.1
Strength data Well known 1.05
*Minimum
Each term directly affects the results. Therefore, for an estimate, we evaluate the product
of each term
nd
= ∏
= n i 1.05 ( 2.0)(1.1)(1.05
= ) 2.43
© McGraw Hill 40
 Reliability Method of Design

Reliability method of design – A method of design that relates the

distribution of stresses with the distribution of strengths to achieve
an acceptable success rate.
Reliability, R – The statistical measure of the probability that a
mechanical element will not fail in use.

© McGraw Hill 41
 Probability of Failure

Probability of Failure, pf – the number of instances of failures per

total number of possible instances.
Probability Density Function, PDF – the distribution of events
within a given range of values.
Two common PDFs.

• Gaussian (normal) distribution.

• Weibull distribution.

© McGraw Hill 42
 Gaussian (Normal) Distribution 1

The probability density function (PDF) of the Gaussian distribution

is expressed in terms of its mean, µx, and its standard deviation σˆ x
1  1  x − µx  
2

f ( x) exp  −    (1 - 4)
σˆ x 2π  2  σˆ x  

small σˆ x large σˆ x

Fig. 1–4 Typical plots of the Gaussian distribution

© McGraw Hill 43
 Gaussian (Normal) Distribution 2

pf is obtained by integrating Eq. (1–4).

The variable x is placed in dimensionless form using the
transformation variate.
x − µx
z= (1 - 5)
σˆ x
The transformation variate is normally distributed, with a mean of
zero and a standard deviation and variance of unity.

© McGraw Hill 44
 Cumulative Density Function for Gaussian Distribution 2

A sketch of the standard normal distribution, showing the z

transform is given below.
The normal cumulative density function is labeled Φ(zα).

Fig. 1–5

zα 1  u2 
Φ ( zα )
= ∫ exp  −  du
−∞ 2π  2
α zα ≤ 0
=
1 − α zα > 0
Access the text alternative for slide images.

© McGraw Hill 45
 Table A–10 Cumulative Distribution Function of Normal
(Gaussian) Distribution (partial)
zα 1  u2 
Φ ( zα )
= ∫ exp  −  du
−∞ 2π  2

α zα ≤ 0
=
1 − α zα > 0
Zα 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
0.0 0.5000 0.4960 0.4920 0.4880 0.4840 0.4801 0.4761 0.4721 0.4681 0.4641
0.1 0.4602 0.4562 0.4522 0.4483 0.4443 0.4404 0.4364 0.4325 0.4286 0.4247
0.2 0.4207 0.4168 0.4129 0.4090 0.4052 0.4013 0.3974 0.3936 0.3897 0.3859
0.3 0.3821 0.3783 0.3745 0.3707 0.3669 0.3632 0.3594 0.3557 0.3520 0.3483
0.4 0.3446 0.3409 0.3372 0.3336 0.3300 0.3264 0.3238 0.3192 0.3156 0.3121
0.5 0.3085 0.3050 0.3015 0.2981 0.2946 0.2912 0.2877 0.2843 0.2810 0.2776
0.6 0.2743 0.2709 0.2676 0.2643 0.2611 0.2578 0.2546 0.2514 0.2483 0.2451
0.7 0.2420 0.2389 0.2358 0.2327 0.2296 0.2266 0.2236 0.2206 0.2177 0.2148
0.8 0.2119 0.2090 0.2061 0.2033 0.2005 0.1977 0.1949 0.1922 0.1894 0.1867
0.9 0.1841 0.1814 0.1788 0.1762 0.1736 0.1711 0.1685 0.1660 0.1635 0.1611
1.0 0.1587 0.1562 0.1539 0.1515 0.1492 0.1469 0.1446 0.1423 0.1401 0.1379
1.1 0.1357 0.1335 0.1314 0.1292 0.1271 0.1251 0.1230 0.1210 0.1190 0.1170
1.2 0.1151 0.1131 0.1112 0.1093 0.1075 0.1056 0.1038 0.1020 0.1003 0.0985
1.3 0.0968 0.0951 0.0934 0.0918 0.0901 0.0885 0.0869 0.0853 0.0838 0.0823
Access the text alternative for slide images.

