Skip to main content

Full text of "Processes and design for manufacturing"

See other formats



Processes and Design 
for Manufacturing 


Processes and Design 
for Manufacturing 

Second Edition 

SherifD. ElWakil 

University of Massachusetts Dartmouth 



Prospect Heights, Illinois 

To the memory ofMamdouh El-Wakil, M.D., Ph.D. 

For information about this book, contact: 
Waveland Press, Inc. 
P.O. Box 400 

Prospect Heights, Illinois 60070 

Copyright © 1998 by Sherif D. El Wakil 
2002 reissued by Waveland Press, Inc. 

ISBN 1-57766-255-5 

All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in 
any form or by any means without permission in writing from the publisher. 

Printed in the United States of America 

7 6 5 4 3 2 1 

Chapter 1 Overview 1 


1.1 Definition of Manufacturing 1 

1.2 Relationship Between Manufacturing and Standard of Living 2 

1.3 Overview of the Manufacturing Processes 2 

1.4 Types of Production 3 

1.5 Fundamentals of Manufacturing Accuracy 4 

1.6 The Production Turn 6 

1.7 Product Life Cycle 7 

1.8 Technology Development Cycle 8 

1.9 The Design Process 10 

1.10 Product Design: The Concept of Design for Manufacturing 14 
Review Questions 16 

Chapter 2 Concurrent Engineering 17 


2.1 Reasons for Adopting Concurrent Engineering 19 

2.2 Benefits of Concurrent Engineering 20 

2.3 Factors Preventing the Adoption of Concurrent Engineering 2 1 

2.4 The Four Pillars of Concurrent Engineering 22 

2.5 Forces of Change 24 

2.6 A Success Story: Nippondenso 30 
Review Questions 32 


iv Contents 

Chapter 3 Casting and Foundry Work 33 


3.1 Classifications of Casting by Mold Material 34 

3.2 Classifications of Casting by Method of Filling the Mold 52 

3.3 Classifications of Casting by Metal to be Cast 58 

3.4 Foundry Furnaces 63 

3.5 Casting Defects and Design Considerations 68 

3.6 Cleaning, Testing, and Inspection of Castings 72 

3.7 Castability (Fluidity) 75 
Review Questions 76 
Design Example 78 
Design Projects 81 

Chapter 4 Joining of Metals 84 


4.1 Riveting 84 

4.2 Welding 84 

4.3 Surfacing and Hard-Facing 120 

4.4 Thermal Cutting of Metals 121 

4.5 Brazing and Soldering 123 

4.6 Sticking of Metals 128 
Review Questions 130 
Problems 133 
Design Example 133 
Design Projects 137 

Chapter 5 Metal Forming 139 


5.1 Plastic Deformation 140 

5.2 Rolling 145 

5.3 Metal Drawing 155 

5.4 Extrusion 158 

5.5 Forging 176 

5.6 Cold Forming Processes 201 
Review Questions 204 
Problems 207 

Design Example 207 
Design Projects 209 


Chapter 6 Sheet Metal Working 211 


6.1 Press Working Operations 212 

6.2 High-Energy-Rate Forming (HERF) 238 

6.3 Spinning of Sheet Metal 241 
Review Questions 242 
Problems 244 

Design Example 245 
Design Projects 246 

Chapter 7 Powder Metallurgy 248 


7.1 Metal Powders 249 

7.2 Powder Metallurgy: The Basic Process 254 

7.3 Operational Flowchart 258 

7.4 Alternative Consolidation Techniques 258 

7.5 Secondary Consolidation Operations 263 

7.6 Finishing Operations 264 

7.7 Porosity in Powder Metallurgy Parts 266 

7.8 Design Considerations for Powder Metallurgy Parts 268 

7.9 Advantages and Disadvantages of Powder Metallurgy 270 

7.10 Applications of Powder Metallurgy Parts 270 
Review Questions 274 

Problems 275 
Design Project 277 

Chapter 8 Plastics 278 


8.1 Classification of Polymers 279 

8.2 Properties Characterizing Plastics and Their Effect on Product Design 282 

8.3 Polymeric Systems 283 

8.4 Processing of Plastics 291 

8.5 Fiber-Reinforced Polymeric Composites 303 
References 328 

Review Questions 328 
Design Projects 330 

vi Contents 

Chapter 9 Physics of Metal Cutting 331 


9.1 Cutting Angles 332 

9.2 Chip Formation 334 

9.3 Cutting Forces 339 

9.4 Oblique Versus Orthogonal Cutting 343 

9.5 Cutting Tools 348 

9.6 Machinability 353 

9.7 Cutting Fluids 354 

9.8 Chatter Phenomenon 356 

9.9 Economics of Metal Cutting 356 
Review Questions 358 
Problems 359 

Design Project 360 

Chapter 10 Machining of Metals 361 


10.1 Turning Operations 362 

10.2 Shaping and Planing Operations 379 

10.3 Drilling Operations 382 

10.4 Milling Operations 392 

10.5 Grinding Operations 400 

10.6 Sawing Operations 405 

10.7 Broaching Operations 407 

10.8 Nontraditional Machining Operations 408 
Review Questions 411 

Problems 413 

Chapter 11 Product Cost Estimation 415 


11.1 Costs: Classification and Terminology 416 

11.2 Labor Cost Analysis 418 

11.3 Material Cost Analysis 421 

11.4 Equipment Cost Analysis 423 

11.5 Engineering Cost 425 

11.6 Overhead Costs 425 

11.7 Design to Cost 427 

Contents vii 

Review Questions 427 
Problems 428 
Design Project 430 

Chapter 12 Design for Assembly 431 


12.1 Types and Characteristics of Assembly Methods 432 

12.2 Selection of Assembly Method 435 

12.3 Product Design for Manual Assembly 436 

12.4 Product Design for Automatic Assembly 438 

12.5 Product Design for Robotic Assembly 445 

12.6 Methods for Evaluating and Improving Product DFA 446 
Review Questions 459 

Design Project 459 

Chapter 13 Environmentally Conscious Design 
and Manufacturing 460 


13.1 Solid- Waste Sources 462 

13.2 Solid-Waste Management 464 

13.3 Guidelines for Environmentally Conscious Product Design 469 

13.4 Environmentally Conscious Manufacturing 472 

13.5 Environmental Protection and Pollution Control Legislation 473 
Review Questions 475 

Chapter 14 Computer-Aided Manufacturing 476 


14.1 Numerical Control (NC) 476 

14.2 Computerized Numerical Control (CNC) 494 

14.3 Direct Numerical Control (DNC) 498 

14.4 Computer-Aided Part Programming 499 

14.5 Other Applications of Computer- Aided Manufacturing 514 
Review Questions 516 

Problems 518 

Chapter 14 Appendix 520 

viii Contents 

Chapter 15 Industrial Robots 523 


15.1 Reasons for Using Robots 524 

15.2 Methods for Classifying Robots 525 

15.3 Components of a Robot 536 

15.4 End Effectors 537 

15.5 Sensors 540 

15.6 Industrial Applications of Robots 541 
Review Questions 545 

Chapter 16 Automated Manufacturing Systems 547 


16.1 Computer-Integrated Manufacturing (CIM) 548 

16.2 Group Technology (GT) 556 

16.3 Computer-Aided Process Planning (CAPP) 562 

16.4 Material-Requirement Planning (MRP) 565 

16.5 The Potential of Artificial Intelligence in Manufacturing 566 

16.6 Flexible Manufacturing System (FMS) 568 
Review Questions 575 

Appendix Materials Engineering 577 


A.l Types of Materials 577 

A.2 Properties of Materials 580 

A.3 Standard Tests for Obtaining Mechanical Properties 580 

A.4 Phase Diagrams 590 

A.5 Ferrous Alloys 595 

A.6 Aluminum Alloys 603 

A.7 Copper Alloys 604 

References 605 

Index 610 

At the time the author's first book on processes and design for manufacturing was 
published, the main concern of the manufacturing/engineering academic commu- 
nity was the erroneous picture of manufacturing as involving little more than manual 
training (i.e., manual skills acquired by on-site training in machine shops and the like). 
Unfortunately, this distorted view of manufacturing was created and fueled by the 
shallow, descriptive, and qualitative manner in which the vast majority of books then 
covered the subject. Now, design for manufacturing is a "hot topic," and engineers in 
all disciplines are beginning to realize its strategic importance. Many government pro- 
grams are aimed at enhancing the efficiency of product development and design. The 
present text serves to provide engineering students with the knowledge and skills re- 
quired for them to become good product designers. 

The design component in this book has been strengthened by adding four new 

• Chapter 2, Concurrent Engineering 

• Chapter 1 1 , Product Cost Estimation 

• Chapter 12, Design for Assembly 

• Chapter 13, Environmentally Conscious Design and Manufacturing 

Also, whenever applicable, chapters have been supplemented by design examples il- 
lustrating the interaction between design and manufacturing and showing how prod- 
ucts can be designed for producibility, taking factors like the lot size into 
consideration. In addition, some design projects, which were previously assigned at 
the University of Massachusetts Dartmouth, have been given at the end of several 
chapters. Students are encouraged to use computational tools like spreadsheets and 
other software for modeling and analysis. 

The text has also been supplemented with an appendix that covers the fundamen- 
tals of materials engineering. It provides a basis for understanding manufacturing 
processes, as well as for selecting materials during the product design process. It is 



aimed at engineering students who have not taken materials science as a prerequisite 
for a course on manufacturing processes but is not meant as a substitute for any mate- 
rials science textbook. 

The author wishes to acknowledge the contributions of the many corporations and 
individuals who supplied various figures and photographs or provided software to aid 
in producing this book, chief among them Silverscreen. Thanks are also extended to 
reviewers of the manuscript: 

Mary C. Kocak, Pellissippi State Technical Community College 

Zhongming (Wilson) Liang, Purdue University — Fort Wayne 

Wen F. Lu, University of Missouri — Rolla 

Antonio Minardi, University of Central Florida 

Charles Mosier, Clarkson University 

Masud Salimian, Morgan State University 

Richard D. Sisson. Jr., Worcester Polytechnic Institute 

Joel W. Troxler, Montana State University 

David C. Zenger, Worcester Polytechnic Institute 

Yuming Zhang, University of Kentucky 

A note of gratitude also goes to Ana Gonzalez for her hard work in typing the 
manuscript. The author wishes to thank Andrea Goldman and Jean Peck for their 
encouragement and support. Finally, the author must express his profound gratitude to 
his wife and children for their patience as the huge task of completing this second edi- 
tion unfolded. God knows the sacrifice they gave. 

SherifD. El Wakil 
North Dartmouth, Massachusetts 

Chapter 1 



Before learning about various manufacturing processes and the concept of de- 
sign for manufacturing, we first must become familiar with some technical terms 
that are used frequently during the planning for and operation of industrial man- 
ufacturing plants. We also must understand thoroughly the meaning of each of 
these terms, as well as their significance to manufacturing engineers. The ex- 
planation of the word manufacturing and its impact on the life-style of the peo- 
ple of industrialized nations should logically come at the beginning. In fact, this 
chapter will cover all these issues and also provide a better understanding of the 
design process, as well as the different stages involved in it. Finally, the concept 
of design for manufacturing and why it is needed will be explained. 

• /^~*\ 


Manufacturing can be defined as the transformation of raw materials into useful prod- 
ucts through the use of the easiest and least expensive methods. It is not enough, there- 
fore, to process some raw materials and obtain the desired product. It is, in fact, of major 
importance to achieve this goal by employing the easiest, fastest, and most efficient 
methods. If less efficient techniques are used, the production cost of the manufactured 
part will be high, and the part will not be as competitive as similar parts produced by 
other manufacturers. Also, the production time should be as short as possible in order to 
capture a larger market share. 

The function of a manufacturing engineer is, therefore, to determine and define the 
equipment, tools, and processes required to convert the design of the desired product 
into reality in an efficient manner. In other words, it is the engineer's task to find out 
the most appropriate, optimal combination of machinery, materials, and methods 


needed to achieve economical and trouble-free production. Thus, a manufacturing en- 
gineer must have a strong background in materials and up-to-date machinery, as well 
as the ability to develop analytical solutions and alternatives for the open-ended prob- 
lems experienced in manufacturing. An engineer must also have a sound knowledge of 
the theoretical and practical aspects of the various manufacturing methods. 


The standard of living in any nation is reflected in the products and services avail- 
able to its people. In a nation with a high standard of living, a middle-class family 
usually owns an automobile, a refrigerator, an electric stove, a dishwasher, a wash- 
ing machine, a vacuum cleaner, a stereo, and, of course, a television set. Such a 
family also enjoys health care that involves modern equipment and facilities. All 
these goods, appliances, and equipment are actually raw materials that have been 
converted into manufactured products. Therefore, the more active in manufacturing 
raw materials the people of a nation are, the more plentiful those goods and ser- 
vices become; as a consequence, the standard of living of the people in that nation 
attains a high level. On the other hand, nations that have raw materials but do not 
fully exploit their resources by manufacturing those materials are usually poor and 
are considered to be underdeveloped. It is, therefore, the know-how and capability 
of converting raw materials into useful products, not just the availability of miner- 
als or resources within its territorial land, that basically determines the standard of 
living of a nation. In fact, many industrial nations, such as Japan and Switzerland, 
import most of the raw materials that they manufacture and yet still maintain a high 
standard of living. 


The final desired shape of a manufactured component can be achieved through one or 
more of the following four approaches: 

1. Changing the shape of the raw stock without adding material to it or taking mater- 
ial away from it. Such change in shape is achieved through plastic deformation, and 
the manufacturing processes that are based on this approach are referred to as metal 
forming processes. These processes include bulk forming processes like rolling, 
extrusion, forging, and drawing, as well as sheet metal forming operations like 
bending, deep drawing, and embossing. Bulk forming operations are covered in 
Chapter 5, and the working of sheet metal is covered in Chapter 6. 

2. Obtaining the required shape by adding metal or joining two metallic parts to- 
gether, as in welding, brazing, or metal deposition. These operations are covered 
in Chapter 4. 

1.4 Types of Production 3 

3. Molding molten or particulate metal into a cavity that has the same shape as the 
final desired product, as in casting and powder metallurgy. These processes are cov- 
ered in Chapters 3 and 7, respectively. 

4. Removing portions from the stock material to obtain the final desired shape. A cut- 
ting tool that is harder than the stock material and possesses certain geometric char- 
acteristics is employed in removing the undesired material in the form of chips. 
Several chip-making (machining) operations belong to this group. They are exem- 
plified by turning, milling, and drilling operations and are covered in Chapter 10. 
The physics of the process of chip removal is covered in Chapter 9. 


Modern industries can be classified in different ways. These classifications may be by 
process, or by product, or based on production volume and the diversity of products. 
Classification by process is exemplified by casting industries, stamping industries, and 
the like. Classification by product indicates that industries may belong to the automo- 
tive, aerospace, and electronics groups. Classification based on production volume 
identifies three distinct types of production: mass, job shop, and moderate. Let us 
briefly discuss the features and characteristics of each type. We will also discuss the 
subjects in greater depth later in the text. 

Mass Production 

Mass production is characterized by the high production volume of the same (or very 
similar) parts for a prolonged period of time. An annual production volume of less than 
50,000 pieces cannot usually be considered as mass production. As you may expect, the 
production volume is based on an established or anticipated sales volume and is not di- 
rectly affected by the daily or monthly orders. The typical example of mass-produced 
goods is automobiles. Because that type attained its modern status in Detroit, it is some- 
times referred to as the Detroit type. 

Job Shop Production 

Job shop production is based on sales orders for a variety of small lots. Each lot may 
consist of up to 200 or more similar parts, depending upon the customer's needs. It is 
obvious that this type of production is most suitable for subcontractors who produce 
varying components to supply various industries. The machines employed must be 
flexible to handle frequent variations in the configuration of the ordered components. 
Also, the personnel employed must be highly skilled in order to handle a variety of 
tasks that differ for the different parts that are manufactured. 

Moderate Production 

Moderate production is an intermediate phase between the job shop and the mass 
production types. The production volume ranges from 10,000 to 20,000 parts, and 
the machines employed are flexible and multipurpose. This type of production is 

1 Overview 

gaining popularity in industry because of an increasing market demand for cus- 
tomized products. 



Modern manufacturing is based on flow-type "mass" assembly of components into 
machines, units, or equipment without the need for any fitting operations performed on 
those components. That was not the case in the early days of the Industrial Revolution, 
when machines or goods were individually made and assembled and there was always 
the need for the "fitter" with his or her file to make final adjustments before assembling 
the components. The concepts of mass production and interchangeability came into 
being in 1798, when the American inventor Eli Whitney (born in Westboro, Massa- 
chusetts) contracted with the U.S. government to make 10,000 muskets. Whitney 
started by designing a new gun and the machine tools to make it. The components of 
each gun were manufactured separately by different workers. Each worker was as- 
signed the task of manufacturing a large number of the same component. Meanwhile, 
the dimensions of those components were kept within certain limits so that they could 
replace each other if necessary and fit their mating counterparts. In other words, each 
part would fit any of the guns he made. The final step was merely to assemble the in- 
terchangeable parts. By doing so, Eli Whitney established two very important concepts 
on which modern mass production is based — namely, interchangeability and fits. Let 
us now discuss the different concepts associated with the manufacturing accuracy re- 
quired for modern mass production technologies. 


A very important fact of the manufacturing science is that it is almost impossible to ob- 
tain the desired nominal dimension when processing a workpiece. This is caused by 
the inevitable, though very slight, inaccuracies inherent in the machine tool, as well as 
by various complicated factors like the elastic deformation and recovery of the work- 
piece and/or the fixture, temperature effects during processing, and sometimes the skill 
of the operator. Because it is difficult to analyze and completely eliminate the effects 
of these factors, it is more feasible to establish a permissible degree of inaccuracy or a 
permissible deviation from the nominal dimension that would not affect the proper 
functioning of the manufactured part in any detrimental way. According to the Inter- 
national Standardization Organization (ISO) system, the nominal dimension is referred 
to as the basic size of the part. 

The deviations from the basic size to each side (in fact, both can also be on the 
same side) determine the high and the low limits, respectively, and the difference be- 
tween these two limits of size is called the tolerance. It is an absolute value without a 
sign and can also be obtained by adding the absolute values of the deviations. As you 
may expect, the magnitude of the tolerance is dependent upon the basic size and is des- 

1.5 Fundamentals of Manufacturing Accuracy 


The relationship 
between tolerance 
and production cost 


ignated by an alphanumeric symbol called the grade. There are 1 8 standard grades of 
tolerance in the ISO system, and the tolerances can be obtained from the formulas or 
the tables published by the ISO. 

Smaller tolerances, of course, require the use of high-precision machine tools in man- 
ufacturing the parts and, therefore, increase production cost. Figure 1 . 1 indicates the rela- 
tionship between the tolerance and the production cost. As can be seen, very small toler- 
ances necessitate very high production cost. Therefore, small tolerances should not be 
specified when designing a component unless they serve a certain purpose in that design. 


Before two components are assembled together, the relationship between the dimensions 
of the mating surfaces must be specified. In other words, the location of the zero line (i.e., 
the line indicating the basic size) to which deviations are referred must be established for 
each of the two mating surfaces. As can be seen in Figure 1 .2a, this determines the degree 
of tightness or freedom for relative motion between the mating surfaces. Figure 1 .2a also 
shows that there are basically three types of fits: clearance, transition, and interference. 

In all cases of clearance fit, the upper limit of the shaft is always smaller than the 
lower limit of the mating hole. This is not the case in interference fit, where the lower limit 
of the shaft is always larger than the upper limit of the hole. The transition fit, as the name 
suggests, is an intermediate fit. According to the ISO, the internal enveloped part is always 


The two systems of fit 
according to the ISO: 

(a) shaft-basis system; 

(b) hole-basis system 

Basic size 





Hole tolerance zone 


Shaft tolerance zone 


referred to as the shaft, whereas the surrounding surface is referred to as the hole. Accord- 
ingly, from the fits point of view, a key is the shaft and the key way is the hole. 

It is clear from Figures 1.2a and b that there are two ways for specifying and 
expressing the various types of fits: the shaft-basis and the hole-basis systems. The 
location of the tolerance zone with respect to the zero line is indicated by a letter, 
which is always capitalized for holes and lowercased for shafts, whereas the tolerance 
grade is indicated by a number, as previously explained. Therefore, a fit designation 
can be H7/h6, F6/g5, or any other similar form. 


When the service life of an electric bulb is over, all you do is buy a new one and re- 
place the bulb. This easy operation, which does not need a fitter or a technician, would 
not be possible without the two main concepts of interchangeability and standardiza- 
tion. Interchangeability means that identical parts must be able to replace each other, 
whether during assembly or subsequent maintenance work, without the need for any 
fitting operations. Interchangeability is achieved by establishing a permissible toler- 
ance, beyond which any further deviation from the nominal dimension of the part is 
not allowed. Standardization, on the other hand, involves limiting the diversity and 
total number of varieties to a definite range of standard dimensions. An example is the 
standard gauge system for wires and sheets. Instead of having a very large number of 
sheet thicknesses in steps of 0.001 inch, the number of thicknesses produced is limited 
to only 45 (in U.S. standards). As you can see from this example, standardization has 
far-reaching economic implications and also promotes interchangeability. Obviously, 
the engineering standards differ for different countries and reflect the quality of tech- 
nology and the industrial production in each case. Germany established the DIN 
(Deutsche Ingenieure Normen), standards that are finding some popularity worldwide. 
The former Soviet Union adopted the GOST, standards that were suitable for the pe- 
riod of industrialization of that country. 


In almost all cases, the main goal of a manufacturing project is to make a profit, the ex- 
ception being projects that have to do with the national security or prestige. Let us es- 
tablish a simplified model that illustrates the cash flow through the different activities 
associated with manufacturing so that we can see how to maximize the profit. As shown 
in Figure 1.3, the project starts by borrowing money from a bank to purchase machines 
and raw materials and to pay the salaries of the engineers and other employees. Next, the 
raw materials are converted into products, which are the output of the manufacturing do- 
main. Obviously, those products must be sold (through the marketing department) in 
order to get cash. This cash is, in turn, used to cover running costs, as well as required 
payment to the bank; any surplus money left is the profit. 

We can see in this model that the sequence of events forms a continuous cycle (i.e., 
a closed circuit). This cycle is usually referred to as the production turn. We can also re- 
alize the importance of marketing, which ensures the continuity of the cycle. If the prod- 

1.7 Product Life Cycle 


Initial money 

The production turn 


borrowed from bank 




Money to purchase 

raw materials and 

for the running cost 


+ profit 


ucts are not sold, the cycle is obviously interrupted. Moreover, we can see that maxi- 
mum profit is obtained through maximizing the profit per turn and/or increasing the 
number of turns per year (i.e., running the cycle faster). Evidently, these two conditions 
are fulfilled when products are manufactured in the easiest and least expensive way. 


It has been observed that all products, from the sales viewpoint, go through the same 
product life cycle, no matter how diverse or dissimilar they are. Whether the product 
is a new-model airplane or a coffeemaker, its sales follow a certain pattern or sequence 
from the time it is introduced in the market to the time it is no longer sold. The main 
difference between the cycles of these two products is the span or duration of the 
cycle, which always depends upon the nature and uses of the particular product. As we 
will see later when discussing concurrent engineering in Chapter 2, it is very important 
for the designer and the manufacturing engineer to fully understand that cycle in order 
to maximize the profits of the production plant. 

It is clear from Figure 1 .4 that the sales, as well as the rate of increase in sales, are ini- 
tially low during the introduction stage of the product life cycle. The reason is that the con- 
sumer is not aware of the performance and the unique characteristics of the product. 


The product life cycle 




Through television and newspaper advertisements and word-of-mouth communication, a 
growing number of consumers learn about the product and its capabilities. Meanwhile, 
the management works on improving the performance and eliminating the shortcomings 
through minor design modifications. It is also the time for some custom tailoring of the 
product for slightly different customer needs, in order to serve a wider variety of con- 
sumers. As a result, the customer acceptance is enhanced, and the sales accordingly in- 
crease at a remarkable rate during this stage, which is known as the growth stage. 
However, this trend does not continue forever, and, at a certain point, the sales level out. 
This is, in fact, the maturity stage of the life cycle. During this stage, the product is usu- 
ally faced with fierce competition, but the sales will continue to be stable if the manage- 
ment succeeds in reducing the cost of the product and/or developing new applications for 
it. The more successful the management is in achieving this goal, the longer the duration 
of the maturity stage will be. Finally, the decline stage begins, the sales fall at a noticeable 
rate, and the product is, at some point, completely abandoned. The decrease in the sales is 
usually due to newer and better products that are pumped into the market by competing 
manufacturers to serve some customer need. It can also be caused by diminishing need for 
the uses and applications of such a product. A clever management would start developing 
and marketing a new product (B) during the maturity stage of the previous one (A) so as 
to keep sales continuously high, as shown in Figure 1.5. 


The proper overlap of 
products' life cycles 






Every now and then, a new technology emerges as result of active research and devel- 
opment (R & D) and is then employed in the design and manufacture of several differ- 
ent products. It can, therefore, be stated that technology is concerned with the industrial 
and everyday applications of the results of the theoretical and experimental studies that 
are referred to as engineering. Examples of modern technologies include transistor, mi- 
crochip, and fiber optics. 

The relationship between the effectiveness or performance of a certain technology 
and the effort spent to date to achieve such performance is shown graphically in Fig- 

1.8 Technology Development Cycle 9 

ure 1.6. This graphical representation is known as the technology development cycle. 
It is also sometimes referred to as the S curve because of its shape. As can be seen in 
Figure 1.6, a lot of effort is required to produce a sensible level of performance at the 
early stage. Evidently, there is a lack of experimental experience since the techniques 
used are new. Next, the rate of improvement in performance becomes exponential, a 
trend that is observed with almost all kinds of human knowledge. At some point, how- 
ever, the rate of progress becomes linear because most ideas are in place; any further 
improvement comes as a result of refining the existing ideas rather than adding new 
ones. Again, as time passes, the technology begins to be "exhausted," and performance 
levels out. A "ceiling" is reached, above which the performance of the existing current 
technology cannot go because of social and/or technological considerations. 

An enlightened management of a manufacturing facility would allocate resources 
and devote effort to an active R&D program to come up with a new technology (B) as 
soon as it realized that the technology on which the products are based (A) was beginning 
to mature. The production activities would then be transferred to another S curve, with a 
higher ceiling for performance and greater possibilities, as shown in Figure 1 .7. Any delay 
in investing in R & D for developing new technology may result in creating a gap between 
the two curves (instead of continuity with the overlap shown in Figure 1.7), with the final 
outcome being to lose the market to competing companies that possess newer technology. 
In fact, the United States dominated the market of commercial airliners because compa- 
nies like Boeing and McDonnell Douglas knew exactly when to switch from propeller- 
driven airplanes to jet-propulsion commercial airliners. This is contrary to what some 
major computer companies did when they continued to develop and produce mainframe 
computers and did not recognize when to make the switch to personal computers. Current 
examples of technological discontinuity include the change from conventional telecom- 
munications cables to fiber optics for communication and information transfer. 


The technology 
development cycle 
(or S curve) 



1 Overview 


Transfer from one S 
curve to another 



An engineer is a problem solver who employs his or her scientific and empirical 
knowledge together with inventiveness and expert judgment to obtain solutions for 
problems arising from societal needs. These needs are usually satisfied by some phys- 
ical device, structure, or process. The creative process by which one or more of the 
fruits of the engineer's effort are obtained is referred to as design. It is, indeed, the core 
of engineering that provides the professional engineer with the chance of creating orig- 
inal designs and watching them become realities. The satisfaction that the engineer 
feels following the implementation of his or her design is the most rewarding experi- 
ence in the engineering profession. Because design is created to satisfy a societal need, 
there can be more than one way to achieve that goal. In other words, several designs 
can address the same problem. Which one is the best and most efficient design? Only 
time will tell because it is actually the one that would be favored by the customers 
and/or the society as a whole. 

Although there is no single standard sequence of steps to create a workable de- 
sign, E. V. Krick has outlined the procedure involved in the design process, and his 
work has gained widespread acceptance. Following is a discussion of the stages of the 
design process according to Krick. (See the references at the back of the book for more 
detailed information.) 

Problem Formulation 

As illustrated in Figure 1.8, problem formulation is the first stage of the design 
process. This phase comes as a result of recognizing a problem and involves defining 
that problem in a broad perspective without getting deep into the details. It is also at 
this stage that the engineer decides whether or not the problem at hand is worth solv- 
ing. In other words, this stage basically constitutes a feasibility study of the problem 
arising from a recognized need. The designer should, therefore, realize the importance 

1.9 The Design Process 



The design process 
(Adapted from Krick, An 
Introduction to 
Engineering and 
Engineering Design, 
2nd ed. New York: John 
Wiley, 1969) 

Recognition of 
a problem to 
be solved 


Problem formulation 
Problem analysis 






The process 
of design 

of this stage. Neglecting it may result in wasting money in an effort to solve a prob- 
lem that is not worth solving or wasting time on details that make it extremely difficult 
to get a broad view of the problem so as to select the appropriate path for solving it. 
The formulation of a problem can take any form that is convenient to the designer, al- 
though diagrammatic sketching (in particular, the black-box method) has proven to be 
a valuable tool. 

Problem Analysis 

The second stage involves much information gathering and processing in order to 
come up with a detailed definition of the problem. Such information may come from 
handbooks, from manufacturers' catalogs, leaflets, and brochures, as well as from per- 
sonal contacts. You are strongly advised to seek information wherever you can find it; 
workers at all levels of a company may have some key information that you can use. 
The end product should be a detailed analysis of the qualitative and quantitative char- 
acteristics of the input and output variables and constraints, as well as the criteria that 
will be used in selecting the best design. 

Search for Alternative Solutions 

In the third stage, the designer actively seeks alternative solutions. A good practice is to 
make a neat sketch for each preliminary design with some notes about its pros and cons. 
All sketches should be kept even after a different final design is selected so that if that 

12 1 Overview 

final design is abandoned for some reason, a designer does not have to start from the be- 
ginning again. It is also important to remember not to end the search for alternative so- 
lutions prematurely, before it is necessary or desirable to do so. Sometimes, a designer 
gets so involved with details of what he or she thinks is a good idea or solution that he 
or she will become preoccupied with these details, spending time on them instead of 
searching for other good solutions. Therefore, you are strongly advised to postpone 
working out the details until you have an appropriate number of viable solutions. 

It is, indeed, highly recommended to employ collaborative methods for enabling 
the mind to penetrate into domains that might otherwise remain unexplored. A typical 
example is the technique of brainstorming, where a few or several people assemble to 
produce a solution for a problem by creating an atmosphere that encourages everyone 
to contribute with whatever comes to mind. After the problem is explained, each mem- 
ber comes up with an idea that is, in turn, recorded on a blackboard, thus making all 
ideas evident to all team members. 

Decision Making 

The fourth stage involves the thorough weighing and judging of the different solutions 
with the aim of being able to choose the most appropriate one. That is, trade-offs have 
to be made during this stage. They can be achieved by establishing a decision matrix, 
as shown in Figure 1 .9. 

As can be seen in Figure 1 .9, each of the major design objectives is in a column, and 
each solution is allocated a row. Each solution is evaluated with regard to how it fulfills 
each of the design objectives and is, therefore, given a grade (on a scale of 1 to 10) in each 
column. Because the design objectives do not have the same weight, each grade must be 
multiplied by a factor representing the weight of the design function for which it was 
given. The total of all the products of multiplication is the score of that particular solution 
and can be considered as a true indication of how that solution fulfills the design objec- 
tives. As you can see, this technique provides a mechanism for rating the various solu- 
tions, thus eliminating most and giving further consideration to only a few. 

The chosen design is next subjected to a thorough analysis in order to optimize 
and refine it. Detailed calculations, whether manual or computational, are involved at 
this point. Both analytical and experimental modeling are also extensively employed 
as tools in refining the design. It is important, therefore, to now discuss modeling and 
simulation. A model can be defined as a simplified representation of a real-life situa- 
tion that aids in the analysis of an associated problem. There are many ways for clas- 
sifying and identifying models. For example, models can be descriptive, illustrating a 
real-world counterpart, or prescriptive, helping to predict the performance of the actual 
system. They can also be deterministic or probabilistic (used when making decisions 
under uncertainty). A simple example of a model is the free-body diagram used to de- 
termine the internal tensile force acting in a wire with a weight attached to its end. 
There are many computer tools (software) that are employed by designers to create 
models easily and quickly. Examples include geometric modeling and finite element 
analysis software packages. On the other hand, simulation can be defined as the 
process of experimenting with a model by subjecting it to various values of input pa- 

1.9 The Design Process 


■ 2 

H '« 


ui I 

a I 

— to 

u. < 


<B cu 


o o 

1- 1- 

| 8 


oo oS 

O </> 




CO 00 



■H T-i 










d d 





O +3 

_ O 


q q 

© 4S 



oo oJ 




•* h- 











d d 


i £ 




_ o 


o o 


oo oS 





o in 



«fr <fr 






in in 





d d 






o ■*■> 

_ o 


o o 



oo oS 


. 3 








c c 







CO en 




rameters and observing the output, which can be taken as an indication of the behav- 
ior of the real-world system under the tested conditions. As you can see, simulation 
can save a lot of time and effort that could be spent on experimental models and pro- 
totypes. This saving is particularly evident when computer simulation is employed. 
Still, simulation would not eliminate design iterations but rather would minimize their 
number. You are, therefore, urged to make use of these tools whenever possible. 


In the fifth and last stage of the design process, the designer organizes the material ob- 
tained in the previous stage and puts it in shape for presentation to his or her superiors. The 
output of this stage should include the attributes and performance characteristics of the re- 
fined design, given in sufficient detail. Accordingly, the designer must communicate all 
information in the form of clear and easy-to-understand documents. Documentation con- 
sists of carefully prepared, detailed, and dimensioned engineering drawings (i.e., assem- 
bly drawings and workshop drawings or blueprints), a written report, and possibly an 
iconic model. With the recent development in rapid prototyping techniques, a prototype 
can certainly be a good substitute for an iconic model. This approach has the advantage of 
revealing problems that may be encountered during manufacturing. 


The conventional procedure for product design, as illustrated in Figure 1.10, used to 
start with an analysis of the desired function, which usually dictated the form as well 
as the materials of the product to be made. The design (blueprint) was then sent to the 
manufacturing department, where the kind and sequence of production operations 
were determined mainly by the form and materials of the product. In fact, the old de- 
sign procedure had several disadvantages and shortcomings: 

FIGURE 1.10 

The old procedure for 
product design 






1.10 Product Design: The Concept of Design for Manufacturing 


FIGURE 1.11 

The new concept of a 
manufacturing system 
for achieving rational 
product designs 

1. In some cases, nice-looking designs were impossible to make; in many other cases, 
the designs had to be modified so that they could be manufactured. 

2. Preparing the design without considering the manufacturing process to be carried out 
and/or the machine tools available would sometimes result in a need for special- 
purpose, expensive machine tools. The final outcome was an increase in the produc- 
tion cost. 

3. When the required production volume was large, parts had to be specially designed 
to facilitate operations involved in mass production (such as assembly). 

4. A group of different products produced by the same manufacturing process has 
common geometric characteristics and features that are dictated by the manufactur- 
ing process employed (forgings, for example, have certain characteristic design fea- 
tures that are different from those of castings, extrusions, or stampings). Ignoring 
the method of manufacturing during the design phase would undermine these char- 
acteristic design features and thus result in impractical or faulty design. 

Because of these reasons and also because of the trend of integrating the activi- 
ties in a manufacturing corporation, the modern design procedure takes into consid- 
eration the method of manufacturing during the design phase. As can be seen in 
Figure 1.11, design, material, and manufacturing are three interactive, interrelated el- 
ements that form the manufacturing system, whose prime inputs are conceptual prod- 
ucts (and/or functions) and whose outputs are manufactured products. In fact, the 
barriers and borders between the design and manufacturing departments are fading 





1 Overview 

out and will eventually disappear. The tasks of the designer and those of the manu- 
facturing engineer are going to be combined and done by the same person. It is, 
therefore, the mission of this text to emphasize concepts like design for manufactur- 
ing and to promote the systems approach for product design. 

Review Questions 



1. What is the definition of manufacturing? 

2. Is there any relationship between the status of 
manufacturing in a nation and the standard of 
living of the people in that nation? Explain 

3. Explain the different approaches for obtaining a 
desired shape and give examples of some man- 
ufacturing processes that belong to each group. 

4. List the different types of production and ex- 
plain the main characteristics of each. Also 
mention some suitable applications for each 

5. Explain the meaning of the term tolerance. 

6. How do we scientifically describe the tightness 
or looseness of two mating parts? 

7. What concepts did Eli Whitney establish to en- 
sure trouble-free running of the mass produc- 
tion of multicomponent products? 

8. What is meant by the production turn? What 
role does marketing play in this cycle? 

9. Using the concept of production turn, how can 
we maximize the profits of a company by two 
different methods? 


Explain the stages involved in the life cycle of 
a product. 

What is the significance of the product life 
cycle during the phase of planning for the pro- 
duction of new products? 

What is the S curve? Explain an American suc- 
cess story in employing it. 

Give some examples of transfer from one tech- 
nology development curve to another. 

What are the stages involved in the design 
process? Explain each briefly. 

What is meant by trade-offs? How can these be 
achieved during the decision-making stage? 

Explain the old approach for product design. 
What are its disadvantages? 

17. Explain the concept of design for manufactur- 
ing. Why is it needed in modern industries? 







Chapter 2 



Concurrent engineering is a manufacturing philosophy that involves managing 
the product development process with the aim of getting new products with the 
highest quality at the best competitive price in the least time to the market. It 
has proven to be a key factor for the survival and, more importantly, for the 
prosperity of companies that are clever enough to adopt its methodology and 
tools (Motorola and Hewlett-Packard are good examples). In fact, many compa- 
nies can, in good faith, argue that they have been using this methodology in 
one form or another for some time — consider the efforts of corporations like 
Xerox, Hewlett-Packard, and Ford in the late 1970s to review and revise their 
product design practices versus those of their foreign competitors. An impor- 
tant milestone in the history of concurrent engineering is considered to be the 
report issued in December 1988 by the Defense Advanced Research Projects 
Agency (DARPA) as a result of a study to improve concurrency in the product de- 
sign process, a study that lasted more than five years. Many professionals in 
this field rightfully believe that the DARPA report is the true foundation for the 
concept of concurrent engineering. Many terms were and still are (though to a 
lesser degree) used to describe this methodology. Examples include team de- 
sign, simultaneous engineering, and integrated product development. The term 
concurrent engineering was first coined by the Institute for Defense Analysis 
(IDA), which also provided the following definition: 

Concurrent engineering is a systematic approach to integrated, concur- 
rent design of products and their related processes, including manufac- 
ture and support. This approach is intended to cause the developers, 


18 2 Concurrent Engineering 

from the outset, to consider all elements of the product life cycle from 
concept through disposal, including quality, cost, schedule, and user re- 

A good way to understand this new concept and what it means is to compare 
the product development process in the traditional engineering approach, which 
is usually referred to as serial or sequential engineering, with the one in the con- 
current engineering environment. In serial engineering, a team of qualified pro- 
fessional engineers designs a product without much interaction with or input from 
other departments within the corporation such as manufacturing, sales, or cus- 
tomer service. A model or prototype is then fabricated in the prototype workshop 
based on the documented design produced by the team. Note that the environ- 
ment in the model shop is ideal and is different from the real one on the shop floor 
during production. Weeks or even months after releasing the design, the testing 
department receives the model and carries out acceptance tests to make sure 
that the model conforms to the documented design and also meets the criteria 
established and agreed upon for the functioning and performance of the prod- 
uct. As you may expect, alterations and modification and/or revisions of the de- 
sign are needed in most cases as a result of the absence of inputs from other 
departments during the design process. As a consequence, revisions call for 
new designs that, in turn, require the fabrication and testing of new prototypes. 

This obviously time-consuming cycle may have to be repeated a few times 
in order to achieve the desired goals. Such a cycle prolongs the new product 
development process to various degrees. Depending upon the complexity of the 
product and the number of iterations, the delay can be excessive, thus causing 
damage to the marketing strategy and sales of the new product. It is, therefore, 
clear that the absence of communication between the different departments 
starting early in the initial phase of the design process and continuing through- 
out that process would result in a larger number of design iterations and de- 
lays in releasing the product to the market. On the contrary, in a concurrent 
engineering environment, all relevant departments, such as design, manufac- 
turing, R&D, and marketing, become involved and participate in the design 
process from its very beginning. This interaction reduces to a large extent or 
even eliminates design iterations, thus compressing the development cycle 
with the final outcome of reducing the time-to-market for a new product. The 

2.1 Reasons for Adopting Concurrent Engineering 19 

products can, therefore, be turned over at a much faster pace. Bearing in mind 
that the larger portion of profit occurs in the early part of the cycle for suc- 
cessful products, it would consequently be possible to allow new products to 
be retired at nearly their optimum profitability. Also, because customers are 
consulted (through the marketing and customer service departments) early in 
the product development process, the new product would most probably pene- 
trate the market easily because it would correctly meet customers' expecta- 
tions in terms of both function and quality. A good example is the case of the 
Boeing 777 jetliner. During the initial development stage, Boeing called repre- 
sentatives of its customers, including BOAC (the British-government-owned air- 
lines), although Britain is a part of the consortium building the air bus. By doing 
so, Boeing got the praise and the support of its customers all over the world 
(including BOAC) in the form of orders of the new product under development. 
More importantly, some serious modifications in the design were made in the 
very early phase, thus saving a lot of money and effort if they had not been 
done. The choice of the height of the wings above ground level is a clear ex- 
ample. Looking at the initial conceptual design, the customers realized that the 
wing height was too high to the extent that it would create difficulties during fu- 
eling and would require the use of a special fueling truck. Boeing was promptly 
advised to lower the level of the wings so that the currently used fueling trucks 
could be easily and successfully employed. 


Modern manufacturing industries are facing many challenges, such as global competi- 
tion and fast-changing consumer demands. These and other challenges call for the 
adoption of the concurrent engineering methodology. Following is a list of some of the 
challenges that can be successfully met by concurrent engineering: 

1. Increasing product complexities that prolong the product development process and 
make it more difficult to predict the impact of design decisions on the functionality 
and performance of the final product. 

2. Increasing global competitive pressures that result from the emerging concept of 
reengineering (which enabled many Asian countries to produce extremely cost- 
effective products because the cost of R & D in this case is almost zero). This def- 
initely creates the need for a cost-effective product development cycle. 

20 2 Concurrent Engineering 

The need for rapid response to fast-changing consumer demands. This phenomenon 
calls for the need to continuously listen to the "voice" of the consumer — one of the 
solid principles of concurrent engineering methodology. 

The need for shorter product life cycles. This phenomenon necessitates the intro- 
duction of new products to the market at a very high pace — something that can 
only be achieved by compressing the product development cycle. Consider the 
changes that the old-fashioned mechanical typewriter has undergone. Its life cycle 
was 20 to 30 years, which then decreased to 10 years for an electromechanical 
typewriter and finally to only 18 months for a word processor. This example clearly 
illustrates the need to solve the problem of time-to-market pressure. 

Large organizations with several departments working on developing numerous 
products at the same time. The amount of data exchanged between these depart- 
ments is extremely large, and unless properly managed in a rational manner, the 
flow and transfer of information is not fast or easy (i.e., a piece of information 
needed by a certain department may be passed to another one). The final outcome 
would certainly be delays in the process of product development and products that 
will not appear in the market at the scheduled time. 

New and innovative technologies emerging at a very high rate, thus causing the 
new products to be technologically obsolete within a short period of time. This phe- 
nomenon is particularly evident in the electronics industry, where the life cycle of 
a typical product is in months (it used to be years during the 1980s). As a conse- 
quence, new products must appear in the market at a very high pace — something 
that definitely necessitates a shorter product development cycle. 



The benefits of adopting concurrent engineering are numerous and positively affect the 
various activities in a corporation. Following is a summary of the important ones: 

1. Because the customer is consulted during the early product development process, 
the product will appear on the market with a high level of quality and will meet the 
expectations of the customer. The product introduction region (or start-up) of the 
product life cycle (see Figure 1.4) will be very short. The sales volume will, there- 
fore, attain maturity in a very short time. As a consequence, large revenues and 
profits would be achieved during the early phase of the product life cycle. This is 
very important because products become technologically obsolete very quickly as a 
result of fast-emerging, innovative technology. 

2. Adopting concurrent engineering will result in improved design quality, which is 
measured by the number of design changes made during the first six months after 
releasing a new product to the market. These design changes are extremely expen- 
sive unless caught early during the product development process. The lower the 

2.3 Factors Preventing the Adoption of Concurrent Engineering 21 

number of these changes, the more robust the design of the product is. In a concur- 
rent engineering environment, these design changes would evidently be minimal. 

3. Reduced product development and design times will result from listening to the 
voice of the customer and from transferring information between the various de- 
partments involved, including those downstream. This benefit is, in fact, a conse- 
quence of the reduction in the number of design iterations necessary to achieve 
optimum product design. Another factor is forsaking sequential methods of product 
development and replacing them with concurrent ones. 

4. Reduced production cost is a consequence of the preceding two benefits — namely, 
the reduction in the number of design changes after releasing the product and the 
reduction in the time of the product development process. Reduced cost, of course, 
provides a manufacturing company with a real advantage in meeting global com- 
petitive pressures. 

5. Elimination of delays when releasing the product to the market will guarantee a 
good market share for the new product. Also, it has been proven that delays in re- 
leasing the product will result in market loss of revenues. 

6. As a result of the reduced design time and effort, new products will be pumped into 
the market more frequently, which is, indeed, the advantage that Japanese au- 
tomakers have over their American counterparts. They can produce more different 
models, with smaller production volumes and shorter life cycles. 

7. Increased reliability and customer satisfaction will result from delivering the prod- 
uct "right the first time" and will also enhance the credibility of the manufacturing 


Now, if adopting concurrent engineering results in all the benefits just listed, why isn't 
it widely applied in all manufacturing corporations? The reason is that concurrent en- 
gineering is based on a manufacturing philosophy that requires breaking barriers be- 
tween departments and establishing multidisciplinary teams. This philosophy clearly 
contradicts the authoritative culture that is currently dominant in the industrial estab- 
lishment. The threat of loss of power and authority makes middle management and bu- 
reaucrats resistant to the idea of implementing concurrent engineering. There is also a 
natural resistance to anything new inherent in the minds of some people. Another fac- 
tor may be the need to build an excellent communication infrastructure for facilitating 
the flow of information throughout the product development cycle. Apparently, a lot of 
money and effort must be invested to create an adequate information system — some- 
thing that many companies either cannot afford or do not want to do. Yet another fac- 
tor that holds up the implementation of concurrent engineering is the temptation to 
come up with temporary short-run solutions to the problem of decreasing revenues, 

22 2 Concurrent Engineering 

without any regard to strategic planning and long-term goals. Examples include cutting 
the work force to increase profits (a faulty, shortsighted approach that would cause a 
company to lose trained employees who are needed to enable quick, on-schedule prod- 
uct delivery) and cutting the price without any basis (a solution that would inevitably 
eliminate or reduce the profits). 



The implementation of concurrent engineering is based on managing forces of change 
and using them as resources or tools in four arenas for efficient, fast, and economical 
product development. These arenas — organization, communication infrastructure, re- 
quirements, and product development — are the pillars on which the methodology of 
concurrent engineering rests. Let us examine each of them and see how each can be 


This arena includes the managers, product development teams, and support teams (i.e., 
the organization itself and the interactions of its components). The role of management 
is vital and includes not only motivating people to change their work habits to match 
the concurrent engineering environment but also ensuring unhindered exchange of in- 
formation between the different disciplines. In fact, management can be used as one of 
the forces of change or tools that guarantees continuous improvement of the product 
development process, as will be discussed later. 

Communication Infrastructure 

The communication infrastructure encompasses the hardware, software, and expertise 
that together form a system that allows the easy transfer of information relating to 
product development. As you may expect, when the product complexity increases, the 
number of disciplines involved also increases, as does the volume of information to be 
transferred. The system to be established must be capable of handling the type and 
amount of data necessary for the product development process. It must retrieve, eval- 
uate, and present the data in an organized format that is easy to understand and to use 
by team members and by management. In fact, many corporations learned the hard 
way that communication technologies are as important as design and manufacturing 
technologies for the success of a new product. The first task to handle, after purchas- 
ing the hardware, is to build a comprehensive and efficient database that has queries 
and that can be accessed by teams and by managers who are in charge of monitoring 
and evaluating the product development process. Electronic mail, interactive browsing 
capabilities, and other modern information transfer technology are also essential in 
order to eliminate the need for shoveling piles of documents and papers between the 
different departments and teams. 

2.4 The Four Pillars of Concurrent Engineering 23 


A broad (but accurate) description of the product requirements involves all product at- 
tributes that affect customer satisfaction. Consequently, customers' needs are considered 
when setting the specifications for the conceptual design. This consideration would, in- 
deed, ensure that the model or prototype meets the original goals from the start. This 
process of creating the conceptual specifications is extremely important and must be car- 
ried out rigorously. The more product attributes and constraints are initially specified, the 
fewer the problems associated with the final product design are, and the fewer the number 
of design changes or iterations will be. Of course, the conceptual design constraints 
should be defined very clearly and subjected to a continuous process of updating, evalua- 
tion, and validation. As is well known, constraints like government regulations, envi- 
ronmental laws, and industry and national standards are changing all the time. 
Continually updating these would certainly improve the product development process. 

Product Development 

In a concurrent engineering environment, the downstream processes of manufacturing, 
maintenance, customer service, sales, and so on, must be considered in the early design 
phase. This consideration, as previously mentioned, is a necessary condition for the im- 
plementation of concurrent engineering. The second important condition is the need for 
continuous improvement and optimization of the product development process. As a 
consequence of these two conditions, there is a continuous drive to develop, evaluate, 
and adopt new design methodologies. Concepts and approaches like design for manu- 
facturing (DFM) and design for assembly (DFA) have been popular in recent days and 
have proven to be valuable tools in adopting and implementing concurrent engineering. 
The reason is that their philosophy is based on using manufacturing (or assembly) as a 
design constraint, thus taking downstream processes like manufacturing and assembly 
into full consideration during the early design phase of the product. In other words, the 
success of these methodologies is linked to their philosophies being compatible to (or 
matching) that of concurrent engineering. 

Another part of the product development arena is what is sometimes referred to as 
the component libraries. The design (and manufacturing) attributes of the different 
components, whether standard ones that were purchased or parts that were previously 
designed and manufactured, are kept in a database. The availability of such a database 
to team members will speed up the design process by providing them with many al- 
ternatives to choose from and, more importantly, by freeing them from reinventing the 
wheel. Using previously tested components, maybe with very slight modification in 
the design, can save a lot of time and effort. Keeping a computerized database has the 
advantages of easy retrieval of designs and simultaneous availability to all team mem- 
bers. This topic will be covered later in the book in detail when we discuss group tech- 
nology and computer-aided process planning. 

Also part of the product development arena is the design process itself. We have 
already covered its stages and methodology in Chapter 1. Here, we want to emphasize 
again that good design has always been based on customer needs, which must be, in 


Concurrent Engineering 

turn, determined by listening to customer concerns. In fact, it was for this reason that 
the concept of quality function deployment (QFD) was developed in Japan's Kyoto 
Shipyard in the 1980s. By including QFD in the design process, teams do not lose 
touch with the customer, and, consequently, the designed products will meet cus- 
tomers' needs and expectations. Although QFD is beyond the scope of this text, a brief 
discussion will enlighten those engineers who must communicate with members in 
charge of QFD in a multidisciplinary team. QFD seeks to identify and evaluate the 
meaning of the word quality from the customer's point of view. The approach involves 
constructing a matrix that is quite similar to the design decision matrix covered in 
Chapter 1 (see Figure 1.9). This matrix is called the house of quality. The attributes, 
functions, and characteristics that the customer wants can be clearly identified and 
used as input constraints or requirements for the process of designing the product as 
previously mentioned. To repeat, remember that one of the goals of the design team is 
to have a decreasing number of design changes with increasing order of design stage. 
In the final design stage, if the appropriate methodology is adopted, the number of 
changes should be zero. 


Implementing concurrent engineering is based on managing some forces of change and 
using them as resources or tools to create the concurrent engineering environment. Fol- 
lowing are some of these forces of change. 


Technology has a very important role to play in each of the four arenas of concurrent 
engineering. It speeds up and optimizes the product development process, minimizes 
the number of design iterations, and facilitates communication and information trans- 
fer between the different teams and departments. Managers should, therefore, take full 
advantage of the most up-to-date technology available and avoid technology that is 
under development or obsolete. Unfortunately, the problem of acquiring up-to-date 
technology is far more complicated than it seems because of the extremely fast pace 
with which technology is advancing and the vast amount of different options available. 
Here are some tips that address the technology problem: 

1. Keep engineers abreast of the latest technological developments by providing them 
with technical journals and periodicals, sending them to international engineering 
conferences and exhibitions, and ensuring a continuous learning process through 
workshops and short courses offered on site. 

2. Ensure that the latest scientific findings are promptly employed in developing a com- 
pany's technology and, therefore, result in high-quality products. Companies should 
focus their efforts on applied research for developing products and processes and in- 
tegrate their R&D with design and development activities. (In fact, this is one of the 
reasons behind the success of several countries in the Far East.) 

2.5 Forces of Change 25 

3. Try to make use of the results of government-funded research and thus save time 
and money spent in obtaining similar findings. 

4. Definitely overcome the "not-invented-here" syndrome. Many industry people 
make the mistake of completely ignoring any technology that was not invented in 
their company. This syndrome leads to isolationism and, eventually, falling behind. 
It is very difficult for a company to fully develop technology starting from scratch. 
Acquiring technology by purchasing it, by establishing partnerships between com- 
panies, and by encouraging technology exchanges is worth exploring. 

It is important here to cast light on one of industry's most difficult problems in the 
United States — the bad effects of having advanced technology geared toward military 
applications. Although there is a wealth of technological information as a result of ac- 
tive R & D in military industries, it is classified and, therefore, not accessible for civilian in- 
dustries and commercial applications. A further obstacle is the difference in the product 
requirements in both cases. Although military criteria specify quality regardless of cost, 
civilian requirements call for both quality and cost. The picture is clear when you compare 
the performance and the cost of a nuclear bomber with those of a commercial jetliner. 


Management in a concurrent engineering environment takes its role from management 
in a traditional serial manufacturing corporation and goes far beyond it. A manager's 
role involves not only setting schedules and work expectations of engineers and assign- 
ing responsibilities but also managing changes and building an organizational structure 
that is flexible and can respond quickly to surprises and sudden changes in demands and 
requirements. You may have already concluded that managers must have a general but 
solid understanding of current and relevant technical issues in order to communicate ef- 
fectively with multidisciplinary teams. In fact, one of the most important tasks of mid- 
dle management in a concurrent engineering environment is the creation of those 
multidisciplinary teams in order to carry out the product development process. In sum- 
mary, the traditional role of management that is based on vertical chain of command, au- 
thoritative decision making, and the "carrot-stick" model of running corporations is 
diminishing continuously, especially in a company that adopts the concurrent engineer- 
ing philosophy. More emphasis is being placed on creating product development teams 
and facilitating information transfer and communication between them. 

Let us look more thoroughly at the process of establishing multidisciplinary teams. 
As you may expect, complex tasks are handled by breaking them into less complex ones 
that are, in turn, dealt with simultaneously but separately with different teams. Good man- 
agers should optimize the size of each team. A team that is too large or too small creates 
communication problems, is less efficient, and is more expensive. Attention must also be 
given to the talents and the quality of team members in terms of choosing the right person 
for the right job. When establishing the teams, the management focus must be to concur- 
rently execute tasks that are normally carried out sequentially and to integrate those ac- 
tivities that are concurrent. Consequently, an appropriate project-modeling tool must be 
used in order to identify and locate the patterns of information flow and interaction. There 


Concurrent Engineering 


The PERT chart 

are basically three approaches or tools — namely, the PERT chart, the GANTT chart, and 
the design structure matrix (DSM). 

The PERT chart, which is illustrated in Figure 2.1, is basically used to determine 
project duration and critical path. On the other hand, the GANTT chart displays the 
relative positioning of tasks on a time scale, as shown in Figure 2.2. In fact, some re- 
searchers believe that the DSM method is far better in displaying the connectivity of 
interacting tasks and improving the product development process. It also clearly illus- 
trates where the integration of tasks should take place. 

The DSM method and its modified version have been extensively used by Smith 
and Eppinger of the Sloan School of Management at the Massachusetts Institute of 
Technology (MIT). The basic method involves representing the relationship among 
project tasks in a matrix form and allows for different tasks to be coupled. As can be 
seen in Figure 2.3, each individual task is represented by a row and by a column of a 
square matrix; the need for information flow between two tasks is indicated by a check 
mark (x). Going horizontally across a task's row, the columns under which there are 
check marks are those from which information must be received in order to complete 
the given task. On the other hand, going vertically down a task's column, the check 


The GANTT chart 

Task A 

/ ' 








Task D 

2.5 Forces of Change 



Initial phase of the 
design structure matrix 



























































































































marks indicate the rows (tasks) that require output from the given column. The diago- 
nal elements are hatched because a task cannot be coupled with itself. Now, structur- 
ing the teams, usually referred to as product development teams (PDTs), can be 
accomplished by identifying highly coupled sets of tasks. First, the rows and columns 
must be rearranged so as to yield "batches" of check marks where a few tasks are cou- 
pled together and where the information of PDTs is most appropriate, as shown in Fig- 
ure 2.4. The process of swapping the rows and columns of the matrix requires 
experience and skill because it is based on trial and error. Nevertheless, the process is 
also amenable to computer manipulation and analysis. As you can see, however, the 
DSM model does not take into consideration the degree of dependence or coupling be- 
tween each two tasks. Recently, Smith and Eppinger replaced the check marks by 
numbers indicating the strength of dependence. The eigenvalue of such a matrix would 
reveal the highly coupled sets of tasks. 

After the teams are established, the next question is how to manage the product 
development project and ensure that it is on target to meet the previously agreed-upon 
milestones and deadlines. Again, PERT and GANTT charts can be employed to 


Concurrent Engineering 


Final phase of the 
design structure matrix 

Tasks A F G D E 























































































































achieve these goals. There are, however, some other methods for project updating that 
visually illustrate the project status in one integrated chart, thus ensuring that different 
PDTs meet their stated goals concurrently. The radar (spider) chart is a popular one. As 
can be seen in Figure 2.5, each task or area of activity is represented by a radial line 
and for a one-year period. Thus, one look at the chart is enough to see whether or not 
the separate goals in the different areas are met. In the ideal case, when all tasks are 
performed exactly according to the planned schedule, this chart will end up having 
concentric circles at the various time periods, as indicated by the dashed lines in Fig- 
ure 2.5. On the other hand, the bug chart is a plot of project expenditures versus prod- 
uct goals or milestones, which are indicated on the time axis as shown in Figure 2.6. 
The scales of both axes are adjusted so that a project that is on track is represented by 
a straight line making 45° with both axes. Although this chart has the advantages of in- 
dicating project-cost updating and individual milestones, it is sometimes misleading (a 
delay in purchasing supplies, for example, might seem or be interpreted as a positive 
indication). Further details are beyond the scope of this text, and interested readers are 
advised and encouraged to seek specialized books on the subject. 

2.5 Forces of Change 



The radar (spider) chart 

Area of activity 



The bug chart 



release date 
and expenditure 


Manufacturing release 
date target 

30 2 Concurrent Engineering 


There is an extremely large number of tools for handling various tasks in the different 
arenas. The selection of the right tool for the right job is, therefore, not easy. In addi- 
tion, most of these tools are undergoing a rapid, continuous, never-ending evolution. 
Tools that were new in the 1980s are now technically obsolete, which indicates the 
need to continually upgrade and replace a company's acquired tools. For example, 
structured analysis software tools manage information systems and present complex 
systems in a clear, easy-to-comprehend way. Instead of the old-fashioned written flow- 
charting procedures, this software provides graphical representation of any complex 
operation by a network of elements (each stands for a particular function) and by ar- 
rows indicating data flow and interaction between those elements. It also enables the 
elimination of redundant loops, thus making an operation more efficient. 

Another example is tools used for design automation. They find increasing appli- 
cation in manufacturing corporations, and it is anticipated that by the year 2000 about 
80 percent of all designs will be electronically done using these tools. Also, integration 
of these islands of automation (i.e., engineering departments in a manufacturing firm) 
is the trend in the 1990s, where local-area networks (LANs) are extensively used to 
transfer information from one department to another in a standardized format. 

The adoption of new tools creates the problem of changing the responsibilities and 
nature of the jobs of employees, who will need retraining and, sometimes, have to be 
swapped. Also, based on the preceding discussion, the level of automation used must 
be appropriate for the company, and automated PDTs must be integrated into a system. 

It is always important to remember that the forces of change discussed herein are 
just examples and that there can be other forces of change depending upon the nature of 
the manufacturing corporation and its production. Nevertheless, in all cases and regard- 
less of the tools used, the change from serial manufacturing to concurrent engineering 
must be well planned and managed so as to take place gradually and smoothly and must 
always be monitored by management. In fact, abrupt changes and employee dissatisfac- 
tion are two factors that can impede the implementation of concurrent engineering. 

> ~ 


Now that you understand the arenas of concurrent engineering and the forces of 
change, it is time to look at a case study indicating how concurrent engineering was 
successfully implemented and resulted in solving tough problems that were facing one 
of the world's largest manufacturers of automotive parts. The original report was given 
in a paper entitled "Nippondenso Co. Ltd: A Case Study of Strategic Product Design," 
authored by Daniel E. Whitney and presented at the Collaborative Engineering Con- 
ference held at MIT in October 1993. This paper contained a wealth of information 
and was based on seven personal visits by the author to Nippondenso Co. Ltd. during 
the period 1974 to 1991, as well as on interviews with the company's personnel and 
papers published by the engineering staff. The information has been rearranged here, 
however, so as to draw parallelism with the previously mentioned concurrent engi- 
neering model and its four arenas. 

2.6 A Success Story: Nippondenso 31 

Nippondenso Co. Ltd. is one of the world's largest manufacturers of automotive 
components, including air conditioners, heaters, relays, alternators, radiators, plus me- 
ters, diesel components, filters, controls, brake systems, and entertainment equipment. 
The company has 20 plants in 15 foreign countries in addition to 10 plants in Japan. In 
1991, almost 43,000 people were employed by the company worldwide. Nippondenso 
is the first-tier supplier to Toyota and other Japanese and foreign car companies, and 
its sales amounted to about $10 billion dollars in 1989. Now that you have a clear idea 
about the size of this company and the diversity of its products, let us see how they 
created a concurrent engineering environment. Following is the company's approach 
in each of the previously mentioned arenas. 


Nippondenso's philosophy is based on developing the product and the process for mak- 
ing it simultaneously. Consequently, multidisciplinary teams are formed through repre- 
sentation from various departments like production engineering, machines and tools, 
product design, and so on. Teams are small at the beginning but become larger as the 
project proceeds from the concept phase to the detailed-design phase. Top management 
promptly steps in when a crisis occurs and when a crucial decision needs to be made. Of 
course, a parallel-task approach is employed by overlapping some of the design steps. 


In addition to product performance specifications and production cost targets, there are 
other severe constraints dictated by the nature of the business of Nippondenso as a 
supplier to large auto manufacturers (i.e., the need to meet ordering patterns). The re- 
quirements of customers (like Toyota) include delivering extremely large amounts of 
products on a just-in-time (JIT) basis, with high variety and an unpredictable model 
mix that is always changing. A further constraint is to achieve all these goals with lit- 
tle or no changeover time. As you will see later, defining customer requirements 
helped Nippondenso to address the problems in a rational, thoughtful manner. 

Communication Infrastructure 

Nippondenso built an excellent system for information exchange. It is used to integrate 
the different machine tools throughout the plant through local- or wide-area networks 
that are, in turn, linked with the engineering departments dealing with computer con- 
trol, scheduling, quality monitoring, and the like. Any change in data by a team mem- 
ber is promptly made available to all members of other teams, thus breaking the 
barriers between departments and between teams. 

Product Development 

The two most important elements upon which Nippondenso's approach in the product 
development arena is based include developing the product and its manufacturing 
processes simultaneously and developing new product design methodologies. In fact, 
this approach is credited for enabling Nippondenso to meet customer requirements. 


Concurrent Engineering 

In order to meet the challenge of high production volume and high variety, the 
first step for Nippondenso was standardization after negotiating with customers and 
listening to their concerns. The next step was to design the products intelligently so as 
to achieve the desired flexibility during assembly, rather than employing complex and 
expensive production methods. In other words, their philosophy was based on using 
assembly rather than manufacturing to make different models. High variety was 
achieved by producing several versions of each component in the product and then as- 
sembling the appropriate component's versions into any desired model. Thus, an ex- 
tremely large number of combinations of component versions resulted in a large 
number of possible models. Moreover, this approach also ensured quick changeover 
from one model to another. 

At this point, the basic concept of concurrent engineering has been thoroughly 
demonstrated. Interested readers are encouraged to consult more specialized books on 
the subject (see the titles provided in the references at the back of the book). 

>w Questions 

1. Define the term concurrent engineering and 
elaborate on its meaning. 

2. In what way does a concurrent engineering en- 
vironment differ from that of serial manufactur- 

3. How did concurrent engineering come into 

4. What are the reasons for adopting concurrent 

5. Discuss three of these reasons in detail. 

6. List the benefits of adopting concurrent engi- 
neering and discuss three of them in detail. 

7. If concurrent engineering is so beneficial, why 
don't all manufacturing companies adopt it? 



What are the four pillars on which concurrent 
engineering rests? 