© McGraw Hill 46
 Example 1–4 (1)
In a shipment of 250 connecting rods, the mean tensile strength is found to be S = 45 kpsi and has a
standard deviation of σˆ S = 5 kpsi.
(a) Assuming a normal distribution, how many rods can be expected to have a strength less than
S = 39.5 kpsi?
(b) How many are expected to have a strength between 39.5 and 59.5 kpsi?
Solution
(a) Substituting in Equation (1–5) gives the transform z variable as
x − µ x S − S 39.5 − 45
z39.5 = = = = −1.10
σ? x σS 5
The probability that the strength is less than 39.5 kpsi can be designated as F(z) = Φ(z39.5) = Φ(−1.10).
Using Table A–10, and referring to Figure 1–6, we find Φ(z39.5) = 0.1357. So the number of rods having
a strength less than 39.5 kpsi is,
Answer N Φ ( z39.5 ) =250(0.1357) =33.9 ≈ 34 rods
because Φ(z39.5) represents the proportion of the population N having a strength less than 39.5 kpsi.

Figure 1–6
Access the text alternative for slide images.

© McGraw Hill 47
 Example 1–4 (2)

(b) Corresponding to S = 59.5 kpsi, we have

59.5 − 45
=z39.5 = 2.90
5
Referring again to Figure 1–6, we see that the probability that the strength is less than 59.5 kpsi is
F(z) = Φ(z59.5) = Φ(2.90). Because the z variable is positive, we need to find the value
complementary to unity. Thus, from Table A–10
Φ (2.90) = 1 − Φ (−2.90) = 1 − 0.001.87 = 0.998 13
The probability that the strength lies between 39.5 and 59.5 kpsi is the area between the ordinates at
z39.5 and z59.5 in Figure 1–6. This probability is found to be
p = Φ ( z59.5 ) − Φ ( z39.5 ) = Φ (2.90) − Φ (−1.10)
= 0.998 13 − 0.1357 = 0.862 43
Therefore the number of rods expected to have strengths between 39.5 and 59.5 kpsi is
Answer Np 250(0.862)
= = 215.5 ≈ 216 rods

Fig. 1–6

© McGraw Hill 48
 Discrete Mean and Standard Deviation

xi is the value of an event (i = 1, 2, … k).

fi is the class frequency, or number of times the event xi occurs
within the class frequency range.
The discrete mean is,
1 k
x = ∑ fi xi (1 - 6)
N i =1
The discrete standard deviation is,
k

∑ fi xi 2 − N x 2
sx = i =1
(1 - 7)
N −1
© McGraw Hill 49
 Example 1–5 (1)
Five tons of 2-in round rods of 1030 hot-rolled steel have been received for workpiece stock.
Nine standard-geometry tensile test specimens have been machined from random locations
in various rods. In the test report, the ultimate tensile strength was given in kpsi. The data in
the ranges 62 to 65, 65 to 68, 68 to 71, and 71 to 74 kpsi is given in histographic form as
follows:
Sut (kpsi) 63.5 66.5 69.5 72.5
f 2 2 3 2
where the values of Sut are the midpoints of each range. Find the mean and standard
deviation of the data.
Solution
Table 1–1 provides a tabulation of the calculations for the solution.
Class Midpoint x, kpsi Class Frequency f Extension fx Extension fx2
63.5 2 127 8 064.50
Table 1–1 66.5 2 133 8 844.50
69.5 3 208.5 14 480.75
72.5 2 145 10 513.50
∑ 9 613.5 41 912.25

© McGraw Hill 50
 Example 1–5 (2)
Class Midpoint x, kpsi Class Frequency f Extension fx Extension fx2
63.5 2 127 8 064.50
66.5 2 133 8 844.50
69.5 3 208.5 14 480.75
72.5 2 145 10 513.50
∑ 9 613.5 41 912.25

From Equation (1–6),

1 k 1
Answer =x = ∑
N i =1
f x
i i
9
=(613.5) 68.16667
= 68.2 kpsi

From Equation (1–7),

k

∑ f x
i i
2
− N x 2

41 912.25 − 9(68.16667 2 )
=
Answer sx =
i =1
N −1 9 −1
= 3.39 kpsi
© McGraw Hill 51
 Reliability 1

The reliability is related to the probability of failure by

R= 1− p f (1 - 8)

Example: If 1000 parts are manufactured, with 6 of the parts

failing, the reliability is
6
1−
R= 0.994
= or 99.4%
1000

© McGraw Hill 52
 Reliability 2

Series System – a system that is deemed to have failed if any

component within the system fails.
The overall reliability of a series system is the product of the
reliabilities of the individual components.
n
R = ∏ Ri (1 - 9)
i =1

Example: A shaft with two bearings having reliabilities of 95% and

98% has an overall reliability of,
R = R1 R2 = 0.95 (0.98) = 0.93 or 93%

© McGraw Hill 53
 Relating Design Factor to Reliability 1

Reliability is the statistical probability that machine systems and

components will perform their intended function satisfactorily
without failure.
Deterministic relations between stress, strength, and design factor
are often used due to simplicity and difficulty in acquiring
statistical data.
Stress and strength are actually statistical in nature.