What is the difference between the role of man- 
agement in a concurrent engineering environ- 
ment and that role in conventional serial 

How does the product development process dif- 
fer in a concurrent engineering environment 
from that in conventional serial manufacturing? 

Explain why new concepts like DFM, DFA, 
and QFD are important and very useful when 
implementing concurrent engineering. 


Definition. The word casting is used both for the process and for the product. 
The process of casting is the manufacture of metallic objects (castings) by 
melting the metal, pouring it into a mold cavity, and allowing the molten metal 
to solidify as a casting whose shape is a reproduction of the mold cavity. This 
process is carried out in a foundry, where either ferrous (i.e., iron-base) or non- 
ferrous metals are cast. 

Casting processes have found widespread application, and the foundry in- 
dustry is considered to be the sixth largest in the United States because it pro- 
duces hundreds of intricately shaped parts of various sizes like plumbing 
fixtures, furnace parts, cylinder blocks of automobile and airplane engines, pis- 
tons, piston rings, machine tool beds and frames, wheels, and crankshafts. In 
fact, the foundry industry includes a variety of casting processes that can be 
classified in one of the following three ways: 

1. By the mold material and/or procedure of mold production 

2. By the method of filling the mold 

3. By the metal of the casting itself 

Historical Background. At the dawn of the metal age, human knowledge was not 
advanced enough to generate the high temperatures necessary for smelting 
metals. Therefore, because casting was not possible, metals were used as 
found or heated to a soft state and worked into shapes. The products of that era 
are exemplified by the copper pendant from Shanidar Cave (northeast of Iraq), 
which dates back to 9500 b.c. and which was shaped by hammering a piece of 


34 3 Casting and Foundry Work 

native metal and finishing with abrasives. Later, copper-smelting techniques 
were developed, and copper castings were produced in Mesopotamia as early as 
3000 b.c. The art of casting was then refined by the ancient Egyptians, who in- 
novated the "lost-wax" molding process. During the Bronze Age, foundry work 
flourished in China, where high-quality castings with intricate shapes could be 
produced. The Chinese developed certain bronze alloys and mastered the lost- 
wax process during the Shang dynasty. Later, that art found its way to Japan with 
the introduction of Buddhism in the sixth century. There were also some signifi- 
cant achievements in the West, where the Colossus of Rhodes, a statue of the 
sun god Helios weighing 360 tons, was considered to be one of the seven won- 
ders of the world. That bronze statue was cast in sections, which were assem- 
bled later, and stood 105 feet high at the entrance of the harbor of Rhodes. 

Although iron was known in Egypt as early as 4000 b.c, the development of 
cast iron was impossible because the high melting temperature needed was not 
achievable then and pottery vessels capable of containing molten iron were not 
available. The age of cast iron finally arrived in 1340 when a flow oven (a crude 
version of the blast furnace) was erected at Marche-Les-Dames in Belgium. It was 
capable of continuous volume production of molten iron. Ferrous foundry practice 
developed further with the invention of the cupola furnace by John Wilkenson in 
England. This was followed by the production of black-heart malleable iron in 
1826 by Seth Boyden and the development of metallography by Henry Sorby of 
England. The relationship between the properties and the microstructure of alloys 
became understood, and complete control of the casting process became feasi- 
ble based on this knowledge. Nevertheless, forming processes developed more 
rapidly than foundry practice because wrought alloys could better meet a wider 
range of applications. Nodular cast iron, which possesses both the castability of 
cast iron and the impact strength of steel, was introduced in 1948, thus paving 
the way for castings to compete more favorably with wrought alloys. 


Molds can be either permanent or nonpermanent. Permanent molds are made of steel, 
cast iron, and even graphite. They allow large numbers of castings to be produced suc- 
cessively without changing the mold. A nonpermanent mold is used for one pouring 

3.1 Classifications of Casting by Mold Material 35 

only. It is usually made of a silica sand mixture but sometimes of other refractory ma- 
terials like chromite and magnesite. 

Green Sand Molds 

Molding materials. Natural deposits taken from water or riverbeds are used as mold- 
ing materials for low-melting-point alloys. Thus, the material is called green sand, 
meaning unbaked or used as found. These deposits have the advantages of availability 
and low cost, and they provide smooth as-cast surfaces, especially for light, thin jobs. 
However, they contain 15 to 25 percent clay, which, in turn, includes some organic im- 
purities that markedly reduce the fusion temperatures of the natural sand mixture, 
lower the initial binding strength, and require a high moisture content (6 to 8 percent). 
Therefore, synthetic molding sand has been developed by mixing a cleaned pure silica 
sand base, in which grain structure and grain-size distribution are controlled, with up 
to 18 percent combined fireclay and bentonite and only about 3 percent moisture. Be- 
cause the amount of clay used as a binding material is minimal, synthetic molding sand 
has higher refractoriness, higher green (unbaked) strength, better permeability, and 
lower moisture content. The latter advantage results in the evolution of less steam dur- 
ing the casting process. Thus, control of the properties of the sand mixture is an im- 
portant condition for obtaining good castings. For this reason, a sand laboratory is 
usually attached to the foundry to determine the properties of molding sands prior to 
casting. Following are some important properties of a green sand mixture: 

1. Permeability. Permeability is the most important property of the molding sand and 
can be defined as the ability of the molding sand to allow gases to pass through. 
This property depends not only on the shape and size of the particles of the sand 
base but also on the amount of the clay binding material present in the mixture and 
on the moisture content. The permeability of molds is usually low when casting 
gray cast iron and high when casting steel. 

2. Green compression strength of a sand mold. Green strength is mainly due to the 
clay (or bentonite) and the moisture content, which both bind the sand particles to- 
gether. Molds must be strong enough not to collapse during handling and transfer 
and must also be capable of withstanding pressure and erosion forces during pour- 
ing of the molten metal. 

3. Moisture content. Moisture content is expressed as a percentage and is important 
because it affects other properties, such as the permeability and green strength. Ex- 
cessive moisture content can result in entrapped steam bubbles in the casting. 

4. Flowability. Flowability is the ability of sand to flow easily and fill the recesses and 
the fine details in the pattern. 

5. Refractoriness. Refractoriness is the resistance of the molding sand to elevated 
temperatures; that is, the sand particles must not melt, soften, or sinter when they 
come in contact with the molten metal during the casting process. Molding sands 
with poor refractoriness may burn when the molten metal is poured into the mold. 
Usually, sand molds should be able to withstand up to 3000°F (1650°C). 

36 3 Casting and Foundry Work 

Sand molding tools. Sand molds are made in flasks, which are bottomless containers. 
The function of a flask is to hold and reinforce the sand mold to allow handling and 
manipulation. A flask can be made of wood, sheet steel, or aluminum and consists of 
two parts: an upper half called the cope and a lower half called the drag. The standard 
flask is rectangular, although special shapes are also in use. For proper alignment of 
the two halves of the mold cavity when putting the cope onto the drag prior to casting, 
flasks are usually fitted with guide pins. When the required casting is high, a middle 
part, called the cheek, is added between the drag and the cope. Also, when a large 
product is to be cast, a pit in the ground is substituted for the drag; the process is then 
referred to as pit molding. 

Other sand molding tools can be divided into two main groups: 

1. Tools (such as molders, sand shovels, bench rammers, and the like) used for fill- 
ing the flask and ramming the sand 

2. Tools (such as draw screws, draw spikes, trowels, slicks, spoons, and lifters) used 
for releasing and withdrawing the pattern from the mold and for making required 
repairs on or putting finishing touches to the mold surfaces 

Patterns for sand molding. The mold cavity is the impression of a pattern, which is 
an approximate replica of the exterior of the desired casting. Permanent patterns 
(which are usually used with sand molding) can be made of softwood like pine, hard- 
wood like mahogany, plastics, or metals like aluminum, cast iron, or steel. They are 
made in special shops called pattern shops. Wood patterns must be made of dried or 
seasoned wood containing less than 10 percent moisture to avoid warping and dis- 
tortion of the pattern if the wood dries out. They should not absorb any moisture 
from the green molding sand. Thus, the surfaces of these patterns are painted and 
coated with a waterproof varnish. A single-piece wood pattern can be used for mak- 
ing 20 to 30 molds, a plastic pattern can be used for 20,000 molds, and a metal pat- 
tern can be used for up to 100,000 molds, depending upon the metal of the pattern. 
In fact, several types of permanent patterns are used in foundries. They include the 

1. Single or loose pattern. This pattern is actually a single copy of the desired cast- 
ing. Loose patterns are usually used when only a few castings are required or when 
prototype castings are produced. 

2. Gated patterns. These are patterns with gates in a runner system. They are used to 
eliminate the hand-cutting of gates. 

3. Match-plate patterns. Such patterns are used for large-quantity production of 
smaller castings, where machine molding is usually employed. The two halves of 
the pattern, with the line of separation conforming to the parting line, are perma- 
nently mounted on opposite sides of a wood or metal plate. This type of pattern al- 
ways incorporates the gating system as a part of the pattern. 

4. Cope-and-drag pattern plates. The function of this type of pattern is similar to that 
of the match-plate patterns. Such a pattern consists of the cope and drag parts of the 

3.1 Classifications of Casting by Mold Material 37 

pattern mounted on separate plates. It is particularly advantageous for preparing 
molds for large and medium castings, where the cope and drag parts of the mold are 
prepared on different molding machines. Therefore, accurate alignment of the two 
halves of the mold is necessary and is achieved through the use of guide and locat- 
ing pins and bushings in the flasks. 

In order for a pattern to be successfully employed in producing a casting having 
the desired dimensions, it must not be an exact replica of the part to be cast. A number 
of allowances must be made on the dimensions of the pattern: 

1. Pattern drafts. This is a taper of about 1 percent that is added to all surfaces per- 
pendicular to the parting line in order to facilitate removal of the pattern from the 
mold without ruining the surfaces of the cavity. Higher values of pattern draft are 
employed in the case of pockets or deep cavities. 

2. Shrinkage allowance. Because molten metals shrink during solidification and con- 
tract with further cooling to room temperature, linear dimensions of patterns must 
be made larger to compensate for that shrinkage and contraction. The value of the 
shrinkage allowance depends upon the metal to be cast and, to some extent, on 
the nature of the casting. The shrinkage allowance is usually taken as 1 percent for 
cast iron, 2 percent for steel, 1.5 percent for aluminum, 1.5 percent for magnesium, 
1 .6 percent for brass, and 2 percent for bronze. In order to eliminate the need for 
recalculating all the dimensions of a casting, pattern makers use a shrink rule. It is 
longer than the standard 1-foot rule; its length differs for the different metals of the 

3. Machine finish allowance. The dimensions on a casting are oversized to compen- 
sate for the layer of metal that is removed through subsequent machining to obtain 
better surface finish. 

4. Distortion allowance. Sometimes, intricately shaped or slender castings distort dur- 
ing solidification, even though reproduced from a defect-free pattern. In such cases, 
it is necessary to distort the pattern intentionally to obtain a casting with the desired 
shape and dimensions. 

Cores and core making. Cores are the parts of the molds that form desired internal 
cavities, recesses, or projections in castings. A core is usually made of the best quality 
of sand to have the shape of the desired cavity and is placed into position in the mold 
cavity. Figure 3.1 shows the pattern, mold, and core used for producing a short pipe 
with two flanges. As you can see, projections, called core prints, are added to both 
sides of the pattern to create impressions that allow the core to be supported and held 
at both ends. When the molten metal is poured, it flows around the core to fill the rest 
of the mold cavity. Cores are subjected to extremely severe conditions, and they must, 
therefore, possess very high resistance to erosion, exceptionally high strength, good 
permeability, good refractoriness, and adequate collapsibility (i.e., the rapid loss of 
strength after the core comes in contact with the molten metal). Because a core is sur- 
rounded by molten metal from all sides (except the far ends) during casting, gases have 
only a small area through which to escape. Therefore, good permeability is sometimes 


3 Casting and Foundry Work 


The pattern, mold, and 
core used for producing 
a short pipe 



(before removal 

from mold) 

assisted by providing special vent holes to allow gases to escape easily. Another re- 
quired characteristic of a core is the ability to shrink in volume under pressure without 
cracking or failure. The importance of this characteristic is obvious when you consider 
a casting that shrinks onto the core during solidification. If the core is made hard 
enough to resist the shrinkage of the casting, the latter would crack as a result of being 
hindered from shrinking. Figure 3.2 is a photograph of a sand core for an automotive 
cam tunnel. 

Core sand is a very pure, fine-grained silica sand that is mixed with different 
binders, depending upon the casting metal with which it is going to be used. The 
binder used with various castings includes fireclay, bentonite, and sodium silicate (in- 
organic binders), as well as oils (cottonseed or linseed oil), molasses, dexstrin, and 
polymeric resins (organic binders). 

Cores are usually made separately in core boxes, which involve cutting or ma- 
chining cavities into blocks of wood, metal, or plastic. The surfaces of each cavity 
must be very smooth, with ample taper or draft, to allow easy release of the green (un- 
baked) core. Sometimes, a release agent is applied to the surfaces of the cavity. Core 
sand is rammed into the cavity, and the excess is then struck off evenly with the top of 
the core box. Next, the green core is carefully rolled onto a metal plate and is baked in 
an oven. Intricate cores are made of separate pieces that are pasted together after bak- 
ing. Sometimes, cores are reinforced with annealed low-carbon steel wires or even 


Core for an automotive 
cam tunnel 

3.1 Classifications of Casting by Mold Material 



A simple core and its 
corresponding core box 




Left half of 
core box 

Right half of 
core box 


Vent hole 

cast-iron grids (in the case of large cores) to ensure coherence and stability. Figure 3.3 
illustrates a simple core and its corresponding core box. 

Large, round cores can be made by means of sweeps or templates, and drawing 
sweeps are employed to produce large cores that are not bodies of revolution. Various 
machines may also be employed in the core-making process. These include die ex- 
truders, jolt-squeeze machines, sandslingers, and pneumatic core blowers. Large cores 
are handled in the foundry and placed into the mold by means of a crane. 

Gating systems. Molds are filled with molten metal by means of channels, called 
gates, cut in the sand of the mold. Figure 3.4 illustrates a typical gating system, which 
includes a pouring basin, a down sprue, a sprue base(well), a runner, and in-gates. The 
design of the gating system is sometimes critical and should, therefore, be based on the 
theories of fluid mechanics, as well as the recommended industrial practice. In fact, a 
gating system must be designed so that the following are ensured: 

1. A continuous, uniform flow of molten metal into the mold cavity must be pro- 
vided without any turbulence. 


A typical gating system 

Down sprue 


Sprue base (well) 


40 3 Casting and Foundry Work 

2. A reservoir of molten metal that feeds the casting to compensate for the shrinkage 
during solidification must be maintained. 

3. The molten metal stream must be prevented from separating from the wall of the 

Let us now break down the gating system into its components and discuss the de- 
sign of each of them. The pouring basin is designed to reduce turbulence. The molten 
metal from the ladle must be poured into the basin at the side that does not have the 
tapered sprue hole. The hole should have a projection with a generous radius around 
it, as shown in Figure 3.4, in order to eliminate turbulence as the molten metal enters 
the sprue. Next, the down sprue should be made tapered (its cross-sectional area 
should decrease when going downward) to prevent the stream of molten metal from 
separating from its walls, which may occur because the stream gains velocity as it trav- 
els downward and, therefore, contracts (remember the continuity equation in fluid me- 
chanics, A x V, = A 2 V 2 ). The important and critical element of the gating system is the 
in-gate, whose dimensions affect those of all other elements. Sometimes, the cross- 
sectional area of the in-gate is reduced in the zone adjacent to the sprue base to create 
a "choke area" that is used mainly to control the flow of molten metal and, conse- 
quently, the pouring time. In other words, it serves to ensure that the rate of molten- 
metal flow into the mold cavity is not higher than that delivered by the ladle and, 
therefore, keeps the gating system full of metal throughout the casting operation. 
On the other hand, gas contamination, slag inclusions, and the like should be elim- 
inated by maintaining laminar flow. Accordingly, the Reynolds number (R„) should 
be checked throughout the gating system (remember that the flow is laminar when 
/?„ < 2000). Use must also be made of Bernoulli's equation to calculate the velocity of 
flow at any cross section of the gating system. 

In some cases, when casting heavy sections or high-shrinkage alloys, extra reser- 
voirs of molten metal are needed to compensate continually for the shrinkage of the 
casting during solidification. These molten-metal reservoirs are called risers and are 
attached to the casting at appropriate locations to control the solidification process. The 
locations of the feeding system and the risers should be determined based on the phe- 
nomenon that sections most distant from those molten-metal reservoirs solidify first. 
Risers are molded into the cope half of the mold to ensure gravity feeding of the 
molten metal and are usually open to the top surface of the mold. In that case, they are 
referred to as open risers. When they are not open to the top of the mold, they are then 
called blind risers. Risers can also be classified as top risers and side risers, depend- 
ing upon their location with respect to the casting. 

Another way to achieve directional solidification is the use of chills; these involve 
inserts of steel, cast iron, or copper that act as a "heat sink" to increase the solidifica- 
tion rate of the metal at appropriate regions of the casting. Depending upon the shape 
of the casting, chills can be external or internal. 

Molding processes. Green sand can be molded by employing a variety of processes, 
including some that are carried out both by hand and with molding machines. Follow- 
ing is a brief survey of the different green sand molding methods: 

3.1 Classifications of Casting by Mold Material 


Flask molding. Flask molding is the most widely used process in both hand- and 
machine-molding practices. Figure 3.5 illustrates the procedure for simple hand- 
molding using a single (loose) pattern. First, the lower half of the pattern is placed 
on a molding board and surrounded by the drag. The drag is then filled with sand 
(using a shovel) and rammed very firmly. Ventilation holes are made using a steel 
wire, but these should not reach the pattern. The drag is turned upside down to 
bring the parting plane up so that it can be dusted. Next, the other half of the pat- 
tern is placed in position to match the lower half, and the cope is located around it, 
with the eyes of the cope fitted to the pins of the drag. Sand is shoveled into the 
cope and rammed firmly, after using a sprue pin to provide for the feeding passage. 
Ventilation holes are made in the cope part of the mold in the same way they were 
made in the other half. The pouring basin is cut around the head of the sprue pin 
using a trowel, and the sprue pin is pulled out of the cope. The cope is then care- 
fully lifted off the drag and turned so that the parting plane is upward. The two 
halves of the pattern are removed from both the cope and the drag. The runner 
and/or gate are cut from the mold cavity to the sprue in the drag part of the mold. 
Then, any damages are repaired by slightly wetting the location and using a slick. 
The cope is then carefully placed on the drag to assemble the two halves of the 


The procedure of flask 
molding using a single 
(loose) pattern 







42 3 Casting and Foundry Work 

mold. Finally, the cope and the drag are fastened together by means of shackles or 
bolts to prevent the pressure created by the molten metal (after pouring) from sep- 
arating them. Enough weight can be placed on the cope as an alternative to using 
shackles or bolts. In fact, the pressure of the molten metal after casting can be given 
by the following equation: 

p = wx h (3.1) 

where: p is the pressure 

w is the specific weight of the molten metal 
h is the height of the cope 

The force that is trying to separate the two halves of the mold can, therefore, be given 
by the following equation: 

F = p x A (3.2) 

where: F is the force 

A is the cross-sectional area of the casting (including the runner, gates, 
etc.) at the parting line 

2. Stack molding. Stack molding is best suited for producing a large number of small, 
light castings while using a limited amount of floor space in the foundry. As can be 
seen in Figure 3.6a and b, there are two types of stack molding: upright and 
stepped. In upright stack molding, 10 to 12 flask sections are stacked up. They all 
have a common sprue that is employed in feeding all cavities. The drag cavity is al- 
ways molded in the upper surface of the flask section, whereas the cope cavity is 
molded in the lower surface. In stepped stack molding, each section has its own 
sprue and is, therefore, offset from the one under it to provide for the pouring basin. 
In this case, each mold is cast separately. 

3. Sweep molding. Sweep molding is used to form the surfaces of the mold cavity 
when a large-size casting must be produced without the time and expenses involved 
in making a pattern. A sweep that can be rotated around an axis is used for produc- 
ing a surface of revolution, contrary to a drawing sweep, which is pushed axially 
while being guided by a frame to produce a surface having a constant section along 
its length (see discussion of the extrusion process in Chapter 5). 

4. Pit molding. Pit molding is usually employed for producing a single piece of a large 
casting when it would be difficult to handle patterns of that size in flasks. Molding 
is done in specially prepared pits in the ground of the foundry. The bottom of the 
pit is often covered with a layer of coke that is 2 to 3 inches (50 to 75 mm) thick. 
Then, a layer of sand is rammed onto the coke to act as a "bed" for the mold. Vent 
pipes connect the coke layer to the ground surface. Molding is carried out as usual, 
and molds are almost always dried before pouring the molten metal. This drying is 
achieved by means of a portable mold drier. A cope that is also dried is then placed 
on the pit, and a suitable weight or a group of weights are located on the cope to 
prevent it from floating when the molten metal is poured. 

3.1 Classifications of Casting by Mold Material 



The two types of stack 
molding: (a) upright; 
(b) stepped 


Molding machines. The employment of molding machines results in an increase in 
the production rate, a marked increase in productivity, and a higher and more con- 
sistent quality of molds. The function of these machines is to pack the sand onto the 
pattern and draw the pattern out from the mold. There are several types of molding 
machines, each with a different way of packing the sand to form the mold. The main 
types include squeezers, jolt machines, and sandslingers. There are also some ma- 
chines, such as jolt-squeeze machines, that employ a combination of the working 


3 Casting and Foundry Work 

principles of two of the main types. Following is a brief discussion of the three main 
types of molding machines (see Figure 3.7): 

1. Squeezers. Figure 3.7a illustrates the working principle of the squeezer type of 
molding machine. The pattern plate is clamped on the machine table, and a flask is 
put into position. A sand frame is placed on the flask, and both are then filled with 


Molding machines: 

(a) squeezer; 

(b) jolt machine; 

(c) sandslinger 




3.1 Classifications of Casting by Mold Material 45 

sand from a hopper. Next, the machine table travels upward to squeeze the sand be- 
tween the pattern plate and a stationary head. The squeeze head enters into the sand 
frame and compacts the sand so that it is level with the edge of the flask. 

2. Jolt machines. Figure 3.7b illustrates the working principle of the jolt type of 
molding machine. As can be seen, compressed air is admitted through the hose to a 
pressure cylinder to lift the plunger (and the flask, which is full of sand) up to a cer- 
tain height, where the side hole is uncovered to exhaust the compressed air. The 
plunger then falls down and strikes the stationary guiding cylinder. The shock wave 
resulting from each of the successive impacts contributes to packing the molding 
sand in the flask. 

3. Sandslingers. Figure 3.7c shows a sandslinger. This type of machine is employed 
in molding sand in flasks of any size, whether for individual or mass production of 
molds. Sandslingers are characterized by their high output, which amounts to 2500 
cubic feet (more than 60 cubic meters) per hour. As can be seen, molding sand is 
fed into a housing containing an impeller that rotates rapidly around a horizontal 
axis. Sand particles are picked up by the rotating blades and thrown at a high speed 
through an opening onto the pattern, which is located in the flask. 

No matter what type of molding machine is used, special machines are employed 
to draw the pattern out of the mold. Basically, these machines achieve that goal by 
turning the flask (together with the pattern) upside down and then lifting the pattern 
out of the mold. Examples of these machines include roll-over molding machines and 
rock-over pattern-draw machines. 

Sand conditioning. The molding sand, whether new or used, must be conditioned be- 
fore being used. When used sand is to be recycled, lumps should be crushed and then 
metal granules or small parts removed (a magnetic field is employed in a ferrous 
foundry). Next, sand (new or recycled) and all other molding constituents must be 
screened in shakers, rotary screens, or vibrating screens. Molding materials are then 
thoroughly mixed in order to obtain a completely homogeneous green sand mixture. 
The more uniform the distribution, the better the molding properties (like permeability 
and green strength) of the sand mixture will be. 

Mixing is carried out in either continuous-screw mixers or vertical-wheel mullers. 
The mixers mix the molding materials by means of two large screws or worm gears; 
the mullers are usually used for batch-type mixing. A typical muller is illustrated in 
Figure 3.8. It consists primarily of a pan in which two wheels rotate about their own 
horizontal axis as well as about a stationary vertical shaft. Centrifugal mullers are also 
in use, especially for high production rates. 

Dry Sand Molds 

As previously mentioned, green sand molds contain up to 8 percent water, depending 
upon the kind and percentage of the binding material. Therefore, this type of mold can 
be used only for small castings with thin walls; large castings with thick walls would 
heat the mold, resulting in vaporization of water, which would, in turn, lead to bubbles 


3 Casting and Foundry Work 


A muller for sand 

Plow blade 


(conditioned sand) 


in the castings. For this reason, molds for large castings should be dried after they 
are made in the same way as green sand molds. The drying operation is carried out 
in ovens at temperatures ranging from 300°F to 650°F (150°C to 350°C) for 8 up to 
48 hours, depending upon the kind and amount of binder used. 

Core-Sand Molds 

When the mold is too big to fit in an oven, molds are made by assembling several 
pieces of sand cores. Consequently, patterns are not required, and core boxes are em- 
ployed instead to make the different sand cores necessary for constructing the mold. 
Because core-sand mixtures (which have superior molding properties) are used, very 
good quality and dimensional accuracy of the castings are obtained. 

Cement-Bonded Sand Molds 

A mixture of silica sand containing 8 to 12 percent cement and 4 to 6 percent water is 
used. When making the mold, the cement-bonded sand mixture must be allowed to 
harden first before the pattern is withdrawn. The obtained mold is then allowed to cure 
for about 3 to 5 days. Large castings with intricate shapes, accurate dimensions, and 
smooth surfaces are usually produced in this way, the only shortcoming being the long 
time required for the molding process. 

Carbon Dioxide Process for Molding 

Silica sand is mixed with a binder involving a solution of sodium silicate (water glass) 
amounting to 6 percent. After the mold is rammed, carbon dioxide is blown through 
the sand mixture. As a result, the gel of silica binds the sand grains together, and no 

3.1 Classifications of Casting by Mold Material 47 

drying is needed. Because the molds are allowed to harden while the pattern is in po- 
sition, high dimensional accuracy of molds is obtained. 

Plaster Molds 

A plaster mold is appropriate for casting silver, gold, magnesium, copper, and alu- 
minum alloys. The molding material is a mixture of fine silica sand, asbestos, and plas- 
ter of paris as a binder. Water is added to the mixture until a creamy slurry is obtained, 
which is then employed in molding. The drying process should be very slow to avoid 
cracking of the mold. 

Loam Molds 

The loam mold is used for very large jobs. The basic shape of the desired mold is con- 
structed with bricks and mortar (just like a brick house). A loam mixture is then used 
as a molding material to obtain the desired fine details of mold. Templates, sweeps, 
and the like are employed in the molding process. The loam mixture used in molding 
consists of 50 percent or more of loam, with the rest being mainly silica sand. Loam 
molds must be thoroughly dried before pouring the molten metal. 

Shell Molds 

In shell molding, a thin mold is made around a heated-metal pattern plate. The mold- 
ing material is a mixture of dry, fine silica sand (with a very low clay content) and 3 
to 8 percent of a thermosetting resin like phenolformaldehyde or ureaformaldehyde. 
Conventional dry-mixing techniques are used for obtaining the molding mixture. Spe- 
cially prepared resin-coated sands are also used. 

When the molding mixture drops onto the pattern plate, which is heated to a tem- 
perature of 350°F to 700°F (180°C to 375°C), a shell about 1/4 inch (6 mm) thick is 
formed. In order to cure the shell completely, it must be heated at 450°F to 650°F 
(230°C to 350°C) for about 1 to 3 minutes. The shell is then released from the pattern 
plate by ejector pins. To prevent sticking of the baked shell, sometimes called the bis- 
cuit, to the pattern plate, a silicone release agent is applied to the plate before the mold- 
ing mixture drops onto it. Figure 3.9 is a photograph of a pattern of a crankshaft used 
in shell molding. 

Shell molding is suitable for mass production of thin-walled, gray cast-iron (and 
aluminum-alloy) castings having a maximum weight between 35 and 45 pounds (15 
and 20 kg). However, castings weighing up to 1000 pounds (450 kg) can be made by 
employing shell molding on an individual basis. The advantages of shell molding in- 
clude good surface finish, few restrictions on casting design, and the fact that this 
process renders itself suitable for automation. 

Ceramic Molds 

In the ceramic molding process, the molding material is actually a slurry consisting 
of refractory grains, ceramic binder, water, alcohol, and an agent to adjust the pH 
value (see discussion of slurry casting in Chapter 7). The slurry is poured around the 


3 Casting and Foundry Work 


A pattern of a 
crankshaft used in shell 

m\ If rlt-LJ I II W I ^H 

permanent (reusable) pattern and is allowed to harden when the pattern is withdrawn. 
Next, the mold is left to dry for some time and then is fired to gain strength. In fact, 
ceramic molds are usually preheated before pouring the molten metal. For this rea- 
son, they are suitable for casting high-pouring-temperature alloys. Excellent surface 
finish and very close tolerances of the castings are among the advantages of this 
molding process and lead to the elimination of the machining operations that are usu- 
ally performed on castings. Therefore, ceramic molds are certainly advantageous 
when casting precious or difficult-to-machine metals as well as for making castings 
with great shape intricacy. 

Precision Molds (Investment Casting) 

Precision molding is used when castings with intricate shapes, good dimensional ac- 
curacy, and very smooth surfaces are required. The process is especially advantageous 
for high-melting-point alloys as well as for difficult-to-machine metals. It is also most 
suitable for producing small castings having intricate shapes, such as the group of in- 
vestment castings shown in Figure 3.10. A nonpermanent pattern that is usually made 
of wax must be prepared for each casting. Therefore, the process is sometimes referred 
to as the lost-wax process. Generally, the precision molding process involves the fol- 
lowing steps (see Figure 3.11): 

1. A heat-disposable pattern, together with its gating system, is prepared by injecting 
wax or plastic into a die cavity. 

2. A pattern assembly that is composed of a number of identical patterns is made. Pat- 
terns are attached to a runner bar made of wax or plastic in much the same manner 
as leaves are attached to branches. A ceramic pouring cup is also attached to the top 
of the pattern assembly, which is sometimes referred to as the tree or cluster (see 
Figure 3.11a). 

3.1 Classifications of Casting by Mold Material 


FIGURE 3.10 

A group of investment 
castings (Courtesy of 
Fansteel ESCAST, 
Addison, Illinois) 



\ \ 

3. The tree is then invested by separately dipping it into a ceramic slurry that is com- 
posed of silica flour suspended in a solution of ethyl silicate and sprinkling it with 
very fine silica sand. A self-supporting ceramic shell mold about 1/4 inch (6 mm) 
thick is formed all around the wax assembly (see Figure 3.1 lb). Alternatively, a thin 
ceramic precoating is obtained, and then the cluster is placed in a flask and a thick 
slurry is poured around it as a backup material. 

4. The pattern assembly is then baked in an oven or a steam autoclave to melt out the 
wax (or plastic). Therefore, the dimensions of the mold cavity precisely match 
those of the desired product. 

5. The resulting shell mold is fired at a temperature ranging from 1600°F to 1800°F 
(900°C to 1000°C) to eliminate all traces of wax and to gain reasonable strength. 

6. The molten metal is poured into the mold while the mold is still hot, and a cluster 
of castings is obtained (see Figure 3.11c). 

Today, the lost-wax process is used in manufacturing large objects like cylinder 
heads and camshafts. The modern process, which is known as the lost-foam method, 
involves employing a styrofoam replica of the finished product, which is then coated 
with a refractory material and located in a box, where sand is molded around it by vi- 
bratory compaction. When the molten metal is finally poured into the mold, the styro- 
foam vaporizes, allowing the molten metal to replace it. 


3 Casting and Foundry Work 

FIGURE 3.11 

Steps involved in investment casting: (a) a cluster of wax patterns; (b) a cluster of ceramic shells; 
(c) a cluster of castings (Courtesy of Fansteel ESCAST, Addison, Illinois) 

Graphite Molds 

Graphite is used in making molds to receive alloys (such as titanium) that can be 
poured only into inert molds. The casting process must be performed in a vacuum to 
eliminate any possibility of contaminating the metal. Graphite molds can be made ei- 
ther by machining a block of graphite to create the desired mold cavity or by com- 
pacting a graphite-base aggregate around the pattern and then sintering the obtained 
mold at a temperature of 18()0°F to 2000°F (1000°C to 1120°C) in a reducing atmos- 
phere (see Chapter 7). In fact, graphite mold liners have found widespread industrial 
application in the centrifugal casting of brass and bronze. 

3.1 Classifications of Casting by Mold Material 


Permanent Molds 

A permanent mold can be used repeatedly for producing castings of the same form and 
dimensions. Permanent molds are usually made of steel or gray cast iron. Figure 3.12a 
and b shows a permanent mold made of alloy steel for molding a cylinder block. Each 
mold is generally made of two or more pieces that are assembled together by fitting 
and clamping. Although the different parts of the mold can be cast to their rough con- 
tours, subsequent machining and finishing operations are necessary to eliminate the 
possibility of the casting's sticking to the mold. Simple cores made of metal are fre- 
quently used. When complex cores are required, they are usually made of sand or plas- 
ter, and the mold is said to be semipermanent. 

Different metals and alloys can successfully be cast in permanent molds. They in- 
clude aluminum alloys, magnesium alloys, zinc alloys, lead, copper alloys, and cast 

FIGURE 3.12 

A permanent mold 
made of alloy steel for 
casting a cylinder block: 
(a) drag; (b) cope 

52 3 Casting and Foundry Work 

irons. It is obvious that the mold should be preheated to an appropriate temperature 
prior to casting. In fact, the operating temperature of the mold, which depends upon 
the metal to be cast, is a very important factor in successful permanent-mold casting. 

Based on the preceding discussion, we can expect the mold life to be dependent 
upon a number of interrelated factors, including the mold material, the metal to be cast, 
and the operating temperature of the mold. Nevertheless, it can be stated that the life 
of a permanent mold is about 100,000 pourings or more when casting zinc, magne- 
sium, or aluminum alloys and not more than 20,000 pourings for copper alloys and 
cast irons. However, mold life can be extended by spraying the surface of the mold 
cavity with colloidal refractories suspended in liquids. 

The advantages of permanent-mold casting include substantial increases in pro- 
ductivity (a mold does not have to be made for each casting), close tolerances, supe- 
rior surface finish, and improved mechanical properties of the castings. A further 
advantage is the noticeable reduction in the percentage of rejects when compared with 
the conventional sand-casting processes. Nevertheless, the process is economically 
feasible for mass production only. There is also a limitation on the size of parts pro- 
duced by permanent-mold casting. A further limitation is that not all alloys are suited 
to this process. 



For all types of molds that we have discussed, the molten metal is almost always fed 
into the mold only by the action of gravity. Therefore, the casting process is referred 
to as gravity casting. There are, however, other special ways of pouring or feeding the 
molten metal into the desired cavities. These casting methods are generally aimed at 
forcing the molten metal to flow and fill the fine details of the mold cavity while elim- 
inating the internal defects experienced in conventional gravity casting processes. Fol- 
lowing is a survey of the commonly used special casting processes. 

Die Casting 

Die casting involves forcing the molten metal into the cavity of a steel mold, called a 
die, under very high pressure (1000 to 30,000 pounds per square inch, or about 70 to 
2000 times the atmospheric pressure). In fact, this characteristic is the major difference 
between die casting and permanent-mold casting, where the molten metal is fed into 
the mold either by gravity or at low pressures. Die casting may be classified according 
to the type of machine used. The two principal types are hot-chamber machines and 
cold-chamber machines. 

Hot-chamber machines. The main components of the hot-chamber die casting ma- 
chine include a steel pot filled with the molten metal to be cast and a pumping system 
that consists of a pressure cylinder, a plunger, a gooseneck passage, and a nozzle. With 
the plunger in the up position, as shown in Figure 3.13a, the molten metal flows by 
gravity through the intake ports into the submerged hot chamber. When the plunger is 

3.2 Classifications of Casting by Method of Filling the Mold 


FIGURE 3.13 

The hot-chamber die casting method: (a) filling the chamber; (b) metal forced into the die cavity 


Hot pot 




die ^ f Stationary 

die half 



pushed downward by the power cylinder (not shown in the figure), it shuts off the in- 
take port. Then, with further downward movement, the molten metal is forced through 
the gooseneck passage and the nozzle into the die cavity, as shown in Figure 3.13b. 
Pressures ranging from 700 to 2000 pounds per square inch (50 to 150 atmospheres) 
are quite common to guarantee complete filling of the die cavity. After the cavity is full 
of molten metal, the pressure is maintained for a preset dwell time to allow the casting 
to solidify completely. Next, the two halves of the die are pushed apart, and the cast- 
ing is knocked out by means of ejector pins. The die cavity is then cleaned and lubri- 
cated before the cycle is repeated. 

The advantages of hot-chamber die casting are numerous. They include high pro- 
duction rates (especially when multicavity dies are used), improved productivity, su- 
perior surface finish, very close tolerances, and the ability to produce intricate shapes 
with thin walls. Nevertheless, the process has some limitations. For instance, only low- 
melting-point alloys (such as zinc, tin, lead, and the like) can be cast because the com- 
ponents of the pumping system are in direct contact with the molten metal throughout 
the process. Also, die casting is usually only suitable for producing small castings that 
weigh less than 10 pounds (4.5 kg). 

Cold-chamber machines. In the cold-chamber die casting machine, the molten-metal 
reservoir is separate from the casting machine, and just enough for one shot of 
molten metal is ladled every stroke. Consequently, the relatively short exposure of 
the shot chamber and the plunger to the molten metal allows die casting of alu- 
minum, magnesium, brass, and other alloys having relatively high melting points. In 
the sequence of operations in cold-chamber die casting, the molten metal is first la- 
dled through the pouring hole of the shot chamber while the two halves of the die are 
closed and locked together, as shown in Figure 3.14. Next, the plunger moves for- 
ward to close off the pouring hole and then forces the molten metal into the die cav- 
ity. Pressures in the shot chamber may go over 30,000 pounds per square inch (2000 


3 Casting and Foundry Work 

FIGURE 3.14 

The cold-chamber die 
casting method 




atmospheres). After the casting has solidified, the two halves of the die are opened, 
and the casting, together with the gate and the slug of excess metal, are ejected from 
the die. 

It is not difficult to see that large parts weighing 50 pounds (23 kg) can be pro- 
duced by cold-chamber die casting. The process is very successful when casting alu- 
minum alloys, copper alloys, and high-temperature aluminum-zinc alloys. However, 
this process has a longer cycle time when compared with hot-chamber die casting. A 
further disadvantage is the need for an auxiliary system for pouring the molten metal. 
It is mainly for this reason that vertical cold-chamber machines were developed. As 
can be seen in Figure 3.15, such a machine has a transfer tube that is submerged into 
molten metal. It is fed into the shot chamber by connecting the die cavity to a vacuum 
tank by means of a special valve. The molten metal is forced into the die cavity when 
the plunger moves upward. 

Centrifugal Casting 

Centrifugal casting refers to a group of processes in which the forces used to distrib- 
ute the molten metal in the mold cavity (or cavities) are caused by centrifugal acceler- 
ation. Centrifugal casting processes can be classified as true centrifugal casting, 

FIGURE 3.15 

A vertical cold-chamber 
die casting machine 



3.2 Classifications of Casting by Method of Filling the Mold 


semicentrifugal casting, and the centrifuging method. Each of these processes is briefly 
discussed next. 

True centrifugal casting. True centrifugal casting involves rotating a cylindrical mold 
around its own axis, with the revolutions per minute high enough to create an effective 
centrifugal force, and then pouring molten metal into the mold cavity. The molten metal 
is pushed to the walls of the mold by centrifugal acceleration (usually 70 to 80 times that 
of gravity), where it solidifies in the form of a hollow cylinder. The outer shape of the 
casting is given by the mold contour, while the diameter of the inner cylindrical surface 
is controlled by the amount of molten metal poured into the mold cavity. The machines 
used to spin the mold may have either horizontal or vertical axes of rotation. Short tubes 
are usually cast in vertical-axis machines, whereas longer pipes, like water supply and 
sewer pipes, are cast using horizontal-axis machines. The basic features of a true cen- 
trifugal casting machine with a horizontal axis are shown in Figure 3.16. 

Centrifugal castings are characterized by their high density, refined fine-grained 
structure, and superior mechanical properties, accompanied by a low percentage of re- 
jects and, therefore, a high production output. A further advantage of the centrifugal 
casting process is the high efficiency of metal utilization due to the elimination of 
sprues and risers and the small machining allowance used. 

Semicentrifugal casting. Semicentrifugal casting is quite similar to the preceding 
type, the difference being that the mold cavity is completely filled with the molten 
metal. But because centrifugal acceleration is dependent upon the radius, the central 
core of the casting is subjected to low pressure and is, therefore, the region where en- 
trapped air and inclusions are present. For this reason, the semicentrifugal casting 
process is recommended for producing castings that are to be subjected to subsequent 
machining to remove their central cores. Examples include cast track wheels for tanks, 

FIGURE 3.16 

A true centrifugal 
casting machine 




56 3 Casting and Foundry Work 

tractors, and the like. A sand core is sometimes used to form the central cavity of the 
casting in order to eliminate the need for subsequent machining operations. 

Centrifuging. In the centrifuging method, a number of mold cavities are arranged on 
the circumference of a circle and are connected to a central down sprue through radial 
gates. Next, molten metal is poured, and the mold is rotated around the central axis of 
the sprue. In other words, each casting is rotated around an axis off (shifted from) its 
own center axis. Therefore, mold cavities are filled under high pressure, so the process 
is usually used for producing castings with intricate shapes; the increased pressure on 
the casting during solidification allows the fine details of the mold to be obtained. 

Continuous Casting 

The continuous casting process is gaining widespread industrial use, especially for 
high-quality alloy steel. In fact, the process itself passed through a few evolutionary 
stages. Although it was originally developed for producing cast-iron sheets, an up-to- 
date version is now being used for casting semifinished products that are to be 
processed subsequently by piercing, forging, extrusion, and the like. 

The continuous casting process basically involves controlling the flow of a stream 
of molten metal that comes out from a water-cooled orifice in order to solidify and 
form a continuous strip (or rod). The new version of this process is usually referred to 
as rotary continuous casting because the water-cooled mold (orifice) is always oscil- 
lating and rotating at about 120 revolutions per minute during casting. Figure 3.17 il- 
lustrates the principles of rotary continuous casting. The steel is melted, refined, and 
degassed and its chemical composition controlled before it is transferred and poured 
into the caster (tundish). The molten metal then enters the rotating mold tangent to the 
edge through the bent tube. The centrifugal force then forces the steel against the mold 
wall, while lighter inclusions and impurities remain in the center of the vortex, where 
they are removed by the operator. Solidification of the metal flowing out of the mold 
continues at a precalculated rate. The resulting bar is then cut by a circular saw that is 
traveling downward at the same speed as the bar. The bar is tilted and loaded onto a 
conveyor to transfer it to the cooling bed and the rolling mill. 

The continuous casting process has the advantages of very high metal yield 
(about 98 percent, compared with 87 percent in conventional ingot-mold practice), 
excellent quality of cast, controlled grain size, and the possibility of casting special 
cross-sectional shapes. 

The V-Process 

The vacuum casting process (V-process for short) involves covering the two halves of 
the pattern with two plastic films that are 0.005 inch (0.125 mm) thick by employing 
vacuum forming (see chapter 8). The pattern is then removed, and the two formed- 
plastic sheets are tightened together to form a mold cavity that is surrounded by a flask 
filled with sand (there is no need for a binder). This mold cavity is kept in a vacuum 
as the molten metal is poured to assist and ensure easy flow. 

3.2 Classifications of Casting by Method of Filling the Mold 


FIGURE 3.17 

The principles of rotary 
continuous casting 

Bent tube 




Ceramic tube 



Guiding rolls 


Hot saw (it travels 


while cutting) 


Conveyor to rolling mill 

The V-process, developed in Japan in the early 1970s, offers many advantages, 
such as the elimination of the need for special molding sands with binders and the 
elimination of the problems associated with green sand molding (like gas bubbles 
caused by excess humidity). Also, the size of risers, vents, and sprues can be reduced 
markedly, thus resulting in an increase in the efficiency of material utilization. Figure 
3.18 shows a plastic mold being prepared for the V-process. 


3 Casting and Foundry Work 

FIGURE 3.18 

A plastic mold being prepared for the V-process (Courtesy of Spectrum Casting, Inc., Flint, 


When classified by metal, castings can be either ferrous or nonferrous. The ferrous 
castings include cast steels and the family of cast irons, whereas the nonferrous cast- 
ings include all other metals, such as aluminum, copper, magnesium, titanium, and 
their alloys. Each of these metals and alloys is melted in a particular type of foundry 
furnace that may not be appropriate for melting other metals and alloys. Also, molding 
methods and materials, as well as fluxes, degassers, and additives, depend upon the 
metal to be cast. Therefore, this classification method is popular in foundry work. Fol- 
lowing is a brief discussion of each of these cast alloys. 

Ferrous Metals 

Cast steels. Steels are smelted in open-hearth furnaces, convertors, electric-arc fur- 
naces, and electric-induction furnaces. Cast steels can be either plain-carbon, low- 
alloy, or high-alloy steel. However, plain-carbon cast steel is the most commonly 
produced type. When compared with cast iron, steel certainly has poorer casting prop- 
erties — namely, higher melting point, higher shrinkage, and poorer fluidity. Steels are 
also more susceptible to hot and cold cracks after the casting process. Therefore, cast 

3.3 Classifications of Casting by Metal to Be Cast 59 

steels are almost always subjected to heat treatment to relieve the internal stresses and 
improve the mechanical properties. 

In order to control the oxygen content of molten steels, aluminum, silicon, or man- 
ganese is used as a deoxidizer. Aluminum is the most commonly used of these ele- 
ments because of its availability, low cost, and effectiveness. 

There is an important difference between cast-steel and wrought products. This in- 
volves the presence of a "skin," or thin layer, just below the surface of a casting, where 
scales, oxides, and impurities are concentrated. Also, this layer may be chemically or 
structurally different from the base metal. Therefore, it has to be removed by machin- 
ing in a single deep cut, which is achieved through reducing the cutting speed to half 
of the conventionally recommended value. 

Gray cast iron. Gray cast iron is characterized by the presence of free graphite flakes 
when its microstructure is examined under the microscope. This kind of microstructure 
is, in fact, responsible for the superior properties possessed by gray cast iron. For in- 
stance, this dispersion of graphite flakes acts as a lubricant during machining of gray 
cast iron, thus eliminating the need for machining lubricants and coolants. When com- 
pared with any other ferrous cast alloy, gray cast iron certainly possesses superior 
machinability. The presence of those graphite flakes is also the reason for its ability to 
absorb vibrations. The compressive strength of this iron is normally four times its ten- 
sile strength. Thus, gray cast iron has found widespread application in machine tool 
beds (bases) and the like. On the other hand, gray cast iron has some disadvantages 
and limitations, such as its low tensile strength, brittleness, and poor weldability. Nev- 
ertheless, gray cast iron has the lowest casting temperature, least shrinkage, and the 
best castability of all cast ferrous alloys. 

The cupola is the most widely used foundry furnace for producing and melting 
gray cast iron. The chemical composition, microstructure, and, therefore, the proper- 
ties of the obtained castings are determined by the constituents of the charge of the 
cupola furnace. Thus, the composition and properties of gray cast iron are controlled 
by changing the percentages of the charge constituents and also by adding inoculants 
and alloying elements. Commonly used inoculants include calcium silicide, ferrosili- 
con, and ferromanganese. An inoculant is added to the molten metal (either in the 
cupola spout or ladle) and usually amounts to between 0.1 and 0.5 percent of the 
molten iron by weight. It acts as a deoxidizer and also hinders the growth of precipi- 
tated graphite flakes. It is important for a product designer to remember that the prop- 
erties of a gray cast-iron product are also dependent upon the dimensions (the 
thicknesses of the walls) of that product because the cooling rate is adversely affected 
by the cross section of the casting. Actually, the cooling rate is high for small castings 
with thin walls, sometimes yielding white cast iron. For this reason, gray cast iron 
must be specified by the strength of critical cross sections. 

White cast iron. When the molten cast-iron alloy is rapidly chilled after being 
poured into the mold cavity, dissolved carbon does not have enough time to precipi- 
tate in the form of flakes. Instead, it remains chemically combined with iron in the 
form of cementite. This material is primarily responsible for the whitish crystalline 
appearance of a fractured surface of white cast iron. Cementite is also responsible for 

60 3 Casting and Foundry Work 

the high hardness, extreme brittleness, and excellent wear resistance of this kind of 
cast iron. Industrial applications of white cast iron involve components subjected to 
abrasion. Sometimes, gray cast iron can be chilled to produce a surface layer of white 
cast iron in order to combine the advantageous properties of the two types of cast 
iron. In this case, the product metal is usually referred to as chilled cast iron. 

Ductile cast iron. Ductile cast iron is also called nodular cast iron and spheroidal- 
graphite cast iron. It is obtained by adding trace amounts of magnesium to a very 
pure molten alloy of gray cast iron that has been subjected to desulfurization. Some- 
times, a small quantity of cerium is also added to prevent the harmful effects of im- 
purities like aluminum, titanium, and lead. The presence of magnesium and cerium 
causes the graphite to precipitate during solidification of the molten alloy in the form 
of small spheroids, rather than flakes as in the case of gray cast iron. This mi- 
crostructural change results in a marked increase in ductility, strength, toughness, and 
stiffness of ductile iron, as compared with gray cast iron, because the stress concen- 
tration effect of a flake is far higher than that of a spheroid (remember what you 
learned in fracture mechanics). The disadvantages of ductile iron, as compared with 
gray cast iron, include lower damping capacity and thermal conductivity. Ductile iron 
is used for making machine parts like axles, brackets, levers, crankshafts, housings, 
die pads, and die shoes. 

Compacted-graphite cast iron. Compacted-graphite (CG) cast iron falls between gray 
and ductile cast irons, both in its microstructure and mechanical properties. The free 
graphite in this type of iron takes the form of short, blunt, and interconnected flakes. 
The mechanical properties of CG cast iron are superior to those of gray cast iron but 
are inferior to those of ductile cast iron. The thermal conductivity and damping capac- 
ity of CG cast iron approach those of gray cast iron. Compacted-graphite cast iron has 
some application in the manufacture of diesel engines. 

Malleable cast iron. Malleable cast iron is obtained by two-stage heat treatment of 
white cast iron having an appropriate chemical composition. The hard white cast 
iron becomes malleable after the heat treatment due to microstructural changes. The 
combined carbon separates as free graphite, which takes the form of nodules. Be- 
cause the raw material for producing malleable iron is actually white cast iron, there 
are always limitations on casting design. Large cross sections and thick walls are 
not permitted because it is difficult to produce a white cast-iron part with these 
geometric characteristics. 

The two basic types of malleable cast iron are the pearlitic and the ferritic (black- 
heart). Although the starting alloy for both types is the same (white cast iron), the heat 
treatment cycle and the atmosphere of the heat-treating furnace are different in each 
case. Furnaces with oxidizing atmospheres are employed for producing pearlitic mal- 
leable cast iron, whereas furnaces with neutral atmospheres are used for producing fer- 
ritic malleable cast iron. When comparing the properties of these two types, the ferritic 
grades normally have higher ductility and better machinability but lower strength and 
hardness. Pearlitic grades can, however, be subjected to further surface hardening 
when the depth of the hardened layer is controlled. 

3.3 Classifications of Casting by Metal to Be Cast 


FIGURE 3.19 

The heat treatment 
sequence for producing 
malleable cast iron 

Temperature ,, 

(850-950°C) 1700°F 
(800°C) 1400°F 


First stage 

per hour 


-X^ cior-nnH stage 5°C 



Time (hours) 

6 hours 

-* Up to 100 hours +~ 

Figure 3.19 shows the heat treatment sequence for producing malleable cast iron. 
Referred to as the malleabilizing cycle, it includes two stages, as shown in Figure 3.19. 
In the first stage, the casting is slowly heated to a temperature of about 1700°F (950°C) 
and is kept at that temperature for about 24 hours. In the second stage, the temperature 
is decreased very slowly at a rate of 5°F to 9°F (3°C to 5°C) per hour from a temper- 
ature of 1400°F (800°C) to a temperature of 1200°F (650°C), where the process ends 
and the casting is taken out of the furnace. The whole malleabilizing cycle normally 
takes about 100 hours. 

Malleable cast iron is usually selected when the engineering application requires 
good machinability and ductility. Excellent castability and high toughness are other 
properties that make malleable cast iron attractive as an engineering material. Typical 
applications of malleable cast iron include flanges, pipe fittings, and valve parts for 
pressure service at elevated temperatures, steering-gear housings, mounting brackets, 
and compressor crankshafts and hubs. 

Alloyed cast irons. Alloying elements like chromium, nickel, and molybdenum are 
added to cast irons to manipulate the microstructure of the alloy. The goal is to im- 
prove the mechanical properties of the casting and also to impart some special proper- 
ties to it, like resistance to wear, corrosion, and heat. A typical example of alloyed 
irons is the white cast iron containing nickel and chromium that is used for corrosion- 
resistant (and abrasion-resistant) applications like water pump housings and grinding 
balls (in a ball mill). 

Nonferrous Metals 

Cast aluminum and its alloys. Aluminum continues to gain wide industrial applica- 
tion, especially in the automotive and electronics industries, because of its distin- 
guished strength-to-weight ratio and its high electrical conductivity. Alloying elements 
can be added to aluminum to improve its mechanical properties and metallurgical 
characteristics. Silicon, magnesium, zinc, tin, and copper are the elements most com- 
monly alloyed with aluminum. In fact, most metallic elements can be alloyed with 

62 3 Casting and Foundry Work 

aluminum, but commercial and industrial applications are limited to those just men- 

A real advantage of aluminum is that it can be cast by almost all casting processes. 
Nevertheless, the common methods for casting aluminum include die casting, gravity 
casting in sand and permanent molds, and investment casting (the lost-foam process). 

The presence of hydrogen when melting aluminum always results in unsound 
castings. Typical sources of hydrogen are the furnace atmosphere and the charge metal. 
When the furnace has a reducing atmosphere because of incomplete combustion of the 
fuel, carbon monoxide and hydrogen are generated and absorbed by the molten metal. 
The presence of contaminants like moisture, oil, or grease, which are not chemically 
stable at elevated temperatures, can also liberate hydrogen. Unfortunately, hydrogen is 
highly soluble in molten aluminum but has limited solubility in solidified aluminum. 
Therefore, any hydrogen that is absorbed by the molten metal is liberated or expelled 
during solidification, causing porosity. Hydrogen may also react with (and reduce) 
metallic oxides to form water vapor, which again causes porosity. Thus, hydrogen must 
be completely removed from molten aluminum before casting. This is achieved by 
using appropriate degassers. Chlorine and nitrogen are considered to be the traditional 
degassers for aluminum. Either of these is blown through the molten aluminum to 
eliminate any hydrogen. However, because chlorine is toxic and nitrogen is not that ef- 
ficient, organic chloride fluxing compounds (chlorinated hydrocarbons) are added to 
generate chlorine within the melt. They are commercially available in different forms, 
such as blocks, powders, and tablets; the most commonly used fluxing degasser is per- 
haps hexachlorethane. Another source of problems when casting aluminum is iron, 
which dissolves readily in molten aluminum. Therefore, care must be taken to spray 
(or cover) iron ladles and all iron surfaces that come into direct contact with the molten 
aluminum with a ceramic coating. This extends the service life of the iron tools used 
and also results in sound castings. 

The most important cast-aluminum alloys are those containing silicon, which 
serves to improve the castability, reduce the thermal expansion, and increase the wear 
resistance of aluminum. Small additions of magnesium make these alloys heat treat- 
able, thus allowing the final properties of the castings to be controlled. Aluminum- 
silicon alloys (with 5 to 13 percent silicon) are used in making automobile parts (e.g., 
pistons) and aerospace components. 

Aluminum-copper alloys are characterized by their very high tensile-strength-to- 
weight ratio. They are, therefore, mainly used for the manufacture of premium-quality 
aerospace parts. Nevertheless, these alloys have poorer castability than the aluminum- 
silicon alloys. Also, amounts of the copper constituent in excess of 12 percent make 
the alloy brittle. Copper additions of up to 5 percent are usually used and result in im- 
proved high-temperature properties and machinability. 

Additions of magnesium to aluminum result in improved corrosion resistance and 
machinability, higher strength, and attractive appearance of the casting when anodized. 
However, aluminum-magnesium alloys are generally difficult to cast. Zinc is also used 
as an alloying element, and the aluminum-zinc alloys have good machinability and mod- 
erately high strength. But these alloys are generally prone to hot cracking and have 
poorer castability and high shrinkage. Therefore, zinc is usually alloyed with aluminum 

3.4 Foundry Furnaces 63 

in combination with other alloying elements and is employed in such cases for pro- 
moting very high strength. Aluminum-tin alloys are also in use. They possess high load- 
carrying capacity and fatigue strength and are, therefore, used for making bearings and 

Cast copper alloys. The melting temperatures of cast copper alloys are far higher 
than those of aluminum, zinc, or magnesium alloys. Cast copper alloys can be grouped 
according to their composition as follows: 

1. Pure copper and high-copper alloys 

2. Brasses (alloys including zinc as the principal alloying element) 

3. Bronzes (alloys including tin as the principal alloying element) 

4. Nickel silvers, including copper-nickel alloys and copper-nickel-zinc alloys 

Cast copper alloys are melted in crucible furnaces, open-flame furnaces, induction 
furnaces, or indirect-arc furnaces. The selection of a furnace depends upon the type of 
alloy to be melted, as well as the purity and quantity required. In melting pure copper, 
high-copper alloys, bronzes, or nickel silver, precautions must be taken to prevent con- 
tamination of the molten metal with hydrogen. It is recommended that the atmosphere 
of the furnace be slightly oxidizing and also that a covering flux be used. Prior to cast- 
ing, however, the molten metal should be deoxidized by adding phosphorus in the form 
of a phosphorous copper flux. On the other hand, brass is usually not susceptible to hy- 
drogen porosity. The problem associated with melting brass is the vaporization and ox- 
idation of the zinc. As a remedy, the atmosphere of the furnace should be slightly 
reducing. Also, a covering flux should be used to prevent vaporization of the zinc; a 
deoxidizing flux (like phosphorous copper) is then added immediately prior to pouring. 
The applications of cast-copper alloys include pipe fitting, ornaments, propeller hubs 
and blades, steam valves, and bearings. 

Zinc alloys. The family of zinc alloys is characterized by low melting temperatures. 
Zinc alloys also possess good fluidity. Therefore, they can be produced in thin sections 
by submerged-hot-chamber die casting. Alloying elements employed include alu- 
minum, copper, and magnesium. 

Magnesium alloys. The main characteristic of magnesium is its low density, which is 
lower than that of any other commercial metal. The potential uses of magnesium are 
many because it is readily available as a component of seawater and most of its disad- 
vantages and limitations can be eliminated by alloying. Magnesium alloys usually are 
cast in permanent molds or are produced by hot-chamber die casting. 


Various furnaces are employed for smelting different ferrous and nonferrous metals in 
foundry work. The type of foundry furnace to be used is determined by the kind of 
metal to be melted, the hourly output of molten metal required, and the purity desired. 
Following is a brief review of each of the commonly used foundry furnaces. 


3 Casting and Foundry Work 

Cupola Furnaces 

Structure. The cupola is the most widely used furnace for producing molten gray 
cast iron. A sketch of a cupola furnace is given in Figure 3.20. As can be seen, the 
cupola is a shaft-type furnace whose height is three to five times its diameter. It is 
constructed of a steel plate that is about 3/8 inch (10 mm) thick and that is internally 
lined with refractory fireclay bricks. The whole structure is erected on legs, or 
columns. Toward the top of the furnace is an opening through which the charge is 
fed. Air, which is needed for the combustion, is blown through the tuyeres located 
about 36 inches (900 mm) above the bottom of the furnace. Slightly above the bot- 
tom and in the front are a tap hole and spout to allow molten cast iron to be col- 
lected. There is also a slag hole located at the rear and above the level of the tap hole 

FIGURE 3.20 

A cupola furnace 

Steel sheet 


Molten metal 

— Molten-metal 

3.4 Foundry Furnaces 


(because slag floats on the surface of molten iron). The bottom of the cupola is 
closed with drop doors to dump residual coke or metal and also to allow for mainte- 
nance and repair of the furnace lining. 

Operation. A bed of molding sand is first rammed on the bottom to a thickness of 
about 6 inches (150 mm) or more. A bed of coke about 40 inches (1.0 m) thick is next 
placed on the sand. The coke is then ignited, and air is blown at a lower-than-normal 
rate. Next, the charge is fed into the cupola through the charging door. Many factors, 
such as the charge composition, affect the final structure of the gray cast iron obtained. 
Nevertheless, it can generally be stated that the charge is composed of 25 percent pig 
iron, 50 percent gray cast-iron scrap, 10 percent steel scrap, 12 percent coke as fuel, 
and 3 percent limestone as flux. These constituents form alternate layers of coke, lime- 
stone, and metal. Sometimes, ferromanganese briquettes and inoculants are added to 
the charge to control and improve the structure of the cast iron produced. 

Direct Fuel-Fired Furnaces 
(Reverberatory Furnaces) 

The direct fuel-fired furnace, or reverberatory furnace, is used for the batch-type melt- 
ing of bronze, brass, or malleable iron. The burners of the furnace are fired with pul- 
verized coal or another liquid petroleum product. Figure 3.21 shows that the roof of 
the reverberatory furnace reflects the flame onto the metal placed on the hearth, thus 
heating the metal and melting it. The gaseous products of combustion leave the furnace 
through the flue duct. The internal surface of the furnace is lined with fire bricks, and 
there are charging and tap holes. When iron is melted, the fuel-air ratio is adjusted to 
produce a completely white iron without free graphite flakes because they lower the 
properties of the resulting malleable iron. 

Crucible (Pot) Furnaces 

Nonferrous metals like bronzes, brasses, aluminum, and zinc alloys are usually melted 
in a crucible, or pot, furnace. Crucible furnaces are fired by liquid, gaseous, or pulver- 
ized solid fuel. Figure 3.22 shows that the products of combustion in a crucible furnace 

FIGURE 3.21 

A reverberatory furnace 




Casting and Foundry Work 

FIGURE 3.22 

A crucible furnace 




do not come in direct contact with the molten metal, thus enabling the production of 
quality castings. Crucible furnaces can be stationary or tilting. When the stationary 
type is employed, crucibles are lifted out by tongs and are then carried in shanks. On 
the other hand, crucibles with long pouring lips are always used with the tilting type. 
Crucibles are made of either refractory material or alloy steels (containing 25 per- 
cent chromium). Refractory crucibles can be of the clay-graphite ceramic-bonded type 
or the silicon-carbide carbon-bonded type. The first type is cheaper, while the second 
one is more popular in industry. Ceramic crucibles are used when melting aluminum, 
bronze, or gray cast iron, whereas brasses are melted in alloy steel crucibles. Different 
alloys must not be melted in the same crucible to avoid contamination of the molten 

Electric Furnaces 

An electric furnace is usually used when there is a need to prevent the loss of any con- 
stituent element from the alloy and when high purity and consistency of casting qual- 
ity are required. An electric furnace is also employed when melting high-temperature 
alloys. In all types of electric furnaces, whether they are electric-arc, resistance, or in- 
duction furnaces, the electric energy is converted into heat. 

Electric-arc furnace. The electric-arc furnace is the most commonly used type of 
electric furnace. Figure 3.23 is a sketch of an electric-arc furnace. The heat generated 
by an electric arc is transferred by direct radiation or by reflected radiation off the in- 

FIGURE 3.23 

An electric-arc furnace 


Gear system (for rotating 
rum at an adequate 
e for pouring the 
molten metal) 

3.4 Foundry Furnaces 


FIGURE 3.24 

An electric-resistance 






ternal lining of the furnace. The electric arc is generated about midway between two 
graphite electrodes. In order to control the gap between the two electrodes and, ac- 
cordingly, control the intensity of heat, one electrode is made stationary and the other 
one movable. Electric-arc furnaces are used mainly for melting steels and, to a lesser 
extent, gray cast iron and some nonferrous metals. 

Resistance furnace. The resistance furnace is employed mainly for melting alu- 
minum and its alloys. Figure 3.24 indicates the basic features of a typical resistance 
furnace. The solid metal is placed on each of the two inclined hearths and is subjected 
to heat radiation from the electric-resistance coils located above it. When the metal 
melts, it flows down into a reservoir. The molten metal can be poured out through the 
spout by tilting the whole furnace. 

Induction furnace. The induction furnace has many advantages, including evenly dis- 
tributed temperatures within the molten metal, flexibility, and the possibility of con- 
trolling the atmosphere of the furnace. In addition, the motor effect of the 
electromagnetic forces helps to stir the molten metal, thus producing more homoge- 
neous composition. Induction furnaces are used to melt steel and aluminum alloys. 
Figure 3.25 shows the construction of a typical induction furnace. It basically involves 
an electric-induction coil that is built into the walls of the furnace. An alternating cur- 
rent in the coil induces current in any metallic object that obstructs the electromagnetic 

FIGURE 3.25 

An electric-induction 



Molten metal 

(under stirring 


coil (copper tubing) 

68 3 Casting and Foundry Work 

flux. Furnaces of both high- and low-frequency current are successfully used in indus- 
try to induce alternating current in solid metal to melt it. 