© McGraw Hill 54
 Probability Density Functions 1
Stress and strength are statistical in nature.
Plots of probability density functions shows distributions.
Overlap is called interference of σ and S, and indicates parts
expected to fail.

Fig. 1–7 (a)

Access the text alternative for slide images.

© McGraw Hill 55
 Dimensions and Tolerances 1

Nominal size – The size we use in speaking of an element.

• Is not required to match the actual dimension.
Limits – The stated maximum and minimum dimensions.
Tolerance – The difference between the two limits.
Bilateral tolerance – The variation in both directions from the basic
dimension, for example, 1.005 ± 0.002 in.
Unilateral tolerance – The basic dimension is taken as one of the
limits, and variation is permitted in only one direction, for example,
+0.004
1.005 −0.000 in

© McGraw Hill 56
 Dimensions and Tolerances 2

Clearance – Refers to the difference in sizes of two mating cylindrical

parts such as a bolt and a hole.
• Assumes the internal member is smaller than the external member.
• Diametral clearance – difference in the two diameters.
• Radial clearance – difference in the two radii.
Interference – The opposite of clearance, when the internal member is
larger than the external member.
Allowance – The minimum stated clearance or the maximum stated
interference or mating parts.
Fit – The amount of clearance or interference between mating parts.
GD&T – Geometric Dimensioning and Tolerancing, a comprehensive
system of symbols, rules, and definitions for defining the theoretically
perfect geometry, along with the allowable variation.
© McGraw Hill 57
 Choice of Tolerances

The designer is responsible for specifying tolerances for every

dimension.
Consideration is given to functionality, fit, assembly,
manufacturing process ability, quality control, and cost.
Excessive precision is a poor design choice, in that it limits
manufacturing options and drives up the cost.
Less expensive manufacturing options should be selected, even
though the part may be less than perfect, so long as the needs are
satisfactorily met.

© McGraw Hill 58
 Choice of Dimensions 1

Dimensioning a part is the designer’s responsibility.

Include just enough dimensions.
Avoid extraneous information that can lead to confusion or multiple
interpretations.
Example of over-specified dimensions. With +/– 1 tolerances, two
dimensions are incompatible.

Fig. 1–8
Access the text alternative for slide images.

© McGraw Hill 59
 Choice of Dimensions 2
Four examples of which dimensions to specify

Fig. 1–9
Access the text alternative for slide images.

© McGraw Hill 60
 Tolerance Stack-up

The cumulative effect of individual tolerances must be allowed to

accumulate somewhere. This is known as tolerance stack-up.
Chain dimensioning allows large stack-up of many small tolerances
in series.

Baseline dimensioning minimizes large tolerance stack-up.

Access the text alternative for slide images.

© McGraw Hill 61
 Example 1–7 (1)

A shouldered screw contains three hollow right circular cylindrical parts on the
screw before a nut is tightened against the shoulder. To sustain the function, the
gap w must equal or exceed 0.003 in. The parts in the assembly depicted in
Figure 1–10 have dimensions and tolerances as follows:
a = 1.750 ± 0.003 in b = 0.750 ± 0.001 in
c = 0.120 ± 0.005 in d = 0.875 ± 0.001 in

Fig. 1–10

All parts except the part with the dimension d are supplied by vendors. The part
containing the dimension d is made in-house.
(a) Estimate the mean and tolerance on the gap w.
(b) What basic value of d will assure that w ≥ 0.003 in?
© McGraw Hill 62
 Example 1–7 (2)

Solution
(a) The mean value of w is given by
Answer w = 1.750 − 0.750 − 0.120 − 0.875 =
a −b −c −d = 0.005 in
For equal bilateral tolerances, the tolerance of the gap is

Answer tw =∑ t =0.003 + 0.001 + 0.005 + 0.001 =0.010 in

all

Then, w = 0.005 ± 0.010 in, and

wmax = w + tw = 0.005 + 0.010 = 0.015 in
wmin =w − tw =0.005 − 0.010 =−0.005 in
Thus, both clearance and interference are possible.
(b) If wmin is to be 0.003 in, then, w= wmin + tw= 0.003 − 0.010= 0.013 in. Thus,

Answer d= 1.750 − 0.750 − 0.120 − 0.013 =

a −b −c −w= 0.867 in
© McGraw Hill 63