Common Defects in Castings 

In order to obtain a sound casting, it is necessary to control adequately the various fac- 
tors affecting the casting process. Casting and pattern designs, molding procedure, and 
melting and pouring of molten metal are among the factors affecting the soundness of 
a casting. Following is a survey of the commonly experienced defects in castings. 

Hot tears. Hot tears can appear on the surface or through cracks that initiate during 
cooling of the casting. They usually are in locations where the metal is restrained from 
shrinking freely, such as a thin wall connecting two heavy sections. 

Cold shut. A cold shut is actually a surface of separation within the casting. It is be- 
lieved to be caused by two "relatively cold" streams of molten metal meeting each 
other at that surface. 

Sand wash. A sand wash can be described as rough, irregular surfaces (hills and val- 
leys) of the casting that result from erosion of the sand mold. This erosion is, in turn, 
caused by the metal flow. 

Sand blow. A sand blow is actually a surface cavity that takes the form of a very 
smooth depression. It can be caused by insufficient venting, lack of permeability, or a 
high percentage of humidity in the molding sand. 

Scab. A scab is a rough "swollen" location in the casting that has some sand embed- 
ded in it. Such a defect is usually encountered when the molding sand is too fine or too 
heavily rammed. 

Shrinkage porosity (or cavity). A shrinkage porosity is a microscopic or macroscopic 
hole formed by the shrinkage of spots of molten metal that are encapsulated by solid- 
ified metal. It is usually caused by poor design of the casting. 

Hard spots. Hard spots are hard, difficult-to-machine areas that can occur at different 

Deviation of the chemical composition from the desired one. Deviation may be due 
to the loss of a constituent element (or elements) during the melting operation. It may 
also be caused by contamination of the molten metal. 

Design Considerations 

A product designer who selects casting as the primary manufacturing process should 
make a design not only to serve the function (by being capable of withstanding the 
loads and the environmental conditions to which it is going to be subjected during its 

3.5 Casting Defects and Design Considerations 


service life) but also to facilitate or favor the casting process. Following are some de- 
sign considerations and guidelines. 

Promote directional solidification. When designing the mold, be sure that the risers 
are properly dimensioned and located to promote directional solidification of the cast- 
ing toward the risers. In other words, the presence of large sections or heat masses in 
locations distant from the risers should be avoided, and good rising practice as previ- 
ously discussed should be followed. Use can also be made of chills to promote direc- 
tional solidification. Failure to do so may result in shrinkage cavities (porosity) or 
cracks in those large sections distant from the risers. It is also very important to re- 
member that a riser will not feed a heavy section through a lighter section. 

Ensure easy pattern drawing. Make sure that the pattern can easily be withdrawn from 
the nonpermanent mold (this does not apply to investment casting). This can be 
achieved through rational selection of the parting line as well as by providing appropri- 
ate pattern draft wherever needed. In addition, undercuts or protruding bosses (espe- 
cially if their axes do not fall within the parting plane) and the like should be avoided. 
Nevertheless, remember that undercuts can be obtained, if necessary, by using cores. 

Avoid the shortcomings of columnar solidification. Dendrites often start to form on 
the cold surface of a mold and then grow to form a columnar casting structure. This 
almost always results in planes of weakness at sharp corners, as illustrated in Fig- 
ure 3.26a. Therefore, rounding the edges is a must for eliminating the development of 
planes of weakness, as shown in Figure 3.26b. Rounded edges are also essential for 
smooth laminar flow of the molten metal. 

Avoid hot spots. Certain shapes, because of their effect on the rate of heat dissipation 
during solidification, tend to promote the formation of shrinkage cavities. This is al- 
ways the case at any particular location where the rate of solidification is slower than 
that at the surrounding regions of the casting. The rate of solidification (and the rate 
of heat dissipation to start with) is slower at locations having a low ratio of surface 
area to volume. Such locations are usually referred to as hot spots in foundry work. 
Unless precautions are taken during the design phase, hot spots and, consequently, 
shrinkage cavities are likely to occur at the L, T, V, Y, and + junctions, as illustrated in 
Figure 3.27a. Shrinkage cavities can be avoided by modifying the design, as shown in 

FIGURE 3.26 

Columnar solidification 
and planes of 
weakness: (a) poor 
design (sharp corner); 
(b) rounded edges to 
eliminate planes of 




3 Casting and Foundry Work 

FIGURE 3.27 

Hot spots: (a) poor 
design, yielding hot 
spots; (b) better 
design, eliminating hot 


<2X Cored 


Figure 3.27b. Also, it is always advisable to avoid abrupt changes in sections and to 
use taper (i.e., make the change gradual), together with generous radii, to join thin to 
heavy sections, as shown in Figure 3.28. 

Avoid the causes of hot tears. Hot tears are casting defects caused by tensile stresses 
as a result of restraining a part of the casting. Figure 3.29a and b shows locations where 
hot tears can occur and a recommended design that would eliminate their formation. 

Distribute the masses of a section to save material. Cast metals are generally 
weaker in tension in comparison with their compressive strengths. Nonetheless, the 
casting process offers the designer the flexibility of distributing the masses of a section 
with a freedom not readily available when other manufacturing processes are em- 
ployed. Therefore, when preparing a design of a casting, try to distribute masses in 
such a manner as to lower the magnitude of tensile stresses in highly loaded areas of 
the cross section and to reduce material in lightly loaded areas. As can be seen in Fig- 
ure 3.30, a T section or an I beam is more advantageous than just a round or square 
one when designing a beam that is to be subjected to bending. 

Avoid thicknesses lower than the recommended minimum section thickness. The 

minimum thickness to which a section of a casting can be designed depends upon such 
factors as the material, the size, and the shape of the casting as well as the specific 

FIGURE 3.28 

Avoiding abrupt 
changes in sections 








3.5 Casting Defects and Design Considerations 


FIGURE 3.29 

Hot tears: (a) a casting 
design that promotes 
hot tears; (b) 
recommended design 
to eliminate hot tears 



casting process employed (i.e., sand casting, die casting, etc.). In other words, strength 
and rigidity calculations may prove a thin section to be sufficient, but casting consid- 
erations may require adopting a higher value for the thickness so that the cast sections 
will fill out completely. This is a consequence of the fact that a molten metal cools 
very rapidly as it enters the mold and may become too cold to fill a thin section far 
from the gate. A minimum thickness of 0.25 inch (6 mm) is suggested for design use 
when conventional steel casting techniques are employed, but wall thicknesses of 
0.060 inch (1.5 mm) are quite common for investment castings. Figure 3.31 indicates 
the relationship between the minimum thickness of a section and its largest dimension. 
It should be pointed out that for a given thickness, steel flows best in a narrow rather 
than in a wide web. For cast-iron and nonferrous castings, the recommended values for 
minimum thicknesses are much lower than those for steel castings having the same 
shape and dimensions. 

Strive to make small projections in a large casting separate. As can be seen in Fig- 
ure 3.32a, a small projection may be subjected to more accidental knocks than a large 

FIGURE 3.30 

Distribution of masses 
to reduce weight 

n bending 


c r7 


T section 

Stress distribution 

I beam 

Stress distribution 

Very low 
stress area 
(material not 
fully utilized) 

Square bar 

Stress distribution 


3 Casting and Foundry Work 

FIGURE 3.31 


Minimum thickness of 


cast steel sections as 

a function of their 



largest dimension 
(Adapted from Steel 
Castings Handbook, 






5th ed. Rocky River, 
Ohio: Steel Founders 



Society of America, 


_ ^^^^-^ 

Length of section (cm) 





400 500 600 700 
i i i i 



I 1 

i i 



30 » 

20 £ 


10 I 

50 100 150 200 

Length of section (in.) 



casting, and if it gets broken, the whole casting will be scrapped. It is, therefore, highly 
recommended to make the small projection separate and attach it to the large casting 
by an appropriate mechanical joining method, as shown in Figure 3.32b. 

Strive to restrict machined surfaces. Whereas some castings are used in their en- 
tirely as-cast condition, some others may require one or more machining operations. It 
is the task of the designer to ensure that machining is performed only on areas where 
it is absolutely necessary. An example of cases where the machining needed involves 
bearing surfaces is shown in Figure 3.33. 

Use reinforcement ribs to improve the rigidity of thin, large webs. A common use of 
brackets or reinforcement ribs is to provide rigidity to thin, large webs (or the like) as 
an alternative to increasing the thickness of the webs. The ribs should be as thin as 
possible (i.e., minimum permissible thickness) and should also be staggered, as shown 
in Figure 3.34. Always remember that parabolic ribs are better than straight ribs in 
terms of economy and uniformity of stress. 

Consider the use of cast-weld construction to eliminate costly cored design. The de- 
sign of some products necessitates the use of complicated steel- wire-reinforced cores 
that are difficult to reach and remove after casting, thus leaving the surfaces unclean. 
An example, a steam ring, is shown in Figure 3.35a. The alternative design would be 
to employ a simple cut plate that is welded into the casting to produce the cast-weld 
construction shown in Figure 3.35b. 



The cleaning process involves the removal of the molding sand adhering to a casting. 
It also includes the elimination of gates, runners, and risers. Generally, surface clean- 
ing can be carried out in rotary separators or by employing sand-blasting and/or metal- 

3.6 Cleaning, Testing, and Inspection of Castings 


FIGURE 3.32 

Large casting with a 
small projection: (a) as 
an integral part; (b) two 
separate parts 

Small projection 

lie shot-blasting machines. The latter two machines use sand particles or shots travel- 
ing at high velocities onto the surface of the casting to loosen and remove the adher- 
ing sand. As you may expect, these machines are particularly suitable when cleaning 
medium and heavy castings. On the other hand, rotary separators are advantageous for 
cleaning light castings. A separator is actually a long, large-diameter drum that ro- 
tates around its horizontal axis into which the castings are loaded together with jack 


3 Casting and Foundry Work 

FIGURE 3.33 

Restriction of surfaces 
to be machined 

FIGURE 3.34 

Use of reinforcement 

1 in. 
(25 mm) 

FIGURE 3.35 

The design of a steam 
ring: (a) cast 
construction; (b) cast- 
weld construction 


/////////////////////// / TT7 

7777/7^) /////// 77^777. _ 


Plate welded 

3.7 Castability (Fluidity) 75 

stars made of white cast iron. A further advantage of rotary separators is that they au- 
tomatically break off gate-and-runner systems and, often, risers. 

Testing and Inspection 

Like any other manufactured parts, castings must be subjected to thorough quality con- 
trol in order to separate defective products and to reduce the percentage of rejects 
through identifying the defects and tracing their sources. Following are some of the 
commonly used tests and inspection methods. 

Testing of the mechanical properties of the casting. Standard tension and hardness 
tests are carried out to determine the mechanical properties of the metal of the casting 
in order to make sure that they conform to the specifications. 

Inspection of the dimensions. Dimensions must fall within the specified limits. 
Therefore, measuring tools and different kinds of gages (e.g., snap, progressive, plug, 
template) are used to check that the dimensions conform to the blueprint. 

Visual examination. Visual inspection is used to reveal only very clear defects. How- 
ever, it is still commonly used in foundries. 

Hydraulic leak testing. The hydraulic leak test is used to detect microscopic shrink- 
age porosity. Various penetrants and testing methods are now available. Details are 
given in the American Society for Testing and Materials (ASTM) standards, designa- 
tion E165. 

Nondestructive testing. There are several nondestructive testing methods that detect 
microscopic and hair cracks. They involve ultrasonic testing, magnetic particle inspec- 
tion, eddy current testing, and radiography. 

Testing for metal composition. Several methods are employed to determine the chem- 
ical composition accurately and to assure product quality. The classical method used to 
be "wet analysis" (i.e., employing acids and reagents in accurate chemical analysis). 
However, because this method is time-consuming, it is being replaced by methods like 
emission spectroscropy, X-ray fluorescence, and atomic absorption spectroscopy. 


The ability of the molten metal to flow easily without premature solidification is a 
major factor in determining the proper filling of the mold cavity. This important prop- 
erty is referred to as castability or, more commonly, fluidity. The higher the fluidity of 
a molten metal, the easier it is for that molten metal to fill thin grooves in the mold and 
exactly reproduce the shape of the mold cavity, thereby successfully producing cast- 
ings with thinner sections. Poor fluidity leads to casting defects such as incomplete fill- 
ing or misruns, especially in the thinner sections of a casting. Because fluidity is 
dependent mainly upon the viscosity of the molten metal, it is clear that higher tem- 
peratures improve the fluidity of molten metal and alloys, whereas the presence of im- 
purities and nonmetallic inclusions adversely affects it. 


3 Casting and Foundry Work 

FIGURE 3.36 

Details of the test for 
measuring fluidity 

0.3 inch 


«-i*A7I[ a3 


section A-A 

Several attempts have been made to quantify and measure the fluidity of metals. 
A commonly used standard test involves pouring the molten metal into a basin so that 
it flows along a spiral channel of a particular cross section, as shown in Figure 3.36. 
Both the basin and the channel are molded in sand, and the fluidity value is indicated 
by the distance traveled by the molten metal before it solidifies in the spiral channel. 

Review Questions 


1. What is meant by the word casting ? 

2. What are the constituents of green molding 

3. List some of the important properties that green 
sand must possess. 

4. What is a flask? What is its function? List the 
parts that form a flask. 

5. Explain the meaning of the word pattern. 

6. List some of the materials used in making pat- 

7. List the different types of permanent patterns 
used in foundries. 

8. What are the different pattern allowances? Dis- 
cuss the function of each. 

9. What are cores? How are they made? 

10. What is meant by a gating system ? What func- 
tions does it serve? 

11. What are the components of a gating system? 

12. What are risers? What function do they serve? 

13. List the various green sand properties and dis- 
cuss each briefly. 

14. Why should weights be located on the cope in 
pit molding? 

Chapter 3 Review Questions 


15. List the various molding machines and discuss 
the operation of each briefly. 

16. Explain sand conditioning and how it is done. 

17. What advantages does dry sand molding have 
over green sand molding? 

18. When are cement-bonded sand molds recom- 

19. What is the main advantage of the carbon diox- 
ide process for molding? 

20. What metals can be cast in plaster molds? 

21. When are loam molds used? 

22. Describe shell molding. What are its advan- 

23. When are ceramic molds recommended? 

24. Explain investment casting and why it is some- 
times called the lost-wax process. 

25. Name a metal that should be cast in a graphite 

26. What are the advantages of employing perma- 
nent molds? Why? 

27. Can molten metals be cast directly into cavities 
of cold permanent molds? Why? 

28. What is the main difference between the hot- 
chamber and the cold-chamber methods of die 

29. List some metals that you think can be cast by 
the hot-chamber method. Justify your answer. 

30. List some metals that you think can be cast by 
the cold-chamber method. Justify your answer. 

31. What are the types of centrifugal casting? 

32. Differentiate between the different types of 
centrifugal casting and discuss the advantages 
and shortcomings of each type. 

33. What are the products that can be manufactured 
by continuous casting? 

34. What does the continuous casting process in- 

35. Discuss some advantages of the continuous 
casting process. 

36. What does the V-process involve? 

37. List some of the merits and advantages of the 

38. Discuss some of the problems encountered in 
casting steels. 

39. What precautions should be taken to eliminate 
the problems in casting steels? 

40. What is gray cast iron? 

41. Discuss some of the properties that make gray 
cast iron attractive for some engineering appli- 

42. Why are inoculants added to gray cast iron? 

43. Differentiate between gray cast iron and white 
cast iron. 

44. What is meant by compacted-graphite cast iron. 

45. What is ductile cast iron? How can it be ob- 

46. What is malleable cast iron? How can it be 
obtained? What are the limitations on produc- 
ing it? 

47. List some alloying elements that are added to 
cast iron. List some applications for alloyed 
cast iron. 

48. What are the problems caused by hydrogen 
when melting and casting aluminum and how 
can these problems be eliminated? 

49. What are the sources of hydrogen when melting 

50. List some cast aluminum alloys and discuss 
their applications. 

51. How are cast copper alloys classified? 

52. What is meant by a deoxidizer? Give an ex- 

53. List some of the characteristics and applica- 
tions of cast zinc alloys. 

54. List some of the characteristics and applica- 
tions of cast magnesium alloys. 

55. For what purpose is the cupola furnace used? 

56. Describe briefly the operation and charge of the 
cupola furnace. 


3 Casting and Foundry Work 



For what purpose is the reverberatory furnace 

List some of the metals that can be melted in 
crucible furnaces. 

59. What are the main differences in construction 
between the stationary and the tilting crucible 

60. List the different types of electric furnaces and 
mention the principles of operation in each case. 

61. List the main advantages and applications of 
electric furnaces. 

62. List some of the common defects of castings 
and discuss the possible causes of each defect. 

63. List and discuss the main design considerations 
for castings. 

64. List and discuss the various testing and inspec- 
tion methods used for the quality control of 

Design Example 



Your company has received an order to manufacture wrenches for loosening and tight- 
ening nuts and bolts of large machines. The plant of the company involves a foundry 
and a machining workshop with a few basic machine tools. Here are the details of the 

Lot size: 
Nut size: 
Required torque: 

500 wrenches 

2 inches (50 mm) 

about 20 lb ft (27.12 N-m) 

You are required to provide a design and a production plan (see the explanation of the 
word design in the design projects section that appears later). 


Before we start solving this design problem, we should make some assumptions. For 
instance, consider the force that can be generated by the ordinary human hand. It will 
allow us to determine the length of the wrench using the following equation: 


where: T is the torque 
F is the force 
(, is the length 

As can be seen from the equation, a low value of F would make the length large and 
thus make the handling of the wrench impractical because of the weight. On the other 
hand, a high value of F is not practical and may not be generated by an ordinary per- 
son. Let us take F = 15 pounds. Therefore, 

Chapter 3 Design Example 79 

^ = T7= 1-33 feet 

Apparently, the force acts at the middle of the fist, and we have to add a couple of 
inches for proper holding: 

length of wrench = 1 8 inches 

Let us now design the section where the maximum bending moment occurs. You 
can assume some dimensions and determine the stress, which will serve as a guide in 
selecting material. Take the section as shown in Figure 3.37a. The moment of inertia 
of the section is 

/ = -j^(0.25)(0.75) 3 + 2 

— (0.375)(0.25) 3 + 0.375 x 0.25 x 0.5 

= 0.008789 + 0.00098 + 0.046875 
= 0.056655 in. 4 

Note that the minimum thickness for steel casting was adhered to. Now, determine the 

20 x 12 x 1.25 .,.. „ „ 2 

max. stress = = 2648 lb/in. 

2 x 0.056655 

That value is very low, and we should try to reduce the section and save material. It is 
always a good idea to make use of spreadsheets to change the dimensions and get the 
stresses acting in each case. Now, take the section as shown in Figure 3.37b: 

/ = ^(0.25)(0.5) 3 + 2 

— (0.375)(0.25) 3 + 0.375 x 0.25 x 0.375 2 

= 0.002604 + 0.00098 + 0.026367 
= 0.038771188 in. 4 

20 x 12 1.25 , , 01ur 2 

max. stress = — — — x — —— = 3868 lb/in. 

0.038771188 2 

As can be seen, we took the minimum thickness to be 0.25 inch, which is the recom- 
mended value for conventional castings of steels. 

The material should be low-carbon steel having 0.25 percent carbon in order to 
possess enough ductility. Also, the steel should be thoroughly killed. A recommended 
material is ASTM A27-77, grade U60-30, which has a yield strength of 30 ksi. When 
taking a factor of safety of 4, the allowable stress would be 7500 lb/in. , which is 
higher than the obtained value of the working stress. 

Now, in order to calculate the thickness of the wrench, let us calculate the bearing 
stress on the nut. A reasonable estimate of the force on the surface of the nut is 

20 x 12 

= 320 pounds 

0.75 F 


3 Casting and Foundry Work 

FIGURE 3.37 

Cross section of the 
wrench: (a) first 
attempt; (b) second 

0.25 inch 

0.75 inch 

0.25 inch 

0.25 inch 

0.5 inch 

0.25 inch 

This is based on the assumption that the torque is replaced by two opposite forces hav- 
ing a displacement of 0.75 inch between the lines of action. Thus, 

bearing stress = 


= 7500 

0.75 x t 
t = 0.056 inch 

Take it as 0.5 inch to facilitate casting the part. 

Because all dimensions are known, a detailed design can be prepared, as shown in 
Figure 3.38. Notice the surface finish marks indicating the surfaces to be machined. 

FIGURE 3.38 

A wrench manufactured by casting 

f?= 1.0 inch 

1.4 inch 


0.5 inch 



375 inch 

Chapter 3 Design Projects 81 

As previously mentioned, conventional sand casting is to be employed, using a 
cope-and-drag pattern plate to cast two wrenches per flask. We can use a single down 
sprue to feed the narrow end of the wrench and a riser at the other end. The parting 
line will pass through the web of the / section. 

Design Projects 

Whenever the word design is mentioned hereafter, you should provide, at least, the 

• Two neatly dimensioned graphical projections of the product (i.e., a blueprint ready 
to be released to the workshop for actual production), including fits (if applicable), 
tolerances, surface finish marks, and so on 

• Material selection with rational justification 

• Selection of the specific manufacturing processes required, as well as their se- 
quence in detail 

• Simple but necessary calculations to check the stresses at the critical sections 

1. Design a bracket for a screw C-clamp that has the following characteristics: 

Maximum clamping force: 22 pounds (100 N) 

Clamping gap: 3 inches (7.5 cm) 
Distance between centerline of screw 

and inner surface of bracket: 2 inches (5 cm) 

Root diameter of screw: 0.25 inch (6 mm) 

Assume that manufacturing is by casting and that production volume is 4000 

2. Design a flat pulley. Its outer diameter is 36 inches (90 cm), and it is to be 
mounted on a shaft that is 2 x /i inches (6.25 cm) in diameter. Its width is 10 inches 
(25 cm), and it has to transmit a torque of 3000 lb ft (4000 Nm). Assume that 500 
pieces are required. Will the design change if only 3 pieces are required? 

3. A connecting lever has two short bosses, each at one of its ends and each with a 
vertical hole that is 3/4 inch (19 mm) in diameter. The lever is straight, and the 
horizontal distance between the centers of the holes is 8 inches (200 mm). The 
lever during functioning is subjected to a bending moment of 50 lb ft (67.8 Nm) 
that acts in the plane formed by the two vertical axes. Provide a detailed design 
for this lever if it is to be produced by casting and 

a. When only 100 pieces are required 

b. When 1 0,000 pieces are required 

82 3 Casting and Foundry Work 

4. Design a micrometer frame for each of the following cases: 

a. The gap of the micrometer is 1.0 inch (25 mm), and the distance from the axis 
of the barrel to the inner side of the frame is 1 .5 inches (37.5 mm). The maxi- 
mum load on the anvil is 22 lb (100 N). 

b. The gap of the micrometer is 6 inches (150 mm), and the distance from the axis 
of the barrel to the inner side of the frame is 4.0 inches (100 mm). The maxi- 
mum load on the anvil is 22 lb (100 N). 

Assume that production volume is 4000 pieces and that one of the various casting 
processes is used. 

TIP: Base your design on rigidity. The maximum deflection must not exceed 
0.1 of the smallest reading of the micrometer. 

5. A pulley transmits a torque of 600 lb ft (813.6 Nm) to a shaft that is \ X A inches 
(31 mm) in diameter. The outer diameter of the pulley is 10 inches (250 mm), and 
it is to be driven by a flat belt that is 2 inches (50 mm) in width. Design this pul- 
ley if it is to be manufactured by casting and 500 pieces are required. 

6. Design a hydraulic jack capable of lifting 1 ton and having a stroke of 6 inches 
(150 mm). The jack is operated by a manual displacement (plunger) pump that 
pumps oil from a reservoir into the high-pressure cylinder through two spring- 
actuated nonreturn valves to push the ram upward. The reservoir and the high- 
pressure cylinder are also connected by a conduit, but the flow of oil is obstructed 
by a screw that, when unscrewed, relieves the pressure of the cylinder by allow- 
ing high-pressure oil to flow back into the reservoir and the ram then to be pushed 
downward. Provide a workshop drawing for each component, as well as an as- 
sembly drawing for the jack. Steel balls and springs are to be purchased. Assume 
production volume is 5000 pieces. 

7. Design a table for the machine shop. That table should be 4 feet (1.2 m) in height, 
with a surface area of 3 by 3 feet (0.90 by 0.9 m), and should be able to carry a 
load of half a ton. Assume production volume is 2000 pieces. 

8. Design a little wrench for loosening and tightening nuts and bolts of a bicycle. The 
nut size is 5/8 inch (15 mm), and the required torque is about 1.0 lb ft (1.356 
Nm). Assume production volume is 10,000 pieces. 

9. A straight-toothed spur-gear wheel transmits 1200 lb ft (1627 Nm) of torque to a 
steel shaft that is 2 inches (50 mm) in diameter. The pitch diameter of the gear is 
8 inches (200 mm), its width is 3 inches (75 mm), and the base diameter is 7.5 
inches (187.5 mm). Design this gear's blank. Assume production volume is 4000 

10. Design a frame for an open-arch (C-type) screw press that can deliver a load of up 
to 2 tons. The open gap is 2 feet (600 mm), and the bed on which workpieces are 
placed is 12 by 12 inches (300 by 300 mm). Assume that the base diameter of the 
screw thread is 1 Vi inches (37.5 mm). 

Chapter 3 Design Projects 83 

11. Design a hydraulic cylinder for earth-moving equipment. It can generate a maxi- 
mum force of 2 tons and has a stroke of 4 feet (1200 mm). Although the maximum 
force is generated only when the plunger rod is moving out, the cylinder is dou- 
ble acting and generates a force of 1 ton during its return stroke. Expected pro- 
duction volume is 2000 pieces, and the pistons, oil rings, and so on, are going to 
be purchased from vendors. 

12. Design a safety valve to be mounted on a high-pressure steam boiler. The pipe on 
which it will be mounted has a bore diameter of 2 inches (50 mm). The pressure 
inside the boiler should not exceed 50 folds of the atmospheric pressure. Expected 
production volume is 5000 pieces, and the stems, springs, bolts, and gaskets are 
going to be purchased from vendors. 

Chapter 4 

Inlng of Metals 

^ s — "*^& 



When two parts of metal are to be attached together, the resulting joint can be 
made dismountable (using screws and the like), or it can be made permanent 
by employing riveting, welding, or brazing processes. The design of dismount- 
able joints falls beyond the scope of this text and is covered in machine de- 
sign. It is, therefore, the aim of this chapter to discuss the design and 
production of permanent joints when various technologies and methods are ap- 
plied. Because the same equipment used in welding is also sometimes em- 
ployed in the cutting of plates, thermal cutting processes will also be 
discussed in this chapter. 

The process of riveting involves inserting a ductile metal pin through holes in two or 
more sheet metals and then forming over (heading) the ends of the metal pin so as to 
secure the sheet metals firmly together. This process can be performed either cold or 
hot, and each rivet is usually provided with one preformed head. Figure 4.1a and b in- 
dicates the sequence of operations in riveting, while Figure 4.2 illustrates different 
shapes of preformed rivet heads. 


Welding is the joining of two or more pieces of metal by creating atom-to-atom bonds 
between the adjacent surfaces through the application of heat, pressure, or both. In 
order for a welding technique to be industrially applicable, it must be reasonable in 
cost, yield reproducible or consistent weld quality, and, more importantly, produce 


4.2 Welding 



Sequence of operations 
in riveting: (a) flat-head 
rivet; (b) regular rivet 


vzz& m% , 

w&. ^m 



— V— 



joints with properties comparable to those of the base material. Various welding tech- 
niques have been developed that are aimed at achieving these three goals. However, no 
matter what welding method is used, the interface between the original two parts must 
disappear if a strong joint is to be obtained. Before we discuss the different methods 
employed to make those surfaces disappear, let us discuss joint design and preparation. 

Joint Design and Preparation 

A weld joint must be designed to withstand the forces to which it is going to be sub- 
jected during its service life. Therefore, the joint design is determined by the type and 
magnitude of the loading that is expected to act on the weldment. In other words, se- 
lection of the type of joint has to be made primarily on the basis of load requirement. 
As Figure 4.3a through e shows, there are five types of weld joints: butt, lap, corner, 
T, and edge. Following is a discussion of each of these different types of joints. 

Butt joint. The butt joint involves welding the edges or end faces of the two original 
parts, as shown in Figure 4.3a. Therefore, the two parts must be aligned in the same 
plane. Usually, when the thickness of the parts falls between 1/8 and 3/8 inch (about 3 
and 9 mm), the two parts are welded without any edge preparation. This type of weld 
is referred to as a square weld and can be either single or double, depending upon the 
thickness of the metal, as shown in Figure 4.4a. As can be seen in Figure 4.4b through 
e, the edges of thicker parts should be prepared with single or double bevels or V-, J-, 
or U-grooves to allow adequate access to the root of the joint. Usually, it is recom- 
mended to adopt the single or double U-groove when the thickness of the parts is more 
than 0.8 inch (20 mm). 

Lap joint. We can see in Figure 4.3b that the lap joint is produced by fillet welding 
overlapping members; the amount of overlap is normally taken to be about three to 


Different shapes of 
preformed rivet heads 

^Z7 c 


4 Joining of Metals 


Types of weld joints: 
(a) butt joint; (b) lap 
joint; (c) corner joint; 
(d) T-joint; (e) edge joint 

1 \ 






five times the thickness of the member. The fillet weld can be continuous and may also 
be of the plug or slot type, as shown in Figure 4.5. 

Corner joint. Figure 4.3c illustrates the corner joint, which can be welded with or 
without edge preparation (see Figure 4.4 for the various possible edge preparations). 

T-joint. The T-joint is shown in Figure 4.3d. T-joints that will be subjected to light 
static loads may not require edge preparation. On the other hand, edge preparations 
(again see Figure 4.4) are often necessary for greater metal thicknesses or when the 
joint is to be subjected to relatively high, alternating, or impulsive loading. 

Edge joint. The edge joint is usually used when welding thin sheets of metal with a 
thickness of up to 1/8 inch (3 mm). Notice in Figure 4.3e that the edges of the mem- 
bers must be bent before the welding process is carried out. 


Different edge 
preparation for butt 
welding: (a) square; (b) 
bevel; (c) V-groove; (d) 
J-groove; (e) U-groove 



Single Single Single Single 







Basic types of fusion 
lap welds 

o © 







4.2 Welding 



Weld symbols 




Groove weld 



Groove welds: 



Plug or slot 



Back or backing Spot or projection Seam 







Weld Symbols and Identification 

Figure 4.6 shows the different weld symbols, whereas Figure 4.7 shows the standard 
identification of welds employed in design drawings. 

Classification of the Welding Processes 

Different methods can be used for classifying industrial welding processes. Each 
method is employed to form groups of welding processes, with each group having 
something in common. For instance, welding processes can be classified according to 
the source of energy required to accomplish welding. In such a case, it is obvious that 
there are four main groups: mechanical, electrical, chemical, or optical. Welding 
processes can also be classified by the degree of automation adopted, which yields 
three groups: manual, semiautomatic, and automatic. The most commonly used 
method of classification is according to the state of the metal at the locations being 


Standard identification 
of welds 

Basic weld symbol 
or detail reference 

Size; size or strength 
for resistance welds 

Reference line 

Root opening; depth 

of filling for plug 

and slot welds 

Specification, process, 
or other reference 

Tail (may be omitted 

when reference 

is not used) 

Basic weld symbol 
or detail reference 



Groove angle included 
angle or countersink 
for plug weld 
Pitch (center-to-center 
spacing) of welds 
Arrow connecting reference 
line to arrow side of joint, 
to grooved member, or both 


Number of spots 
or projection welds 

Elements in this area 

remain as shown 

when tail and arrow 

are reversed 


Field weld 


all around 


88 4 Joining of Metals 

welded, thus splitting the welding processes into two main categories: pressure weld- 
ing and fusion welding. We now discuss each of these two categories in detail. 

Pressure Welding Processes 

Pressure welding involves processes in which the application of external pressure is 
indispensable to the production of weld joints formed either at temperatures below the 
melting point (solid-state welding) or at temperatures above the melting point (fusion 
welding). In both cases, it is important to have very close contact between the atoms 
of the parts that are to be joined. The atoms must be moved together to a distance that 
is equal to or less than the equilibrium interatomic-separation distance. Unfortunately, 
there are two obstacles that must be overcome so that successful pressure welding can 
be carried out and a sound weldment can be obtained. First, surfaces are not flat when 
viewed on a microscopic scale. Consequently, intimate contact can be achieved only 
where peaks meet peaks, as can be seen in Figure 4.8, and the number of bonds would 
not be enough to produce a strong welded joint. Second, the surfaces of metals are usu- 
ally covered with oxide films that inhibit direct contact between the two metal parts to 
be welded. Therefore, those oxide and nonmetallic films must be removed (cleaned 
with a wire brush) before welding in order to ensure a strong welded joint. Pressure 
welding processes are applied primarily to metals possessing high ductility or those 
whose ductility increases with increasing temperatures; thus, the peaks that keep the 
surfaces of the two metallic members apart are leveled out under the action of me- 
chanical stresses or the combined effect of high temperatures and mechanical stresses. 
In fact, a wide variety of pressure welding processes are used in industry. The com- 
monly used ones are listed in Figure 4.9. 

Cold-pressure welding. Cold-pressure welding is a kind of solid-state welding used 
for joining sheets, wires, and small electric components. As previously discussed, the 
surfaces to be welded must be cleaned with a wire brush to remove the oxide film 
and must be carefully degreased before welding. As Figure 4.10 shows, a special 
tool is used to produce localized plastic deformation, which results in coalescence be- 
tween the two parts. This process, which can replace riveting, is usually followed by 


A microscopic view of 

two mating surfaces Surface 1 

Surface 2 

4.2 Welding 



Classification of the 
commonly used 
pressure welding 

Pressure welding processes 

Cold-pressure welding 

* Cold-pressure welding 
of sheets and wires 

* Ultrasonic welding 

* Explosive welding 

Hot-pressure welding 

Molten-metal bonding 

* Percussion 

* Resistance flash 

* Resistance spot 

* Resistance seam 

* Resistance projection 

* Thermit 

Hot solid-state 
pressure welding 

* Diffusion bonding 

* Friction welding 

* Inertia welding 

* Induction welding 
« Resistance upset 

(butt) welding 

annealing of the welded joint. Figure 4.10 also shows that recrystallization takes place 
during the annealing operation. This is added to diffusion, which finally results in com- 
plete disappearance of the interface between the two parts. 

Cold-pressure welding of wires is performed by means of a special-purpose ma- 
chine. Figure 4. 1 1 illustrates the steps involved in this process. As can be seen, the 
wires' ends are clamped and pressed repeatedly against each other in order to ensure 
adequate plastic deformation. The excess upset metal is then trimmed by the sharp 
edges of the gripping jaws. This technique is used when welding wires of nonferrous 
metals such as aluminum, copper, or aluminum-copper alloys. 

Explosive welding. Explosive welding is another technique that produces solid-state 
joints and, therefore, eliminates the problems associated with fusion welding methods, 
like the heat-affected zone and the microstructural changes. The process is based on 
using high explosives to generate extremely high pressures that are, in turn, used to 
combine flat plates or cylindrical shapes metallurgically. Joints of dissimilar metals 
and/or those that are extremely difficult to combine using conventional methods can 
easily be produced by explosive welding. 

During explosive welding, a jet of soft (or fluidlike) metal is formed (on a micro- 
scopic scale) and breaks the oxide film barrier to bring the two metal parts into inti- 
mate contact. That metal jet is also responsible for the typical wavy interface between 

FIGURE 4.10 

Cold-pressure welding 
of sheets 








Welded joint after 

Welded joint after 


4 Joining of Metals 

FIGURE 4.11 

Cold-pressure welding 
of wires 


Upset metal 


S £ 

After trimming 

the two metal parts, thus creating mechanical interlocking between them and, finally, 
resulting in a strong bond. Figure 4. 1 2 illustrates an arrangement for explosive welding 
two flat plates, and Figure 4.13 is a magnified sketch of the wavy interface between 
explosively welded parts. 

FIGURE 4.12 

An arrangement for 
explosive welding two 
flat plates 




^Ml^M ^ 

Plate 1 

Plate 2 


Explosive welding and explosive cladding are popular in the manufacture of heat 
exchangers and chemical-processing equipment. Armored and reinforced composites 
with a metal matrix are also produced by explosive welding. Nevertheless, a clear 
limitation is that the process cannot be used successfully for welding hard, brittle met- 
als. Research is being carried out in this area, and new applications are continuously 

Ultrasonic welding. The ultrasonic welding method of solid-state welding is com- 
monly used for joining thin sheets or wires of similar or dissimilar metals in order 
to obtain lap-type joints. Mechanical vibratory energy with ultrasonic frequencies is 
applied along the interfacial plane of the joint, while a nominal static stress is applied, 
normal to that interface, to clamp the two components together. Oscillating shear 

FIGURE 4.13 

A sketch of the wavy 
interface between 
explosively welded 


4.2 Welding 91 

stresses are, therefore, initiated at the interface and disperse surface films, allowing in- 
timate contact between the two metals, and, consequently, producing a strong joint. Ul- 
trasonic welding does not involve the application of high pressures or temperatures 
and is accomplished within a short time. Therefore, this process is especially suitable 
for automation and has found widespread application in the electrical and microelec- 
tronics industries in the welding of thin metal foils for packaging and splicing and in 
the joining of dissimilar materials in the fabrication of nuclear reactor components. It 
must be noted, however, that the process is restricted to joining thin sheets or fine 
wires. Nevertheless, this restriction applies only to thinner pieces, and the process is 
often used in welding thin foils to thicker sheets. 

Different types of ultrasonic welding machines are available, each constructed to 
produce a certain type of weld, such as spot, line, continuous seam, or ring. A sketch 
of a spot-type welding machine that is commonly used in welding microcircuit ele- 
ments is illustrated in Figure 4.14. As we can see, the machine consists basically of a 
frequency convenor that transforms the standard 60-Hz (or 50-Hz in Europe) electric 
current into a high-frequency current (with a fixed frequency in the range of 15 to 75 
kHz), a transducer that converts the electrical power into elastic mechanical vibrations, 
and a horn that magnifies the amplitude of these vibrations and delivers them to the 
weld zone. Other associated elements include the anvil, a force-application and clamp- 
ing device, a sonotrode (as compared with the electrode in resistance welding), and ap- 
propriate controls to set up optimal values for the process variables, such as vibratory 
power and weld time. 

Friction welding. In friction welding, a type of hot solid-state welding, the parts to be 
welded are tightly clamped, one in a stationary chuck and the other in a rotatable 
chuck that is mounted on a spindle. External power is employed to drive the spindle at 
a constant speed, with the two parts in contact under slight pressure. Kinetic energy is 
converted to frictional heat at the interface. When the mating edges of the workpieces 
attain a suitable temperature (in the forging range) that permits easy plastic flow, the 
spindle rotation is halted, and high axial pressure is applied to plastically deform the 
metal, obtain intimate contact, and produce a strong, solid weld. This is clearly shown 
in Figure 4.15, which indicates the stages involved in friction welding. 

Several advantages have been claimed for the friction welding process. These in- 
clude simplicity, high efficiency of energy utilization, and the ability to join similar as 

FIGURE 4.14 Clamping 

A sketch of an Transducer H ° m *?* 

ultrasonic spot-type 
welding machine 


92 4 Joining of Metals 

FIGURE 4.15 

Stages involved in 
friction welding 




Rotating Heating stage Upsetting stage 

well as dissimilar metal combinations that cannot be joined by conventional welding 
methods (e.g., aluminum to steel or aluminum to copper). Also, since contaminants 
and oxide films are carried away from the weld area where grain refinement takes 
place, a sound bond is obtained and usually has the same strength as the base metal. 
Nevertheless, a major limitation of the process is that at least one of the two parts to 
be joined must be a body of revolution around the axis of rotation (like a round bar or 
tube). A further limitation is that only forgeable metals that do not suffer from hot 
shortness can successfully be friction welded. Also, care must be taken during welding 
to ensure squareness of the edges of workpieces as well as concentricity of round bars 
or tubes. 

Inertia welding. Inertia welding is a version of friction welding that is recom- 
mended for larger workpieces or where high-strength alloys (i.e., superalloys) are to 
be joined together. Inertia welding, as the name suggests, efficiently utilizes the ki- 
netic energy stored in a rotating flywheel as a source for heating and for much of the 
forging of the weld. As is the case with friction welding, the two workpieces to 
be inertia welded are clamped tightly in stationary and rotatable chucks, the differ- 
ence being that the rotatable chuck is rigidly coupled to a flywheel in the case of 
inertia welding. The process involves rotating the flywheel at a predetermined angu- 
lar velocity and then converting the kinetic energy of the freely rotating flywheel to 
frictional heat at the weld interface by applying an axial load to join the abutting 
ends under controlled pressure. The process requires shorter welding time than that 
taken in conventional friction welding, especially for larger workpieces. Examples of 
inertia-welded components include hydraulic piston rods for agricultural machinery, 
carbon steel shafts welded to superalloy turbocharger wheels, and bar stock welded 
to small forgings. 

Induction welding. As the name suggests, induction welding is based on the phe- 
nomenon of induction. We know from physics (electricity and magnetism) that when 
an electric current flows in an inductor coil, another electric current is induced in any 
conductor that intersects with the magnetic flux. In induction welding, the source of 
heat is the resistance, at the abutting workpieces' interface, to the flow of current in- 
duced in the workpieces through an external induction coil. Figure 4.16 illustrates the 
principles of induction welding. For efficient conversion of electrical energy into heat 
energy, high-frequency current is employed, and the process is usually referred to as 
high-frequency induction welding (HFIW). Frequencies in the range of 300 to 450 kHz 
are commonly used in industry, although frequencies as low as 10 kHz are also in use. 
It is always important to remember the "skin effect" when designing an induction- 
welded joint. This effect refers to the fact that the electric current flows superficially 
(i.e., near the surface). In fact, the depth of the layer through which the current flows 
is dependent mainly upon the frequency and the electromagnetic properties of the 

4.2 Welding 


FIGURE 4.16 

Principles of induction 




Coil ^I - 


workpiece metal. Industrial applications of induction welding include butt welding of 
pipes and continuous-seam welding for the manufacture of seamed pipes. 

Thermit welding. Thermit welding makes use of an exothermic chemical reaction to 
supply heat energy. That reaction involves the burning of thermit, which is a mixture 
of fine aluminum powder and iron oxide in the form of rolling-mill scale, mixed at a 
ratio of about 1 to 3 by weight. Although a temperature of 5400°F (3000°C) may be 
attained as a result of the reaction, localized heating of the thermit mixture up to at 
least 2400°F (1300°C) is essential in order to start the reaction, which can be given by 
the following chemical formula: 

8A1 + 3Fe 3 4 -> 9Fe + 4A1 2 3 + heat (4.1) 

As we can see from the formula, the outcome is very pure molten iron and slag. In fact, 
other oxides are also used to produce pure molten metals; these include chromium, 
manganese, or vanadium, depending upon the parent metals to be welded. 

Usually, the thermit welding process requires the application of pressure in order 
to achieve proper coalescence between the parts to be joined. However, fusion thermit 
welding is also used; it does not require the application of force. In this case, the re- 
sulting molten metal is a metallurgical joining agent and not just a means for heating 
the weld area. 

Thermit welding is used in joining railroad rails, pipes, and thick steel sections, as 
well as in repairing heavy castings. The procedure involves fitting a split-type refrac- 
tory mold around the abutting surfaces to be welded, igniting the thermit mixture using 
a primer (ignition powder) in a special crucible, and, finally, pouring the molten metal 
(obtained as a result of the reaction) into the mold. Because the temperature of the 
molten metal is about twice the melting point of steel, the heat input is enough to fuse 
the abutting surfaces, which are usually pressed together to give a sound weld. 

Diffusion bonding. Diffusion bonding is a solid-state welding method in which the 
surfaces to be welded are cleaned and then maintained at elevated temperatures under 
appropriate pressure for a long period of time. No fusion occurs, deformation is lim- 
ited, and bonding takes place principally due to diffusion. As we know from metal- 
lurgy, the process parameters are pressure, temperature, and time, and they should be 
adjusted to achieve the desired results. 

Butt welding. Butt welding belongs to the resistance welding group, which also con- 
sists of the spot, seam, projection, percussion, and flash welding processes. All of these 


4 Joining of Metals 

FIGURE 4.17 

Upset-butt welding 

Clamping dies 



Ac power supply 

operate on the same principle, which involves heating the workpieces as a result of 
being a part of a high-amperage electric circuit and then applying external pressure to 
accomplish strong bonding. Consequently, all the resistance welding processes belong 
to the larger, more general group of pressure welding; without the application of ex- 
ternal pressure, the weld joint cannot be produced. 

In butt welding, sometimes called upset-butt welding or just upset welding, the 
parts are clamped and brought in solid contact, and low-voltage (1 to 3 V) alternating 
current is switched on through the contact area, as illustrated in Figure 4.17. As a re- 
sult of the heat generated, the metal in the weld zone assumes a plastic state (above the 
solidus) and is gradually squeezed and expelled from the contact area. When enough 
upset metal becomes evident, the current is switched off and the welded parts are re- 
leased. Figure 4.18 indicates a typical upset welding cycle. Note that upset welding 
would not be successful for larger sections because these cannot be uniformly heated 
and require extremely high-amperage current. Therefore, the process is limited to 
welding wires and rods up to 3/8 inch (about 10 mm) in diameter. Also, a sound joint 
can be ensured only when the two surfaces being welded together have the same cross- 
sectional area as well as negligible or no eccentricity. 

Flash welding. Flash welding is somewhat similar to upset welding. The equipment 
for flash welding includes a low-voltage transformer (5 to 10 V), a current timing 

FIGURE 4.18 

A typical upset welding 


Solid contact 

Upset metal 

4.2 Welding 


FIGURE 4.19 

Stages in a flash 
welding cycle 

2 £ 


1 I 




Flashing Upsetting 


device, and a mechanism to compress the two workpieces against each other. Figure 
4.19 illustrates the different stages involved in a flash welding cycle. We can see that 
the pressure applied at the beginning is low. Therefore, there are a limited number of 
contact points that act as localized bridges for the electric current. Consequently, metal 
is heated at those points when the current is switched on, and the temperature increases 
with the increasing current until it exceeds the melting point of the metal. At this stage, 
the molten metal is expelled from the weld zone, causing "flashing." New bridges are 
formed and move quickly across the whole interface, resulting in uniform heating all 
over. When the whole contact area is heated above the liquidus line, electric current is 
switched off, and the pressure is suddenly increased to squeeze out the molten metal, 
upset the abutted parts, and weld them together. 

Flash welding is used for joining large sections, rails, chain links, tools, thin- 
walled tubes, and the like. It can also be employed for welding dissimilar metals. The 
advantages claimed for the process include its higher productivity and its ability to 
produce high-quality welds. The only disadvantage is the loss of some metal in 

Percussion welding. In percussion welding, a method of resistance welding, a high- 
intensity electric current is discharged between the parts before they are brought in 
solid contact. This results in an electric arc in the gap between the two surfaces. That 
electric arc lasts only for about 0.001 second and is enough to melt the surfaces to a 
depth of a few thousandths of an inch. The two parts are then impacted against each 
other at a high speed to obtain a sound joint. The major limitation of this process is 
the cross-sectional area of the welded joint. It should not exceed 0.5 square inch 
(300 mm 2 ) in order to keep the intensity of current required at a practical level. In in- 
dustry, percussion welding is limited to joining dissimilar metals that cannot be welded 

Spot welding. Figure 4.20a illustrates the principles of operation of spot welding, a re- 
sistance welding process. Electric current is switched on between the welding electrodes 


4 Joining of Metals 

FIGURE 4.20 

Resistance spot 
welding: (a) principles 
of operation; (b) a 
cross section through a 
spot weld 



Ac power 








to flow through the lapped sheets (workpieces) that are held together under pressure. As 
can be seen in Figure 4.20b, the metal fuses in the central area of the interface between 
the two sheets and then solidifies in the form of a nugget, thus providing the weld joint. 
Heat is also generated at the contact areas between the electrodes and the workpieces. 
Therefore, some precautions must be taken to prevent excessive temperatures and fus- 
ing of the metal at those spots. The electrodes used must possess good electrical and 
thermal conductivities. They are usually hollow and are water-cooled. In addition, areas 
of workpieces in contact with the electrodes must be cleaned carefully. 

Spot welding is the most widely used resistance welding process in industry. Car- 
bon steel sheets having a thickness up to 0. 1 25 inch (4 mm) can be successfully spot 
welded. Spot-welding machines have ratings up to more than 600 kVA and use a volt- 
age of 1 to 12 V obtained from a step-down transformer. Multispot machines are used, 
and the process can be fully automated. Therefore, spot welding has found widespread 
application in the automobile, aircraft, and electronics industries, as well as in sheet 
metal work. 

Seam welding. Seam welding and projection welding are modifications of spot weld- 
ing. In seam welding, the lapped sheets are passed between rotating circular electrodes 
through which the high-amperage current flows, as shown in Figure 4.21. Electrodes 
vary in diameter from less than 2 up to 14 inches (40 to 350 mm), depending upon the 
curvature of the workpieces being welded. Welding current as high as 5000 A may be 
employed, and the pressing force acting upon the electrodes can go up to 6 kN (more 
than half a ton). A welding speed of about 12 feet per minute (4 m/min.) is quite com- 
mon. Seam welding is employed in the production of pressure-tight joints used in con- 
tainers, tubes, mufflers, and the like. Advantages of this process include low cost, high 

4.2 Welding 


FIGURE 4.21 

Principles of seam 



Sheet metal 

Sheet metal 

Overlapping nuggets 

production rates, and suitability for automation. Nevertheless, the thickness of the 
sheets to be seam welded is limited to 0.125 inch (4 mm) in the case of carbon steels 
and much less for more conductive alloys due to the extremely high amperage required 
(0.125-inch-thick steel sheets require 19,000 A, whereas aluminum sheets having the 
same thickness would require 76,000 A). 

Projection welding. In projection welding, one of the workpieces is purposely pro- 
vided with small projections so that current flow and heating are localized at those 
spots. The projections are usually produced by die pressing, and the process calls for 
the use of a special upper electrode. Figure 4.22 illustrates an arrangement of two parts 
to be projection welded, as well as the resulting weld nugget. As you may expect, the 
projections collapse under the externally applied force after sufficient heating, thus 
yielding a well-defined, fused weld nugget. When the current is switched off, the weld 
cools down and solidification takes place under the applied force. The electrode force 
is then released, and the welded workpiece is removed. As is the case with spot weld- 
ing, the entire projection welding process takes only a fraction of a second. 

Projection welding has some advantages over conventional spot welding. For in- 
stance, sheets that are too thick to be joined by spot welding can be welded using this 
process. Also, the presence of grease, dirt, or oxide films on the surface of the work- 
pieces has less effect on the weld quality than in the case of spot welding. A further 

FIGURE 4.22 

An arrangement for 
projection welding two 


I 7 

k\\\\\\\\\\\\\\\\\\\^ ^,,,„.. r 





After welding 



4 Joining of Metals 

advantage of projection welding is the accuracy of locating welds inherent in that 

Fusion Welding Processes 

Fusion welding includes a group of processes that all produce welded joints as a result 
of localized heating of the edges of the base metal above its melting temperature, 
wherein coalescence is produced. A filler metal may or may not be added, and no ex- 
ternal pressure is required. The welded joint is obtained after solidification of the fused 
weld pool. 

In order to join two different metals together by fusion welding, they must possess 
some degree of mutual solubility in the solid state. In fact, metals that are completely 
soluble in the solid state exhibit the highest degree of weldability. Metals with limited 
solid solubility have lower weldability, and metals that are mutually insoluble in the 
solid state are completely unweldable by any of the fusion welding methods. In that 
case, an appropriate pressure welding technique should be employed. An alternative 
solution is to employ an intermediate metal that is soluble in both base metals. 

Metallurgy of fusion welding. Before surveying the different fusion welding 
processes, let us discuss the metallurgy of fusion welding. Important microstructural 
changes take place in and around the weld zone during and after the welding operation. 
Such changes in the microstructure determine the mechanical properties of the welded 
joint. Therefore, a study of the metallurgy of fusion welding is essential for good de- 
sign of welded joints, as well as for the optimization of the process parameters. 

During fusion welding, three zones can be identified, as shown in Figure 4.23, 
which indicates a single V-weld in steel after solidification and the corresponding tem- 
perature distribution during welding. In the first zone, called the fusion zone, the base 
metal and deposited metal (if a filler rod is used) are brought to the molten state dur- 
ing welding. Therefore, when this zone solidifies after welding, it generally has a 
columnar dendritic structure with haphazardly oriented grains. In other words, the mi- 
crostructure of this zone is quite similar to that of the cast metal. Nevertheless, if the 
molten metal is overheated during welding, this results in an acicular structure that is 
brittle, has low strength, and is referred to as the Widmanstatten structure. Also, the 

FIGURE 4.23 

The three zones in a 
fusion-welded joint and 
the temperature 
distribution during 



2700° F 


Fusion zone 

Parent metal 


4.2 Welding 99 

chemical composition of the fusion zone may change, depending upon the kind and 
amount of filler metal added. 

The second zone, which is referred to as the heat-affected zone (HAZ), is that por- 
tion of the base metal that has not been melted. Therefore, its chemical composition 
before and after welding remains unchanged. Nevertheless, its microstructure is al- 
ways altered because of the rapid heating during welding and subsequent cooling. In 
fact, the HAZ is subjected to a normalizing operation during welding and may conse- 
quently undergo phase transformations and precipitation reactions, depending upon the 
nature (chemical composition and microstructure) of the base metal. The size of the 
HAZ is dependent upon the welding method employed and the nature of the base 
metal. This can be exemplified by the fact that the HAZ is 0.1 inch (2.5 mm) when au- 
tomatic submerged arc welding is used, ranges from 0.2 to 0.4 inch (5 to 10 mm) for 
shielded-metal arc welding, and may reach 1 inch (25 mm) in conventional gas weld- 
ing. This evidently affects the microstructure of the weld, which is generally fine- 
grained. The effect of these structural changes on the mechanical properties of the 
weld differs for different base metals. For instance, the structural changes have negli- 
gible effect on the mechanical properties of low-carbon steel, regardless of the weld- 
ing method used. On the contrary, when welding high-carbon alloy steel, hardened 
structures like maternsite are formed in the HAZ of the weld that result in a sharp re- 
duction in the ductility of the welded joint and/or crack formation. (Remember the ef- 
fect of alloying elements on the critical cooling rate in the TTT diagram that you 
studied in metallurgy.) 

The third zone involves the unaffected parent metal adjacent to the HAZ that is 
subjected to a temperature below AC 3 (a critical temperature) during welding. In this 
zone, no structural changes take place unless the base metal has been subjected to plas- 
tic deformation prior to welding, in which case recrystallization and grain growth 
would become evident. 

Arc welding. Arc welding is based on the thermal effect of an electric arc that is act- 
ing as a powerful heat source to produce localized melting of the base metal. The elec- 
tric arc is, in fact, a sustained electrical discharge (of electrons and ions in opposite 
directions) through an ionized, gaseous path between two electrodes (i.e., the anode 
and the cathode). In order to ionize the air in the gap between the electrodes so that the 
electric arc can consequently be started, a certain voltage is required. (The voltage re- 
quired depends upon the distance between the electrodes.) The ionization process re- 
sults in the generation of electrons and positively charged ions. Next, the electrons 
impact on the anode, and the positively charged ions impact on the cathode. The col- 
lisions of these particles, which are accelerated by the arc voltage, transform the ki- 
netic energy of the particles into thermal and luminous energy, and the temperature at 
the center of the arc can reach as high as 11,000°F (6000°C). Actually, only a com- 
paratively low potential difference between the electrodes is required to start the arc. 
For instance, about 45 V is usually sufficient for direct current (dc) welding equipment, 
and up to 60 V for an alternating current (ac) welder. Also, the voltage drops after the 
arc is started, and a stable arc can then be maintained with a voltage in the range of 15 
to 30 V. Generally, arc welding involves using a metal electrode rod and attaching the 
other electrode to the workpiece. The electrode rod either melts during the process 


4 Joining of Metals 

(consumable electrode) and provides the necessary filler metal for the weld, or the 
electrode does not melt and the filler metal is separately provided. 

As just mentioned, either alternating current or direct current can be used in arc 
welding, although each has its distinct advantages. While arc stability is much better 
with alternating current than with direct current, the ac welding equipment is far less 
expensive, more compact in size, and simpler to operate. A further advantage of ac arc 
welding is the high efficiency of the transformer used, which goes up to 85 percent, 
whereas the efficiency of dc welding systems usually varies between only 30 and 
60 percent. 

In dc arc welding, the degree to which the work is heated can be regulated by 
using either straight or reversed polarity. As can be seen in Figure 4.24a and b, the 
cathode is the electrode rod and the anode is the workpiece in straight polarity (DCSP), 
whereas it is the other way around in reversed polarity (DCRP). When using DCSP, 
more heat is concentrated at the cathode (the electrode rod) than at the anode (the 
workpiece). Therefore, melting and deposition rates (of consumable electrodes) are 
high, but penetration in the workpiece is shallow and narrow. Consequently, DCSP is 
recommended when welding sheet metal, especially at higher welding speeds. With 
DCRP, heat is concentrated at the cathode (the workpiece) and results in deeper pene- 
tration for a given welding condition. It is, therefore, preferred for groove welds and 
similar applications. 

During the welding operation, heat is generated in the transformer as well as in 
other elements of the welding circuit, resulting in a temperature rise that may cause 
damage to those elements. There is, therefore, a time limitation when using the weld- 
ing equipment at a given amperage. That time limitation is usually referred to as the 
rated duty cycle. Consider the following numerical example. A power supply for arc 
welding rated at a 150- A 40-percent duty cycle means that it can be used only 40 
percent of the time when welding at 150 A. The idle or unused time is required to 
allow the equipment to cool down. The percentage of duty cycles at currents other than 
the rated current can be calculated using the following equation: 

% duty cycle 


rated current V 

\ load current / 

x rated duty cycle 


FIGURE 4.24 

Straight and reversed 
polarities in dc arc 
welding: (a) dc straight 
polarity (DCSP); (b) dc 
reversed polarity 









4.2 Welding 


Therefore, for this power supply, the percentage of the duty cycle at 100 A is as 

% duty cycle at 100 A = [ — V x 40% = 90% 

There are various types of arc welding. They include the following methods: 

1. Shielded-metal arc welding (SMAW) 

2. Carbon arc welding (CAW) 

3. Flux-cored arc welding (FCAW) 

4. Stud arc welding (SW) 

5. Submerged arc welding (SAW) 

6. Gas-metal arc welding (GMAW, usually called MIG) 

7. Gas-tungsten arc welding (GTAW, usually called TIG) 

8. Plasma arc welding (PAW) 

In addition, there is another welding process, electroslag welding (EW), which is not 
based on the phenomenon of the electric arc but, nevertheless, employs equipment 
similar to that used in gas-metal arc. flux-cored arc, or submerged arc welding. 

1. Shielded-metal arc welding. Shielded-metal arc welding (SMAW) is a manual arc 
welding process that is sometimes referred to as stick welding. The source of heat 
for welding is an electric arc maintained between a flux-covered, consumable metal 
electrode and the workpiece. As can be seen in Figure 4.25, which indicates the op- 
erating principles of this process, shielding of the electrode tip, weld puddle, and 
weld area in the base metal is ensured through the decomposition of the flux cov- 
ering. A blanket of molten slag also provides shielding for the molten-metal pool. 
The filler metal is provided mainly by the metal core of the electrode rod. 

Shielded-metal arc welding can be used for joining thin and thick sheets of 
plain-carbon steels, low-alloy steels, and even some alloy steels and cast iron, pro- 
vided that the electrode is properly selected and also that preheating and postheat- 
ing treatments are performed. It is actually the most commonly used welding 
process and has found widespread application in steel construction and shipbuild- 
ing. Nevertheless, it is uneconomical and/or impossible to employ shielded-metal 
arc welding to join some alloys, such as aluminum alloys, copper, nickel, copper- 
nickel alloys, and low-melting-point alloys such as zinc, tin, and magnesium alloys. 


4 Joining of Metals 

Another clear shortcoming of the process is that welding must be stopped each time 
an electrode stick is consumed to allow mounting a new one. This results in idle 
time and, consequently, a drop in productivity. 

The core wires of electrodes used for shielded-metal arc welding have many dif- 
ferent compositions. The selection of a particular electrode material depends upon 
the application for which it is going to be used and the kind of base metal to be 
welded. Consumable electrodes are usually coated with flux but can also be un- 
coated. The metal wire can have a diameter of up to 1 5/ 32 inch ( 1 2 mm) and a length 
of about 18 inches (450 mm). Although various metals are used as wire materials, by 
far the most commonly used electrode materials involve low-carbon steel (for weld- 
ing carbon steels) and low-alloy steel (for welding alloy steels). Electrodes can be 
bare, lightly coated, or heavily coated. The electrode covering, or coating, results in 
better-quality welds as it improves arc stability, produces gas shielding to prevent 
oxidation and nitrogen contamination, and also provides slag, which, in turn, re- 
tards the cooling rate of the weld's fusion zone. Therefore, electrode coatings are 
composed of substances that serve these purposes. Table 4.1 indicates the composi- 
tion of typical electrode coatings, together with the function of each constituent. 

2. Carbon arc welding. In carbon arc welding (CAW), nonconsumable electrodes 
made of carbon or graphite are used. Only a dc power supply can be employed, 
and the electric arc is established either between a single carbon electrode and 
the workpiece (Bernardos method) or between two carbon electrodes (independent 
arc method). In both cases, no shielding is provided. A filler metal may be used, 
especially when welding sheets with thicknesses more than 1/8 inch (3 mm). The 
carbon electrodes have diameters ranging from 3/8 to 1 inch (10 to 25 mm) and are 
used with currents that range from 200 to 600 A. 

TABLE 4.1 

The constituents of 
typical electrode 
coatings and their 

Main Function 



Gas generating 




Calcium carbonate 

Slag forming 

Titanium dioxide 





Sodium silicate 
Potassium silicate 





Arc stabilizing 

Potassium titanate 
Titanium oxide 


Increasing deposition rate 

Iron powder 


Improving weld strength 

Different alloying elements 


4.2 Welding 


Carbon arc welding is not commonly used in industry. Its application is limited 
to the joining of thin sheets of nonferrous metals and to brazing. 

3. Flux-cored arc welding. Flux-cored arc welding (FCAW) is an arc welding process 
in which the consumable electrode takes the form of a tubular, flux-filled wire, that 
is continuously fed from a spool. Shielding is usually provided by the gases evolv- 
ing during the combustion and decomposition of the flux contained within the 
tubular wire. The process is, therefore, sometimes called inner-shielded, or self- 
shielded, arc welding. Additional shielding may be acquired through the use of an 
auxiliary shielding gas, such as carbon dioxide, argon, or both. In the latter case, the 
process is a combination of the conventional flux-cored arc welding and the gas- 
metal arc welding methods and is referred to as dual-shielded arc welding. 

Flux-cored arc welding is generally applied on a semiautomatic basis, but it can 
also be fully automated. In that case, the process is normally used to weld medium- 
to-thick steel plates and stainless steel sheets. Figure 4.26 illustrates the operating 
principles of flux-cored arc welding. 

4. Stud arc welding. Stud arc welding (SW) is a special-purpose arc welding process 
by which studs are welded to flat surfaces. This facilitates fastening and handling 
of the components to which studs are joined and meanwhile eliminates the drill- 
ing and tapping operations that would have been required to achieve the same 
goal. Only dc power supplies are employed, and the process also calls for the 
use of a special welding gun that holds the stud during welding. Figure 4.27 shows 
the stages involved in stud arc welding, a process that is entirely controlled by the 
timer of the gun. As can be seen in the figure, shielding is accomplished through the 
use of a ceramic ferrule that surrounds the end of the stud during the process. Stud 
arc welding requires a low degree of welding skill, and the whole welding cycle 
usually takes less than a second. 

5. Submerged arc welding. Submerged arc welding (SAW) is a fairly new automatic 
arc welding method in which the arc and the weld area are shielded by a blanket of 
a fusible granular flux. A bare electrode is used and is continuously fed by a special 
mechanism during welding. Figure 4.28 shows the operating principles of sub- 
merged arc welding. As can be seen from the figure, the process is used to join flat 
plates in the horizontal position only. This limitation is imposed by the nature of the 
flux and the way it is fed. 

As is the case with previously discussed arc welding processes, gases evolve as 
a result of combustion and decomposition of the flux, due to the high temperature 

FIGURE 4.26 

Operating principles of 
flux-cored arc welding 


Electrode (tube) 
Flux core 

Base metal 


4 Joining of Metals 

FIGURE 4.27 

Stages involved in stud 
arc welding 







Base metal 

of the arc, and form a pocket, or gas bubble, around the arc. As Figure 4.29 shows, 
this gas bubble is sealed from the arc by a layer of molten flux. This isolates the arc 
from the surrounding atmosphere and, therefore, ensures proper shielding. 

The melting temperature of the flux must be lower than that of the base metal. 
As a result, the flux always solidifies after the metal, thus forming an insulating 
layer over the solidifying molten metal pool. This retards the solidification of the 
fused metal and, therefore, allows the slag and nonmetallic inclusions to float off 
the molten pool. The final outcome is always a weld that is free of nonmetallic in- 
clusions and entrapped gases and has a homogeneous chemical composition. The 
flux should also be selected to ensure proper deoxidizing of the fused metal and 

FIGURE 4.28 

Operating principles of 
submerged arc welding 

FIGURE 4.29 

The mechanics of 
shielding in submerged 
arc welding 

Hose from hopper 

to supply 

granulated flux 

Direction of weld 

Base meta 




weld metal 

Base metal 



4.2 Welding 


should contain additives that make up for the elements burned and lost during the 
welding process. 

Electric currents commonly used with submerged arc welding range between 
3000 and 4000 A. Consequently, the arc obtained is extremely powerful and is ca- 
pable of producing a large molten-metal pool as well as achieving deeper penetra- 
tion. Other advantages of this process include its high welding rate, which is five to 
ten times that produced by shielded-metal arc welding, and the high quality of the 
welds obtained. 

6. Gas-metal arc welding. The GMAW process is commonly called metal-inert-gas 
(MIG) welding. It employs an electric arc between a solid, continuous, consumable 
electrode and the workpiece. As can be seen in Figure 4.30, shielding is obtained 
by pumping a stream of chemically inert gas, such as argon or helium, around the 
arc to prevent the surrounding atmosphere from contaminating the molten metal. 
(The electrode is bare, and no flux is added.) Dry carbon dioxide can sometimes be 
employed as a shielding gas, yielding fairly good results. 

Gas-metal arc welding is generally a semiautomatic process. However, it can 
also be applied automatically by machine. In fact, welding robots and numerically 
controlled MIG welding machines have gained widespread industrial application. 
The gas-metal arc welding process can be used to weld thin sheets as well as rela- 
tively thick plates in all positions, and the process is particularly popular when 
welding nonferrous metals such as aluminum, magnesium, and titanium alloys. The 
process is also used for welding stainless steel and critical steel parts. 

The penetration for gas-metal arc welding is controlled by adopting DCRP and 
adjusting the current density. The higher the current density is, the greater the pen- 
etration is. The kind of shielding gas used also has some effect on the penetration. 
For instance, helium gives the maximum penetration; carbon dioxide, the least; 
argon, intermediate penetration. Thus, it is clear that higher current densities and the 

FIGURE 4.30 

Operating principles of 
gas-metal arc welding 


for feeding 

electrode wire 

Inert gas cylinder 
for providing 
shielding gas 

Base metal 


Joining of Metals 

appropriate shielding gas can be employed in welding thick plates, provided that the 
edges of these plates are properly prepared. 

The electrode wires used for MIG welding must possess close dimensional tol- 
erances and a consistent chemical composition appropriate for the desired applica- 
tion. The wire diameter varies between 0.02 and 0.125 inch (0.5 and 3 mm). 
Usually, MIG wire electrodes are coated with a very thin layer of copper to protect 
them during storage. The electrode wire is available in the form of a spool weigh- 
ing from 2V2 to 750 pounds (1 to 350 kg). As you may expect, the selection of the 
composition of the electrode wire for a given material depends upon other factors, 
such as the kind of shielding gas used, the conditions of the metal being welded 
(i.e., whether there is an oxide film, grease, or contaminants), and, finally, the re- 
quired properties of the weldment. 

Gas-tungsten arc welding. Gas-tungsten arc welding (GTAW), which is usually 
called tungsten-inert-gas (TIG) welding, is an arc welding process that employs the 
heat generated by an electric arc between a nonconsumable tungsten electrode and 
the workpiece. Figure 4.31 illustrates the operating principles of this process. As 
can be seen, a filler rod may (or may not) be fed to the arc zone. The electrode, arc, 
weld puddle, and adjacent areas of the base metal are shielded by a stream of either 
argon or helium to prevent any contamination from the atmosphere. TIG welding is 
normally applied manually and requires a relatively high degree of welder skill. It 
can also be fully automated, in which case the equipment used drives the welding 
torch at a preprogrammed path and speed, adjusts the arc voltage, and starts and 
stops it. 

Gas-tungsten arc welding is capable of welding nonferrous and exotic metals 
in all positions. The list of metals that can be readily welded by this process is 
long and includes alloy steels, stainless steels, heat-resisting alloys, refractory met- 
als, aluminum alloys, magnesium alloys, titanium alloys, copper and nickel alloys, 
and steel coated with low-melting-point alloys. The process is recommended for 

FIGURE 4.31 

Operating principles of 
gas-tungsten arc 

weld metal 

Torch electrode 



Filler rod 

The welding torch 

Base metal 



4.2 Welding 


welding very thin sheets, as thin as 0.005 inch (about 0.125 mm), for the root and 
hot pass on tubing and pipes, and wherever smooth, clean welds are required (e.g., 
in food-processing equipment). Ultrahigh-quality welds can be obtained in the nu- 
clear, rocketry, and submarine industries by employing a modified version of TIG 
welding that involves placing carefully selected and prepared inserts in the gap be- 
tween the sections to be joined and then completely fusing the inserts together with 
the edges of base metal using a TIG torch. 

All three types of current supplies (i.e., ac, DCSP, and DCRP) can be used with 
gas-tungsten arc welding, depending upon the metal to be welded. Thin sheets of 
aluminum or magnesium alloys are best welded by using DCRP, which prevents 
burn-throughs, as previously explained. Nevertheless, it is recommended that an ac 
power supply be used when welding normal sheets of aluminum and magnesium. 
DCSP is best suited for welding high-melting-point alloys such as alloy steels, 
stainless steels, heat-resisting alloys, copper alloys, nickel alloys, and titanium. In 
addition to these considerations, DCRP is also helpful in removing surface oxide 
films due to its cleaning action (the impacting of ions onto the surface like a grit 

8. Plasma arc welding. Figure 4.32 is a sketch of the torch employed in plasma arc 
welding (PAW). The electric arc can take either of two forms: a transferred arc that 
is a constricted arc between a tungsten electrode and the workpiece or a nontrans- 
ferred arc between the electrode and the constricting nozzle. The gas flowing 
around the arc heats up to extremely high temperatures like 60,000°F (33,000°C) 
and becomes, therefore, ionized and electrically conductive; it is then referred to as 
plasma. The main shielding is obtained from the hot ionized gas emerging from the 
nozzle. Additional inert-gas shielding can be used when high-quality welds are re- 
quired. In fact, plasma arc welding can be employed to join almost all metals in all 
positions, although it is usually applied to thinner metals. Generally, the process is 
applied manually and requires some degree of welder skill; however, the process is 
sometimes automated in order to increase productivity. 

Electroslag welding. Electroslag welding (ESW), which was developed by the Rus- 
sians, is not an arc welding process but requires the use of equipment similar to that 

FIGURE 4.32 

The torch employed in 
plasma arc welding 






4 Joining of Metals 

used in arc welding. Although an electric arc is used to start the process, heat is con- 
tinuously generated as a result of the current flow between the electrode (or electrodes) 
and the base metal through a pool of molten slag (flux). As we will see later, the 
molten-slag pool also serves as a protective cover for the fused-metal pool. 

The electroslag welding process is shown in Figure 4.33. As can be seen in the fig- 
ure, the parts to be joined are set in the vertical position, with a gap of 1/2 to l!/2 
inches (12 to 37 mm) between their edges. (The gap is dependent upon the thickness 
of the parts.) The welding electrode (or electrodes) and the flux are fed automatically 
into the gap, and an arc is established between the electrodes and the steel backing 
plate to provide the initial molten-metal and slag pools. Next, the electrical resistivity 
of the molten slag continuously produces the heat necessary to fuse the flux and the 
filler and the base metals. Water-cooled copper plates travel upward along the joint, 
thus serving as dams and cooling the fused metal in the cavity to form the weld. 

Electroslag welding is very advantageous in joining very thick parts together in a 
single pass without any need for beveling the edges of those parts. Therefore, the 
process is widely used in industries that fabricate beds and frames for heavy machin- 
ery, drums, boilers, and the like. 

Gas welding. Gas welding refers to a group of oxyfuel gas processes in which the 
edges of the parts to be welded are fused together by heating them with a flame ob- 
tained from the combustion of a gas (such as acetylene) in a stream of oxygen. A filler 
metal is often introduced into the flame to melt and, together with the base metal, form 
the weld puddle. Gas welding is usually applied manually and requires good welding 
skill. Common industrial applications involve welding thin-to-medium sheets and sec- 
tions of steels and nonferrous metals in all positions. Gas welding is also widely used 
in repair work and in restoring cracked or broken components. 

The fuel gases used for producing the flame during the different gas welding 
processes include acetylene, hydrogen, natural gas (94 percent methane), petroleum 
gas, and vaporized gasoline and kerosene. However, acetylene is the most commonly 
used gas for gas welding because it can provide a flame temperature of about 5700°F 
(3150°C). Unfortunately, acetylene is ignited at a temperature as low as 790°F (420°C) 
and becomes explosive in nature at pressures exceeding 1 .75 atmospheres. Therefore, 
it is stored in metal cylinders, in which it is dissolved in acetone under a pressure of 

FIGURE 4.33 

The electroslag welding 


Base metal 

Molten slag 
Molten metal 

Base plate 

4.2 Welding 


about 19 atmospheres. For more safety, acetylene cylinders are also filled with a 
porous filler (such as charcoal) in order to form a system of capillary vessels that are 
then saturated with the solution of acetylene in acetone. 

The oxygen required for the gas welding process is stored in steel cylinders in the 
liquid state under a pressure of about 150 atmospheres. It is usually prepared in spe- 
cial plants by liquefying air and then separating the oxygen from the nitrogen. 

The equipment required in gas welding, as shown in Figure 4.34, includes oxygen 
and acetylene cylinders, regulators, and the welding torch. The regulators serve to re- 
duce the pressure of the gas in the cylinder to the desired working value and keep it 
that way throughout the welding process. Thus, the proportion of the two gases is con- 
trolled, which determines the characteristics of the flame. Next, the welding torch 
serves to mix the oxygen and the acetylene together and discharges the mixture out at 
the tip, where combustion takes place. 

Depending upon the ratio of oxygen to acetylene, three types of flames can be ob- 
tained: neutral, reducing, and oxidizing. Figure 4.35 is a sketch of a typical oxyacety- 
lene welding flame. As can be seen, the welding flame consists of three zones: the 
inner luminous cone at the tip of the torch, the reducing zone, and the oxidizing zone. 

The first zone, the luminous cone, consists of partially decomposed acetylene as a 
result of the following reaction: 

C 2 H 2 -> 2C + H 2 

The carbon particles obtained are incandescent and are responsible for the white lumi- 
nescence of that brightest part of the flame. Those carbon particles are partly oxidized 
in the second zone, the reducing zone, yielding carbon monoxide and a large amount 
of heat that brings the temperature up to about 5400°F (3000°C). Gases like hydrogen 
and carbon monoxide are capable of reducing oxides. Next, complete combustion of 
those gases yields carbon dioxide and water vapor that together with the excess oxy- 
gen (if any) result in the third zone, the oxidizing zone. Those gases, however, form a 
shield that prevents the atmosphere from coming in contact with the molten-metal 

As can be expected, the extent (as well as the appearance) of each of the zones 
depends upon the type of flame (i.e., the oxygen-to-acetylene ratio). When the ratio 
is about 1, the flame is neutral and distinctively has the three zones just outlined. If 
the oxygen-to-acetylene ratio is less than 1, a reducing, or carbonizing, flame is ob- 
tained. In this case, the luminous cone is longer than that obtained with the neutral 

FIGURE 4.34 

The equipment required 
in gas welding 




4 Joining of Metals 

FIGURE 4.35 

A sketch of a typical 
oxyacetylene welding 

flame, and the outline of the flame is not sharp. This type of flame is employed in 
welding cast iron and in hard-surfacing with high-speed steel and cemented carbides. 
The third type of flame, the oxidizing flame, is obtained when the oxygen-to-acety- 
lene ratio is higher than 1. In this case, the luminous cone is shorter than that ob- 
tained with the neutral flame, and the flame becomes light blue in color. The 
oxidizing flame is employed in welding brass, bronze, and other metals that have 
great affinity to hydrogen. 

Another method that utilizes the heat generated as a result of the combustion of 
a fuel gas is known as pressure-gas welding. As the name suggests, this is actually a 
pressure welding process in which the abutting edges to be welded are heated with 
an oxyacetylene flame to attain a plastic state; then, coalescence is achieved by ap- 
plying the appropriately high pressure. In order to ensure uniform heating of the sec- 
tions, a multiple-flame torch that surrounds the sections is used. The shape of that 
torch is dependent upon the outer contour of the sections to be welded, and the torch 
is usually made to oscillate along its axis. Upsetting is accomplished by a special 
pressure mechanism. This method is sometimes used for joining pipeline mains, rails, 
and the like. 

Electron-beam welding. Electron-beam welding (EBW) was developed by Dr. 
Jacques Stohr (CEA-France, the atomic energy commission) in 1957 to solve a prob- 
lem in the manufacturing of fuel elements for atomic power generators. The process is 
based upon the conversion of the kinetic energy of a high-velocity, intense beam of 
electrons into thermal energy as the accelerated electrons impact on the joint to be 
welded. The generated heat then fuses the interfacing surfaces and produces the de- 
sired coalescence. 

Figure 4.36 shows the basic elements and working principles of an electron-beam 
welding system. The system consists of an electron-beam gun (simply an electron 
emitter such as a hot filament) that is electrically placed at a negative potential with re- 
spect to an anode and that together with the workpiece is earth-grounded. A focus coil 
(i.e., an electromagnetic lens) is located slightly below the anode in order to bring the 
electron beam into focus upon the work. This is achieved by adjusting the current of 
the focus coil. Additional electromagnetic coils are provided to deflect the beam from 
its neutral axis as required. Because the electrons impacting the work travel at an ultra- 
high velocity, the process should be carried out in a vacuum in order to eliminate any 
resistance to the traveling electrons. Pressures on the order of 10 torr (1 atmosphere = 
760 torr) are commonly employed, although pressures up to almost atmospheric can be 
used. Nevertheless, it must be noted that the higher the pressure is, the wider and more 
dispersed the electron beam becomes, and the lower the energy density is. (Energy 
density is the number of kilowatts per unit area of the spot being welded.) 

4.2 Welding 


FIGURE 4.36 

The basic elements and 
working principles of an 
electron-beam welding 



~ voltage 
+ control 

Electron-beam welding machines can be divided into two groups: low-voltage and 
high-voltage machines. Low-voltage machines are those operating at accelerating volt- 
ages up to 60 kV, whereas high-voltage machines operate at voltages up to 200 kV. Al- 
though each of these two types has its own merits, the main consideration should be 
the beam-power density, which is, in turn, dependent upon the beam power and the 
(focused) spot size. In the early days of electron-beam welding, machines were usually 
built to have a rating of 7.5 kW and less. Today, a continuous-duty rating of 60 kW is 
quite common, and the trend is toward still higher ratings. 

There are several advantages to the electron-beam welding process. They include 
the following five: 

1. Because of the high intensity of the electron beam used, the welds obtained are 
much narrower, and the penetration in a single pass is much greater than that ob- 
tained by conventional fusion welding processes. 

2. The high intensity of the electron beam can also develop and maintain a bore- 
hole in the workpiece, thus yielding a parallel-sided weld with a very narrow heat- 
affected zone. As a consequence, the welds produced by this method have almost 
no distortion, have minimum shrinkage, and are stronger than welds produced by 
conventional fusion welding processes. 

3. Because parallel-sided welds are obtained by this process, there is no need for edge 
preparation of the workpieces (such as V- or J-grooves). Square butt-type joints are 
commonly produced by electron-beam welding. 

4. High welding speeds can be obtained with this process. Speeds up to 200 inches per 
minute (0.09 m/s) are common, resulting in higher productivity. 

5. Because the process is usually performed in a vacuum chamber at pressures on the 
order of 10 torr, the resulting weld is excellent, is metallurgically clean, and has 
an extremely low level of atmospheric contamination. Therefore, electron-beam 


4 Joining of Metals 

welding is especially attractive for joining refractory metals whose properties are 
detrimentally affected by even low levels of contamination. 

Because of the ultrahigh quality of the joints produced by electron-beam welding, 
the process has found widespread use in the atomic power, jet engine, aircraft, and 
aerospace industries. Nevertheless, the time required to vacuum the chamber before 
each welding operation results in reduced productivity, and, therefore, the high cost of 
the electron-beam welding equipment is not easily justified. This apparently kept the 
process from being applied in other industries until it was automated. Today, electron- 
beam welding is becoming popular for joining automotive parts such as gear clusters, 
valves, clutch plates, and transmission components. 

Laser-beam welding. The term laser stands for light amplified by stimulated emission 
of radiation. It is, therefore, easy to see that a laser beam is actually a controlled, in- 
tense, highly collimated, and coherent beam of light. In fact, a laser beam proved to be 
a unique source of high-intensity energy that can be used in fusing metals to produce 
welded joints having very high strength. 

Figure 4.37 shows the working principles of laser-beam welding (LBW). In this 
laser system, energy is pumped into a laser medium to cause it to fluoresce. This fluo- 
rescence, which has a single wavelength or color, is trapped in the laser medium (laser 
tube) between two mirrors. Consequently, it is reflected back and forth in an optical 
resonator path, resulting in more fluorescence, which amplifies the intensity of the 
light beam. The amplified light (i.e., the laser beam) finds its way out through the 
partly transparent mirror, which is called the output mirror. The laser medium can be 
a solid, such as a crystal made of yttrium aluminum garnet (YAG). It can also be a gas, 
such as carbon dioxide, helium, or neon. In the latter case, the pumping energy input 
is usually introduced directly by electric current flow. 

Let us now consider the mechanics of laser-beam welding. The energy intensity of 
a laser beam is not high enough to fuse a metal such as steel or copper. Therefore, it 
must be focused by a highly transparent lens at a very tiny spot, 0.01 inch (0.25 mm) 
in diameter, in order to increase the intensity of energy up to a level of 10 million W 
per square inch (15,500 W/mm") at the focal point. The impacting laser energy is 

FIGURE 4.37 

The working principles 
of a laser-beam welding 

energy input 

Laser media 






Output mirror 
(partially transparent) 




4.2 Welding 113 

converted into heat as it strikes the surface of a metal, causing instantaneous fusion of 
the metal at the focal point. Next, a cylindrical cavity, known as a keyhole, that is full 
of vaporized, ionized metallic gas is formed and is surrounded by a narrow region of 
molten metal. As the beam moves relative to the workpiece, the molten metal fills be- 
hind the keyhole and subsequently cools and solidifies to form the weld. It is worth 
mentioning that a stream of a cooling (and shielding) gas should surround the laser 
beam to protect the focusing lens from vaporized metal. Usually, argon is used for this 
purpose because of its low cost, although helium is actually the best cooling gas. 

In spite of the high initial capital cost required, laser-beam welding has gained 
widespread industrial application because of several advantages that the process pos- 
sesses. Among these advantages are the following six: 

1. Based on the preceding discussion of the mechanics of laser-beam welding, we 
would always expect to have a very narrow heat-affected zone with this welding 
method. Consequently, the chemical, physical, and mechanical properties of the 
base metal are not altered, thus eliminating the need for any postwelding heat 

2. The ultrahigh intensity of energy of the laser beam at the focal point allows metals 
having high melting points (refractory metals) to be welded. 

3. The process can be successfully used to weld both nonconductive as well as mag- 
netic materials that are almost impossible to join even with electron-beam welding. 

4. The laser beam can be focused into a chamber through highly transparent windows, 
thus rendering laser-beam welding suitable for joining radioactive materials and for 
welding under sterilized conditions. 

5. The process can be used for welding some materials that have always been consid- 
ered unweldable. 

6. The process can be easily automated. Numerically controlled laser-beam welding 
systems are quite common and are capable of welding along a complex contour. 

Since the Apollo project, laser-beam welding has become popular in the aerospace 
industry. Today, the process is mainly employed for joining exotic metals such as tita- 
nium, tantalum, zirconium, columbium, and tungsten. The process is especially advan- 
tageous for making miniature joints as in tiny pacemaker cans, integrated-circuit 
packs, camera parts, and batteries for digital watches. Nevertheless, laser-beam weld- 
ing is not recommended for joining brass, zinc, silver, gold, or galvanized steel. 

Welding Defects 

In fusion welding processes, considerable thermal stresses develop during heating and 
subsequent cooling of the workpiece, especially with those processes that result in 
large heat-affected zones. Also, metallurgical changes and structural transformations 
take place in the weld puddle as well as in the heat-affected zone, and these may be 
accompanied by changes in the volume. Therefore, if no precautions are taken, defects 

114 4 Joining of Metals 

that are damaging to the function of the weldment may be generated. It is the com- 
bined duty of the manufacturing engineer, the welder, and the inspector to make sure 
that all weldments are free from all kinds of defects. Following is a brief survey of the 
common kinds of welding defects. 

Distortion. Distortion, warping, and buckling of the welded parts are welding defects 
involving deformation (which can be plastic) of the structures as a result of residual 
stresses. They come as a result of restraining the free movement of some parts or mem- 
bers of the welded structure. They can also result from nonuniform expansion and 
shrinkage of the metal in the weld area as a consequence of uneven heating and cool- 
ing. Although it is possible to predict the magnitude of the residual stresses in some 
simple cases (e.g., butt welding of two plates), an analysis to predict the magnitude 
of these stresses and to eliminate distortion in the common case of a welded three- 
dimensional structure is extremely complicated. Nevertheless, here are some recom- 
mendations and guidelines to follow to eliminate distortion: 

1. Preheat the workpieces to a temperature dependent on the properties of the base 
metal in order to reduce the temperature gradient. 

2. Clamp the various elements (to be welded) in a specially designed rigid welding 
fixture. Although no distortion occurs with this method, there are always inher- 
ent internal stresses. The internal stresses can be eliminated by subsequent 
stress-relieving heat treatment. 

3. Sometimes, it is adequate just to tack-weld the elements securely in the right posi- 
tion (relative to each other) before actual-strength welds are applied. It is also ad- 
visable to start by welding the section least subject to distortion first in order to 
form a rigid skeleton that contributes to the balance of assembly. 

4. Create a rational design of weldments (e.g., apply braces to sections most likely to 

Porosity. Porosity can take the form of elongated blowholes in the weld puddle, 
which is known as wormhole porosity, or of scattered tiny spherical holes. In both 
cases, porosity is due mainly to either the evolution of gases during welding or the re- 
lease of gases during solidification as a result of their decreasing solubility in the so- 
lidifying metal. Excess sulfur or sulfide inclusions in steels are major contributors to 
porosity because they generate gases that are often entrapped in the molten metal. 
Other causes of porosity include the presence of hydrogen (remember the problem 
caused by hydrogen in casting), contamination of the joint, and contaminants in the 
flux. Porosity can be eliminated by maintaining clean workpiece surfaces, by properly 
conditioning the electrodes, by reducing welding speed, by eliminating any moisture 
on workpieces, and, most importantly, by avoiding the use of a base metal containing 
sulfur or electrodes with traces of hydrogen. 

Cracks. Welding cracks can be divided into two main groups: fusion zone cracks 
and heat-affected zone cracks. The first group includes longitudinal and transverse 
cracks as well as cracks appearing at the root of the weld bead. This type of cracking 

4.2 Welding 


is sometimes called hot cracking because it occurs at elevated temperatures just 
after the molten metal starts to solidify. It is especially prevalent in ferrous alloys with 
high percentages of sulfur and phosphorus and in alloys having large solidification 

The second type of cracking, heat-affected zone cracks, is also called cold crack- 
ing. This defect is actually due to aggravation by excessive brittleness of the heat- 
affected zone that can be caused by hydrogen embrittlement or by martensite 
formation as a result of rapid cooling, especially in high-carbon and alloy-steel welded 
joints. (Remember the effect of alloying elements on the TTT curve; they shift it to the 
right, thus decreasing the critical cooling rate.) Cold cracks can be eliminated by using 
a minimum potential source of hydrogen and by controlling the cooling rate of the 
welded joint to keep it at a minimum (e.g., keep joints in a furnace after welding or 
embed them in sand). 

The use of multiple passes in welding can sometimes eliminate the need for 
prewelding or postwelding heat treatment. Each pass would provide a sort of preheat- 
ing for the pass to follow. This technique is often effective in the prevention of weld 

Slag inclusions. Slag entrapment in the weld zone can occur in single-pass as well as 
in multipass welds. In single-pass arc welding, slag inclusions are caused by improper 
manipulation of the electrode and/or factors such as too high a viscosity of the molten 
metal or too rapid solidification. Some slag pushed ahead of the arc is drawn down by 
turbulence into the molten-metal pool, where it becomes entrapped in the solidifying 
weld metal. In multipass welds, slag inclusions are caused by improper removal of the 
slag blanket after each pass. 

Lack of fusion. Lack of fusion, shown in Figure 4.38, can result from a number of 
causes. These include inadequate energy input, which leads to insufficient temperature 
rise; improper electrode manipulation; and failure to remove oxide films and clean the 
weld area prior to welding. 

Lack of penetration. Lack of penetration, shown in Figure 4.39, is due to a low en- 
ergy input, the wrong polarity, or a high welding speed. 

Undercutting. Undercutting, shown in Figure 4.40, is a result of a high energy input 
(excessive current in arc welding), which, in turn, causes the formation of a recess. As 
we know, such sharp changes in the weld contour act as stress raisers and often cause 
premature failure. 

FIGURE 4.38 

Lack of fusion 

FIGURE 4.39 

Lack of penetration 



4 Joining of Metals 

FIGURE 4.40 


FIGURE 4.41 


Underfilling. Underfilling, shown in Figure 4.41, involves a depression in the weld 
face below the surface of the adjoining base metal. More filler metal has to be added 
in order to prevent this defect. 

Testing and Inspection of Welds 

Welds must be evaluated by being subjected to testing according to codes and specifi- 
cations that are different for different countries. The various types of tests can be di- 
vided into two groups: destructive and nondestructive. Destructive testing always 
results in destroying the specimen (the welded joint) and rendering it unsuitable for its 
design function. Destructive tests can be mechanical, metallurgical, or chemical. We 
next review various destructive and nondestructive testing methods. 

Visual inspection. Visual inspection involves examination of the weld by the naked 
eye and checking its dimensions by employing special gages. Defects such as cracks, 
porosity, undercuts, underfills, or overlaps can be revealed by this technique. 

Mechanical tests. Mechanical tests are generally similar to the conventional me- 
chanical tests, the difference being the shape and size of the test specimen. Tensile, 
bending, impact, and hardness tests are carried out. Such tests are conducted either on 
the whole welded joint or on the deposited metal only. 

Metallurgical tests. Metallurgical tests involve metallurgical microstructure and 
macrostructure examination of specimens. Macrostructure examination reveals the 
depth of penetration, the extent of the heat-affected zone, and the weld bead shape, as 
well as hidden cracks, porosity, and slag inclusions. Microstructure examination can 
show the presence of nitrides, martensite, or other structures that cause metallurgically 
oriented welding problems. 

Chemical tests. Chemical tests are carried out to ensure that the composition of the 
filler metal is identical to that specified by the manufacturing engineer. Some are crude 
tests, such as spark analysis or reagent analysis; however, if accurate data are required, 
chemical analysis or spectrographic testing must be carried out. 

Radiographic inspection. Radiographic inspection is usually performed by employ- 
ing industrial X rays. This technique can reveal hidden porosity, cracks, and slag in- 
clusions. It is a nondestructive test that does not destroy the welded joint. 
High-penetration X rays are sometimes also employed for inspecting weldments hav- 
ing thicknesses up to l!/2 inches (37 mm). 

4.2 Welding 


Pressure test. Hydraulic (or air) pressure is applied to welded conduits that are 
going to be subjected to pressure during their service lives to check their tightness and 

Ultrasonic testing. Ultrasonic waves with frequencies over 20 kHz are employed to 
detect various kinds of flaws in the weld, such as the presence of nonmetallic inclu- 
sions, porosity, and voids. This method is reliable even for testing very thick parts. 

Magnetic testing. As we know from physics, the lines of magnetic flux are distorted 
in such a way as to be concentrated at the sides of a flaw or a discontinuity, as seen in 
Figure 4.42a and b. This test, therefore, involves magnetizing the part and then using 
fine iron-powder particles that were uniformly dispersed on the surface of the part to 
reveal the concentration of the flux lines at the location of the flaw. This method is suc- 
cessful in detecting superficial hair cracks and pores in ferrous metal. 

Ammonia penetrant test. The ammonia penetrant test is used to detect any leakage 
from welded vessels. It involves filling the vessel with a mixture of compressed air and 
ammonia and then wrapping it with paper that has been impregnated in a 5 percent so- 
lution of mercuric nitrate. Any formation of black spots is an indication of leakage. 

Fluorescent penetrant test. The part is immersed for about half an hour in oil (or 
an oil mixture) and then dipped in magnesia powder. The powder adheres at any 
crack location. 

Design Considerations 

As soon as the decision is made to fabricate a product by welding, the next step is to 
decide which welding process to use. This decision should be followed by selection of 
the types of joints, by determination of the locations and distribution of the welds, and, 
finally, by making the design of each joint. Following is a brief discussion of the fac- 
tors to be considered in each design stage. 

Selection of the joint type. We have previously discussed the various joint designs 
and realized that the type of joint depends upon the thickness of the parts to be welded. 
In fact, there are other factors that should also affect the process of selecting a partic- 
ular type of joint. For instance, the magnitude of the load to which the joint is going 

FIGURE 4.42 

Magnetic testing of 
welds: (a) defective 
weld; (b) sound weld 




Joining of Metals 

to be subjected during its service life is one other important factor. The manner in 
which the load is applied (i.e., impact, steady, or fluctuating) is another factor. Whereas 
the square butt, simple- V, double-V, and simple-U butt joints are suitable only for 
usual loading conditions, the double-U butt joint is recommended for all loading con- 
ditions. On the other hand, the square-T joint is appropriate for carrying longitudinal 
shear under steady-state conditions. When severe longitudinal or transverse loads are 
anticipated, other types of joints (e.g., the single-bevel-T, the double-bevel-T, and the 
double-J) have to be considered. In all cases, it is obvious that cost is the decisive fac- 
tor whenever there is a choice between two types of joints that would function equally 

Location and distribution of welds. It has been found that the direction of the linear 
dimension of the weld with respect to the direction of the applied load has an effect on 
the strength of the weld. In fact, it has been theoretically and experimentally proven 
that a lap weld whose linear direction is normal to the direction of the applied load, as 
is shown in Figure 4.43a, is 30 percent stronger than a lap weld whose linear direction 
is parallel to the direction of the applied load, as shown in Figure 4.43b. In the first 
case, the maximum force F that the joint can carry without any signs of failure can be 
approximated by the following equation: 

F = 0.707 Xhfx G allowable (4.3) 

where: £ is the weld leg 

W is the length of the weld 

^allowable is the allowable tensile stress of the filler material (e.g., 

In the second case (Figure 4.43b), the strength of the joint is based on the fact that the 
throat plane of the weld is subjected to pure shear stress and is given by the following 

F = 0.707 x € x W x x allowable 

where: € is the weld leg 

W is the length of the weld 

^allowable ' s the allowable shear stress of the electrode 


FIGURE 4.43 

Location and 
distribution of welds: 
(a) weld linear direction 
normal to the applied 
load; (b) weld linear 
direction parallel to the 
applied load 





4.2 Welding 


From the theory of plasticity, assuming you adopt the same safety factor in both cases, 
it is easy to prove that 


^allowable — / ^allowable 

= 0.565 G : , 


On the other hand, the strength of a butt-welded joint can be given by the following 

F = € x W x a a 


where t, W, and a allowable are as previously mentioned. A product designer should, 
therefore, make use of this characteristic when planning the location and distribution 
of welds. 

Another important point to consider is the prevention of any tendency of the 
welded elements to rotate when subjected to mechanical loads. A complete force 
analysis must be carried out in order to determine the proper length of each weld. Let 
us now consider a practical example to see the cause and the remedy for this tendency 
to rotate. Figure 4.44 shows an L angle welded to a plate. Any load applied through the 
angle will pass through its center of gravity. Therefore, the resisting forces that act 
through the welds will not be equal; the force closer to the center of gravity of the 
angle will always be larger. Consequently, if any tendency to rotate is to be prevented, 
the weld that is closer to the center of gravity must be longer than the other one. Using 
simple statics, it can easily be seen that 

W 2 


It is also recommended that very long welds be avoided. It has been found that two 
small welds, for example, are much more effective than a single long weld. 

Joint design. In addition to the procedures and rules adopted in common design prac- 
tice, there are some guidelines that apply to joint design: 

FIGURE 4.44 

Preventing the tendency 
of the welded element 
to rotate by appropriate 
distribution of welds 




■><■ 7' 



Center of 

Center of gravity 
-of the angle 


4 Joining of Metals 

FIGURE 4.45 

Designs that promote 
or eliminate distortion 
in welding: 

(a) distortion caused by 
unbalanced weld; 

(b) and (c) methods for 
reducing distortion 




1. Try to ensure accessibility to the locations where welds are to be applied. 

2. Try to avoid overhead welding. 

3. Consider the heating effect on the base metal during the welding operation. Balance 
the welds to minimize distortion. Use short, intermittent welds. Figure 4.45a shows 
distortion caused by an unbalanced weld, whereas Figure 4.45b and c shows meth- 
ods for reducing that distortion. 

4. Avoid crevices around welds in tanks as well as grooves (and the like) that would 
allow dirt to accumulate. Failure to do so may result in corrosion in the welded 

5. Do not employ welding to join steels with high hardenability. 

6. Do not weld low-carbon steels to alloy steels by the conventional fusion welding 
methods because they have different critical cooling rates and hence cannot be suc- 
cessfully welded. 

7. When employing automatic welding (e.g., submerged arc), the conventional joint 
design of manual welding should be changed. Wider Vs (for butt joints) are used, 
and single-pass welds replace multipass welds. 


Surfacing involves the application of a thin deposit on the surface of a metallic work- 
piece by employing a welding method such as oxyacetylene-gas welding, shielded- 
metal arc welding, or automatic arc welding. The process is carried out to increase the 
strength, the hardness, and the resistance to corrosion, abrasion, or wear. For the last 
reason, the process is commonly known as hard-facing. 

Good hard-facing practice should be aimed at achieving a strong bond between 
the deposit and the base metal and also at preventing the formation of cracks and other 
defects in the deposited layer. Therefore, the deposited layer should not generally ex- 
ceed 3/32 inch (2 mm) and will rarely exceed 1/4 inch (6 mm). Also, the base metal 

4.4 Thermal Cutting of Metals 121 

should be heated to a temperature of 500°F to 950°F (350°C to 500°C) to ensure a 
good metallurgical bond and to allow the deposited layer to cool down slowly. 

Hard-facing permits the use of very hard wear- and corrosion-resisting com- 
pounds. The materials used in this process are complex. They involve hard com- 
pounds, like carbides and borides, that serve as the wear-resisting elements, and a 
tough matrix composed of air-hardening steel or iron-base alloys. Such deposited ma- 
terials increase the service life of a part three- or fourfold. The process is also em- 
ployed in restoring worn parts. 

The process of hard-facing has found widespread application in the heavy con- 
struction equipment industry, in mining, in agricultural machinery, and in the petro- 
leum industry. The list of parts that are usually hard-faced is long and includes, for 
example, the vulnerable surfaces of chemical-process vessels, pump liners, valve seats, 
drive sprockets, ripper teeth, shovel teeth, chutes, and the edges of coal recovery 


In this section, we discuss the thermal cutting of metals, specifically oxyfuel flame cut- 
ting and the different arc cutting processes. Although all these processes do not, by any 
means, fall under the topic of joining (the action involved is opposite to that of join- 
ing), they employ the same equipment as the corresponding welding process in each 
case. The thermal cutting processes are not alternatives to sawing but rather are used 
for cutting thick plates, 1 to 10 inches (25 to 250 mm) thick, as well as for difficult-to- 
machine materials. Thermal cutting may be manual, using a hand-operated cutting 
torch (or electrode), or the cutting element can be machine driven by a numerically 
controlled system or by special machines called radiographs. 

Oxyfuel Cutting 

Oxyfuel cutting (OFC) is similar to oxyfuel welding except that an oxidizing flame 
must always be used. The process is extensively used with ferrous metal having thick- 
nesses up to 10 inches (250 mm). During the process, red-hot iron, directly subjected 
to the flame, is oxidized by the extra oxygen in the flame; it then burns up, leaving just 
ashes or slag. Also, the stream of burning gases washes away any molten metal in the 
region being cut. Generally, there is a relationship between the speed of travel of the 
torch or electrode and the smoothness of the cut edge: The higher the speed of travel, 
the coarser the cut edge. 

Although acetylene is commonly used as a fuel in this process, other gases are 
also employed, including butane, methane, propane, natural gas, and a newly devel- 
oped gas with the commercial name Mapp. Hydrogen is sometimes used as a fuel, es- 
pecially underwater to provide a powerful preheating flame. In this case, compressed 
air is used to keep water away from the flame. 

The oxyfuel cutting process can be successfully employed only when the ignition 
temperature of the metal being cut is lower than its melting point. Another condition for 
the successful application of the process involves ensuring that the melting points of the 

122 4 Joining of Metals 

formed oxides are lower than that of the base metal itself. Therefore, oxyfuel cutting is 
not recommended for cast iron because its ignition temperature is higher than its melt- 
ing point. The process is also not appropriate for cutting stainless steel, high-alloy 
chromium and chrome-nickel alloys, and nonferrous alloys because the melting points 
of the oxides of these metals are higher than the melting points of the metals themselves. 

Arc Cutting 

There are several processes based upon utilization of the heat generated by an electric 
arc. These arc cutting processes are generally employed for cutting nonferrous metals, 
medium-carbon steel, and stainless steel. 

Conventional arc cutting. Conventional arc cutting is similar to shielded-metal arc 
welding. It should always be remembered, however, that the electrode enters the gap 
of the cut, so the coating must serve as an insulator to keep the electric arc from short- 
ing out. Consequently, electrodes with coatings containing iron powder are not recom- 
mended for use with this process. 

Air arc cutting. Air arc cutting involves preheating the metal to be cut by an electric 
arc and blowing out the resulting molten metal by a stream of compressed air. The arc- 
air torch is actually a steel tube through which compressed air is blown. 

Oxygen arc cutting. Oxygen arc cutting is similar to air arc cutting except that oxy- 
gen is blown instead of air. The process is capable of cutting cast irons and stainless 
steels with thicknesses up to 2 inches (50 mm). 

Carbon arc cutting. In carbon arc cutting, a carbon or graphite electrode is used. The 
process has the disadvantage of consuming that electrode quickly, especially if contin- 
uous cutting is carried out. 

Tungsten arc cutting. The electrode used in tungsten arc cutting is made of tungsten 
and has, therefore, a service life that is far longer than that of the carbon or graphite 
electrodes. Tungsten arc cutting is commonly employed for stainless steel, copper, 
magnesium, and aluminum. 

Air-carbon arc cutting. Air-carbon arc cutting is quite similar to carbon arc cutting, 
the difference being the use of a stream of compressed air to blow the molten metal 
(that has been fused by the arc) out of the kerf (groove). The process cuts almost all 
metals because its mechanics involve oxidation of the metal. Its applications involve 
removal of welds, removal of defective welds, and dismantling of steel structures. 

Plasma arc cutting. A plasma arc is employed to cut metals in plasma arc cutting 
(PAC). The temperature of the plasma jet is extremely high (ten times higher than that 
obtained with oxyfuel), thus enabling high-speed cutting rates to be achieved. Also, as 
a consequence, the heat-affected zone formed along the edge of the kerf is usually less 
than 0.05 inch (1.3 mm). Plasma arc cutting can be used for cutting stainless steel as 
well as hard-to-cut alloys. A modification of the process involves using a special noz- 
zle to generate a whirlpool of water on the workpiece, thus increasing the limit on the 

4.5 Brazing and Soldering 


FIGURE 4.46 

Laser-beam cutting of 
sheets and plates 


Material— ^^^ ^^^ 

Molten material 

thickness of the workpiece up to 3 inches (75 mm) and meanwhile improving the qual- 
ity of the cut. The only limitation on plasma arc cutting is that the workpiece must be 
electrically conductive. 

Laser-beam cutting. The basic principles of laser-beam cutting are similar to those of 
laser-beam welding. Nevertheless, laser cutting is achieved by the pressure from a jet 
of gas that is coaxial with the laser beam, as shown in Figure 4.46. The function of the 
gas jet is to blow away the molten metal that has been fused by the laser beam. Laser 
beams can be employed in cutting almost any material, including nonconductive poly- 
mers and ceramics. Also, the process is usually automated by using computerized nu- 
merical control systems to control the movements of the machine table under the laser 
beam so that workpieces can be cut to any desired contour. Other advantages of the 
laser-beam cutting process include the straight-edged kerfs obtained, the very narrow 
heat-affected zone that results, and the elimination of the part distortion experienced 
with other conventional thermal cutting processes. 


Brazing and soldering are processes employed for joining solid metal components by 
heating them to the proper temperature and then introducing between them a molten 
filler alloy (brazing metal or solder). The filler alloy must always have a melting point 
lower than that of the base metal of the components. The filler alloy must also possess 
high fluidity and wettability (i.e., be able to spread and adhere to the surface of the 
base metal). As you may expect, the mechanics of brazing or soldering are different 
from those of welding. A strong brazed joint is obtained only if the brazing metal can 
diffuse into the base metal and form a solid solution with it. Figure 4.47a and b is a 
sketch of the microstructure of two brazed joints and is aimed at clarifying the me- 
chanics of brazing and soldering. 

Brazing and soldering can be employed to join carbon and alloy steels, nonfer- 
rous alloys, and dissimilar metals. The parts to be joined together must be carefully 
cleaned, degreased, and clamped. Appropriate flux is applied to remove any re- 
maining oxide and to prevent any further oxidation of the metals. It is only under 


4 Joining of Metals 

FIGURE 4.47 

A sketch of the 
micro-structure of two 
brazed joints: (a) gold 
base metal; (b) low- 
carbon steel base 


along grain 



such conditions that the filler metal can form a strong metallic bond with the base 

The main difference between soldering and brazing is the melting point of the 
filler metal in each case. Soft solders used in soldering have melting points below 
930°F (500°C) and produce joints with relatively low mechanical strength, whereas 
hard solders (brazing metals) have higher melting points, up to 1650°F (900°C), and 
produce joints with high mechanical strength. 

Soft solders are low-melting-point eutectic alloys. They are basically tin, lead, 
cadmium, bismuth, or zinc alloys. On the other hand, brazing filler metals are alloys 
consisting mainly of copper, silver, aluminum, magnesium, or nickel. Table 4.2 gives 
the recommended filler alloys for different base metals. The chemical composition 
and the field of application for the commonly used soft and hard solders are given in 
Table 4.3. 

TABLE 4.2 

Recommended filler alloys for different base metals 

Filler Alloys 


40% Zinc 


Base Metal 




Silver Copper 





* X 

Stainless steel 





Nickel alloys 





Copper (pure) 




Brass or bronze 



Silver (pure) 



Not allowed 


Either Sn-Zn 10 

Al-Mg 3, Al-Si 5 

or Al-Si 12 

* = may be used 
x = best used 

4.5 Brazing and Soldering 


TABLE 4.3 

Most commonly used soft and hard brazing filler metals 


Approximate Chemical 

Brazing or Soldering 






Tin solder 

Sn-Pb 70 

360-490°F (183-255°C) 

General-purpose solder 

Sn-Pb 50 

360-420°F (183-216°C) 

Electrical application 

Sn-Zn 10 

509-750°F (265-400°C) 

Soft soldering of 

Silver solder 

Ag 25-50, Cu 20-40, Sn 0-35, 

1175-1550°F (635-845°C) 

Brazing of copper alloys 

Cd 0-20, Zn 0-20 

and silver in electronics 

Brass solder 

Cu-Zn 40 

1670-1750°F (910-955°C) 

General-purpose hard 

Nickel silver 

Cu balance, Zn 20-30, 

1720-1800° F (938-982°C) 

Nickel alloys and steel 

Ni 10-20 



99.9% copper 

2000-2100°F (1093-1150°C) 

Brazing of steel 


Al-Si 12 

1080-1120°F (582-605°C) 

Brazing all aluminum 
alloys except silumin 


Fluxes are employed in soft soldering as well as in brazing in order to protect the 
cleaned surfaces of the base metal against oxidation during those processes. In addi- 
tion, fluxes enable proper wetting of the surfaces of the base metals by the molten filler 

There are two kinds of fluxes for soft soldering operations: organic and inor- 
ganic. The inorganic fluxes are mostly aqueous solutions of zinc and/or ammonium 
chlorides. They must, however, be completely removed after the soldering operation 
because of their corrosive effect. It is, therefore, completely forbidden to use inor- 
ganic fluxes in soldering electronic components. On the other hand, organic fluxes do 
not have such corrosive effects and are, therefore, widely used for fine soldering in 
electronic circuits. The commonly used organic fluxes involve colophony, a kind of 
resin with a melting point between 350°F and 390°F (180°C and 200°C), as well as 
some fats. 

The fluxes employed in brazing include combinations of borox, boric acid, bo- 
rates, fluorides, and fluoborates together with a wetting agent. The flux can be in the 
form of a liquid, slurry, powder, or paste, depending upon the brazing method used. 

Soldering Techniques 

The manual soldering method involves using a hand-type soldering iron that is made 
of copper and has to be tinned each time before use. The iron is first heated to a tem- 
perature of about 570°F (300°C), and its tip is then dipped into the flux and tinned with 

126 4 Joining of Metals 

the solder. Next, the iron is used for heating the prepared surfaces of the base metal 
and for melting and distributing the soft solder. When the solder solidifies, it forms the 
required solder seam. 

Several other methods are also used for soldering. These include dip soldering and 
induction soldering as well as the use of guns (blowtorches). Nevertheless, electric sol- 
dering irons are still quite common. 

Brazing Techniques 

The selection of a preferred brazing method has to be based on the size and shape of 
the joined components, the base metal of the joint, the brazing filler metal to be used, 
and the production rate. When two brazing techniques are found to be equally suit- 
able, cost is the deciding factor. The following brazing methods are commonly used 
in industry. 

Torch brazing. Torch brazing is still the most commonly used method. It is very 
similar to oxyfuel flame welding in that the source of heat is a flame obtained from 
the combustion of a mixture of a fuel gas (e.g., acetylene) and oxygen. The process 
is very popular for repair work on cast iron and is usually applied manually, al- 
though it can be used on a semiautomatic basis. In this process, however, a reduc- 
ing flame should be used to heat the joint area to the appropriate brazing 
temperature. A flux is then applied, and as soon as it melts, the filler metal (braz- 
ing alloy) is hand-fed to the joint area. When the filler metal melts, it flows into the 
clearance between the base components by capillary attraction. The filler metal 
should always be melted by the heat gained by the joint and not by directly apply- 
ing the flame. 

Furnace brazing. Furnace brazing is performed in either a batch or a continuous con- 
veyor-type furnace and is, therefore, best suited for mass production. The atmosphere 
of the furnace is controlled to prevent oxidation and to suit the metals involved in the 
process. That atmosphere can be dry hydrogen, dissociated ammonia, nitrogen, argon, 
or any other inert gas. Vacuum furnaces are also employed, especially with brazing 
materials containing titanium or aluminum. Nevertheless, a suitable flux is often em- 
ployed. The filler metal must be placed in the joint before the parts go inside the fur- 
nace. The filler metal can, in this case, take the form of a ring, washer, wire, powder, 
or paste. 

Induction brazing. In induction brazing, the components to be brazed are heated by 
placing them in an alternating magnetic field that, in turn, induces an alternating cur- 
rent in the components that rapidly reverses its direction. Special coils made of copper, 
referred to as inductors, are employed for generating the magnetic field. The filler 
metal is often placed in the joint area before brazing but can also be hand-fed by the 
operator. This technique has a clear advantage, which is the possibility of obtaining a 
very closely controlled heating area. 

Dip brazing. Dip brazing involves dipping the joint to be brazed in a molten filler 
metal. The latter is maintained in a special externally heated crucible and is covered 

4.5 Brazing and Soldering 127 

with a layer of flux to protect it from oxidation. Because the filler metal coats the en- 
tire workpiece, this process is used only for small parts. 

Salt-bath brazing. The source of heating in salt-bath brazing is a molten bath of fluo- 
ride and chloride salts. The filler metal is placed in the joint area before brazing and is 
also sometimes cladded. Next, the whole assembly is preheated to an appropriate tem- 
perature and then dipped for 1 to 6 minutes in the salt bath. Finally, the hot brazed joint 
is rinsed thoroughly in hot and cold water to remove any remaining flux or salt. Gen- 
erally, this process is employed for brazing aluminum and its alloys. There is, however, 
a problem associated with the process, and that is the pollution caused by the effluent 
resulting from the rinsing operation. 

Resistance brazing. Low-voltage, high-amperage current is used as the source of 
energy in resistance brazing, as is the case with spot welding. In fact, a spot welder 
can be employed to carry out this process, provided that the pressure is carefully ad- 
justed so as to be just enough to secure the position of the contact where heat devel- 
ops. The workpiece is held between the two electrodes, with the filler metal 
preloaded at the joint area. This process is normally used for brazing of electrical 
contacts and in the manufacture of copper transformer leads. 

Design of Brazed Joints 

For the proper design of brazed joints, two main factors have to be taken into consid- 
eration. The first factor involves the mechanics of the process in that the brazing filler 
metal flows through the joint by capillary attraction. The second factor is that the 
strength of the filler metal is poorer than that of the base metals. The product designer 
should aim for the following: 

1. Ensuring that the filler metal is placed on one side of the joint and allocating a 
space for locating the filler metal before (or during) the process. 

2. Adjusting the joint clearance in order to ensure optimum conditions during brazing. 
That clearance is dependent upon the filler metal used and normally takes a value 
less than 0.005 inch (0.125 mm), except for silumin, in which case it can go up to 
0.025 inch (0.625 mm). 

3. Ensuring that the distance to be traveled by the filler metal is shorter than the limit 
distant, as dictated by the physics of capillarity. 

4. Providing enough filler metal. 

5. Increasing the area of the joint because the filler metal is weaker than the base metal. 

There are three types of joint-area geometries: butt, scarf, and lap. The butt joint 
is the weakest, and the lap is the strongest. Nevertheless, when designing lap joints, 
make sure that the joint overlap is more than 3f, where t is the thickness of the thinner 
parent metal. Examples of some good and poor practices in the design of brazed joints 
are shown in Figure 4.48 as guidelines for beginners in product design. Also, always 
remember that brazed joints are designed to carry shear stress and not tension. 


4 Joining of Metals 

FIGURE 4.48 

Good and poor 
practices in the design 
of brazed joints 






< 3t 

ggF 23 





77 t\ 








S KS/////A 




tsssizzzzzj IzzzzS 








Sticking, or adhesive bonding, of metals is becoming very popular in the automotive, 
aircraft, and packaging industries because of the advantages that this technique can 
offer. Thanks to the recent development in the chemistry of polymers, adhesives are 
now cheap, can be applied easily and quickly, and can produce reasonably strong 
joints. Adhesive bonding can also be employed in producing joints of dissimilar met- 
als or combinations of metals and nonmetals like ceramics or polymers. This certainly 
provides greater flexibility when designing products and eliminates the need for com- 
plicated, expensive joining processes. 

As we know, it is possible to stick entirely smooth metal surfaces together. It is 
obvious, therefore, that the sticking action is caused by adhesive forces between the 
sticking agent and the workpiece and not by the flowing and solidification of the stick- 
ing agent into the pores of the workpiece as occurs, for instance, with wood. In other 
words, adhesion represents attractive intermolecular forces under whose influence the 
particles of a surface adhere to those of another one. There are also many opinions sup- 
porting the theory that mechanical interlocking plays a role in bonding. 


Structural adhesives are normally systems including one or more polymeric materials. 
In their unhardened state (i.e., before they are applied and cured), these adhesives 
can take the form of viscous liquids or solids with softening temperatures of about 
212°F (100°C). The unhardened adhesive agents are often soluble in ketones, esters, 
and higher alcohols, as well as in aromatic and chlorine hydrocarbons. The hardened 

4.6 Sticking of Metals 


FIGURE 4.49 

The three types of 
adhesive-bonded joints 

adhesives, however, resist nearly all solvents. Adhesives that find industrial application 
in bonding two nonmetallic workpieces include cyanacrylates, acrylics, and poly- 
urethanes. Following is a brief description of the adhesives that are commonly used in 

Epoxies. Epoxies are thermosetting polymers (see Chapter 8) that require the addi- 
tion of a hardener or the application of heat so that they can be cured. Epoxies are con- 
sidered to be the best sticking agents because of their versatility, their resistance to 
solvents, and their ability to develop strong and reliable joints. 

Phenolics. Phenolics are characterized by their low cost and heat resistance of up to 
about 930°F (500°C). They can be cured by a hardener or by heat or can be used in 
solvents that evaporate and thus allow setting to occur. Like epoxies, phenolics are 
thermosetting polymers with good strength, but they generally suffer from brittleness. 

Polyamide. The polyamide group of polymers is characterized by its oil and water re- 
sistance. Polyamides are usually applied in the form of hot melts but can also be used 
by evaporation of solvents in which they have been dissolved. Polyamides are nor- 
mally used as can-seam sealants and the like. They are also used as hot-melt for shoes. 

Silicones. Silicones can perform well at elevated temperatures; however, cost and 
strength are the major limitations. Therefore, silicones are usually used as high- 
temperature sealants. 

Joint Preparation 

The surfaces to be bonded must be clean and degreased because most adhesives do not 
amalgamate with fats, oils, or wax. Joint preparation involves grinding with sandpaper, 
grinding and filling, sand blasting, and pickling and degreasing with trichlorethylene. 
Oxide films, electroplating coats, and varnish films need not be removed (as long as 
they are fixed to the surface). Roughening of the surface is advantageous, provided that 
it is not overdone. 

Joint Design 

There are basically three types of adhesive-bonded joints. They are shown in Figure 
4.49 and include tension, shear, and peel joints. Most of the adhesives are weaker in 
peel and tension than in shear. Therefore, when selecting an adhesive, you should al- 
ways keep in mind the types of stresses to which the joint is going to be subjected. It 
is also recommended that you avoid using tension and peel joints and change the de- 
sign to replace these by shear joints whenever possible. 


Sheer Peel 


4 Joining of Metals 

Review Questions 


1. What does the riveting process involve? 

2. What are rivets usually made of? 

3. List some applications of riveting. 

4. Spot welding could not completely replace riv- 
eting. List some applications of riveting that 
cannot be done by spot welding. 

5. How would you define welding? 

6. List five types of welded joint designs and dis- 
cuss suitable applications for each type. 

7. What are the types of different methods for 
classifying the welding processes? 

8. How would you break all the manufacturing 
methods into groups according to each of 
these classifying methods? 

9. Explain briefly the mechanics of solid-state 

10. What are the two main obstacles that must be 
overcome so that successful pressure welding 
can be achieved? 

11. What is cold-pressure welding? Give two ex- 
amples, using sketches. 

12. Discuss briefly the mechanics of explosive 
welding and draw a sketch to show the inter- 
face between the welded parts. 

13. List some industrial applications for explosive 

14. Discuss briefly the mechanics of ultrasonic 

15. What are the typical applications of ultrasonic 

16. What are the different types of ultrasonic 
welding machines? List the main components 
common to all these machines. 

17. What is the basic idea on which friction weld- 
ing is based? 

18. Explain briefly the stages involved in a fric- 
tion welding operation. 

19. List the various advantages claimed for fric- 
tion welding. 

20. List the limitations of friction welding. 

21. What is the difference between friction weld- 
ing and inertia welding? 

22. What advantages does inertia welding have 
over friction welding? 

23. Give examples of some parts that are fabri- 
cated by inertia welding. 

24. Explain briefly the mechanics of induction 

25. List some of the common industrial applica- 
tions of the induction welding process. 

26. What is the source of energy in thermit weld- 
ing? Explain. 

27. How is thermit welding performed? 

28. What are the common applications of thermit 

29. How does bonding take place in diffusion 

30. List the different processes that belong to re- 
sistance welding. 

31. Explain briefly the stages involved in a resis- 
tance-butt welding process. 

32. Using a sketch, explain the pressure-time and 
current-time relationships in resistance-butt 

33. List some of the applications of resistance-butt 

34. Clarify the difference between flash welding 
and butt welding. 

35. Draw a graph illustrating current versus time 
and pressure versus time in flash welding. 

Chapter 4 Review Questions 


36. When is flash welding recommended over butt 

37. What is the major disadvantage of flash weld- 

38. Explain the basic idea of percussion welding. 

39. Explain briefly the mechanics of spot welding. 

40. Draw a sketch of a section through a spot- 
welded joint. 

41. Draw a sketch to show a typical cycle for a 
spot-welding machine. 

42. How do you compare seam welding with spot 

43. List some of the advantages of seam welding. 

44. What are the industrial applications of seam 

45. What is the basic idea of projection welding? 

46. What are the advantages of projection weld- 

47. What is the condition for two metals to be 
joined together by fusion welding? 

48. How many zones can be identified in a joint 
produced by a conventional fusion welding 
process? Discuss briefly the microstructure in 
each of these zones. 

49. For what do the letters HAZ stand? 

50. Explain briefly the phenomenon of the electric 
arc and how it can be employed in welding. 

51. What are the advantages of alternating current 
over direct current in arc welding? 

52. What is the difference between DCSP and 
DCRP? When would you recommend using 
each of them? 

53. What is meant by the rated duty cycle? 

54. What shields the molten metal during 
shielded-metal arc welding? 

55. What is the main shortcoming of shielded- 
metal arc welding? 

56. List some of the functions of electrode coat- 

57. Explain briefly the Bernardos welding method. 

58. What is the main feature of the electrodes in 
flux-cored arc welding? 

59. What provides the shielding in flux-cored arc 

60. What is stud arc welding? How is it per- 
formed? List the main applications of this 

61. How is shielding achieved in submerged arc 

62. Why must the plates to be joined by sub- 
merged arc welding be horizontal only? 

63. Why does submerged arc welding always 
yield very high quality welds? 

64. List some of the advantages of submerged arc 

65. What provides shielding in MIG welding? 

66. Why does the MIG welding process render it- 
self suitable for automation? 

67. How can the penetration for gas-metal arc 
welding be controlled? 

68. What is the main difference between the MIG 
and the TIG welding processes? 

69. List some of the applications of TIG welding. 

70. In TIG welding, when would you use an ac 
power supply and when would you use DCSP 
and DCRP? 

71. Explain briefly the mechanics and the basic 
idea of plasma arc welding. 

72. When is plasma arc welding most recom- 

73. Do you consider electroslag welding to be a 
true arc welding process? 

74. How does welding take place in the elec- 
troslag welding process? 

75. When is electroslag welding usually recom- 

76. What is the source of energy in oxyacetylene 
flame welding? 


4 Joining of Metals 

77. How is acetylene stored for use in welding 

78. What does the equipment required in gas 
welding include? Explain the function of each 

79. What are the types of flames that can be ob- 
tained in gas welding? How is each one ob- 

80. What are the zones of a neutral flame? Discuss 
the effect of the oxygen-to-acetylene ratio on 
the nature of the flame obtained. 

81. Explain briefly the operating principles of 
electron-beam welding. 

82. What are the major limitations of the electron- 
beam welding process? 

83. List some of the advantages of electron-beam 

84. What are the major applications of electron- 
beam welding? 

85. For what do the letters in the word laser 

86. Using a sketch, explain how a laser beam ca- 
pable of carrying out welding can be gener- 

87. Explain briefly the mechanics of laser-beam 

88. What are the main advantages of laser-beam 

89. List some of the applications of laser-beam 

90. Using sketches, illustrate the commonly expe- 
rienced welding defects. How can each be 

91. What are the main tests for the inspection of 
welds? Discuss each briefly. 

92. What are the factors affecting the selection of 
the joint type? 

93. On what basis are the location and distribution 
of welds planned? 

94. What rules would you consider when design- 
ing a welded joint? 

95. What is meant by hard-facing ? 

96. What are the main applications of hard-fac- 

97. List the main types of thermal cutting 
processes. Discuss briefly the advantages and 
limitations of each. 

98. How do the mechanics of brazing differ from 
those of welding? 

99. What is the main difference between brazing 
and soft soldering? 

100. List some of the alloys used as brazing fillers 
and mention the base metals that can be 
brazed with each one. 

101. List some of the commonly used soft solders. 

102. What is the main function of brazing fluxes? 

103. List some of the fluxes used in brazing. 

104. List some of the fluxes used in soft soldering. 
Discuss the limitations and applications of 

105. In soft soldering, how should the solder be 

106. List the different brazing techniques used in 
industry. Discuss the advantages and limita- 
tions of each. 

107. As a product designer, what factors should 
you take into consideration when designing a 
brazed joint? 

108. In what case can sticking of metals not be re- 
placed by other welding and brazing tech- 

109. List some of the commonly used adhesives. 
Discuss the characteristics and common appli- 
cations of each. 

110. What are the types of adhesive-bonded joints? 
Which one is usually the strongest? 

Chapter 4 Design Example 


Pr oblem 


1. Two steel slabs, each 1/4 inch (6.35 mm) thick, 
are to be joined by two fillet welds (i.e., at both 
edges). If the width of each slab is 2.5 inches 
(62.5 mm) and the joint is to withstand a load of 
35,000 pounds (156,000 N), determine the al- 
lowable tensile strength of the electrode type to 
be used in welding. 

2. Two steel plates, each 1/4 inch (6.35 mm) thick, 
are to be joined by two fillet welds. If the joint is 
to withstand a load of 50,000 pounds (222,500 N) 
and an E7014 electrode (allowable tensile 
strength = 21,000 lb/in. 2 i.e., 145,000 KN/m 2 ) is 
used, determine the length of weld at each edge. 
If the plate width is 10 inches (254 mm), how 
would you distribute the weld? Draw a sketch. 

3. Two steel plates, each 5/16 inch thick (7.9 mm), 
are to be fillet-welded to a third one that is sand- 
wiched between them. The width of each of the 
first two plates is 4 inches (100 mm), whereas 
the width of the third one is 6 inches (150 mm). 
The two plates overlap the third one by 6 inches 
(150 mm), and an E7014 electrode (allowable 
tensile strength = 21,000 lb/in. 2 , 145,000 
KN/m 2 ) is to be used. If the joint is to withstand 
a load of 190,000 pounds (846 KN), use a sketch 
to illustrate a design for this joint and provide all 

An equal-leg-angle steel section 3 by 3 by 1/4 
inch (75 by 75 by 6 mm) is to be welded to a 
plate using an E7014 electrode (allowable tensile 
strength = 21,000 lb/in. 2 i.e., 145,000 KN/m 2 ). If 
the joint is to withstand a load of 10,000 pounds 
(44.5 KN) coinciding with the axis of the angle, 
design the joint and make a sketch indicating the 
distribution of the weld to eliminate any ten- 
dency of the angle to rotate. 

Two mild steel pipes, each having a 3/4-inch 
(19 mm) outer diameter, are to be joined to- 
gether by brazing. Assuming that the joint is to 
withstand an axial load of 6 tons, give a detailed 
design of this joint. (Take allowable shear stress 
of copper to be 6000 lb/in. 2 i.e., 41,430 KN/m 2 .) 

Two mild steel sheets, each 3/32 inch (2.4 mm) 
thick, are to be brazed together using copper as a 
filler material. Calculate the strength of the joint 
when it is manufactured according to each of the 
designs given in Figure 4.48. Compare the re- 
sults and recommend the design that gives max- 
imum strength. 

A power supply for arc welding is rated at a 
150-A 30-percent duty cycle. What will be the 
percentage of actual time utilized in welding to 
the total time the power supply is on if the cur- 
rent employed in welding is only 125 A? 


r- W&4TA 


You are required to design a flat-belt pulley so that it can be fabricated by welding. 
The pulley is to be mounted on a shaft that is 1 l A inches (3 1 mm) in diameter, and the 
outside diameter of the rim is 10 inches (250 mm). The rim of the pulley is to provide 


4 Joining of Metals 

a surface to transmit a torque of 600 lb ft (816 Nm) from a 2-inch- wide (50-mm) flat 
belt to the shaft. The number of pulleys required is only 5. 


It is advisable to start by gathering information about guidelines for the construc- 
tional features of flat-belt pulleys (e.g., width of rim for a certain belt width and thick- 
ness of rim). Information about the safe speeds of various sizes of pulleys should also 
be collected. 

Key. The best strategy is to design the key so that it will be the weak link in the 
pulley-key-shaft assembly because it is easy to replace. A suitable key material is 
AISI 1020 CD steel, which is commercially available as a key stock material. It has 
the following mechanical properties: 

Ultimate Tensile Strength (UTS) = 78,000 lb/in. 2 

yield stress = 66,000 lb/in." 

yield stress in shear = 38,000 lb/in. 2 

Consider Figure 4.50. The force acting on the key is given by 

„ T 600 lb ft x 12 in/ft ,, c „ n 

P = — = 1 1 ,520 pounds 

r 0.625 inch 

Take the key cross section to be 1/4 by 1/4 inch (6 by 6 mm), and its length € in inches: 

. . 11,520 pounds . 
shear stress in the key = < x a ii wabie 


Take a safety factor of 2 for the key: 
38,000 , _ „„„ ,. .. 2 

Xallowable = ~ ~ = 19,000 lb/in.~ 

FIGURE 4.50 

Forces acting on the 

0. 25 inch 

fl= 0.625 inch 

Chapter 4 Design Example 135 


= 19.000 


£ = 2.4 inches (60 mm) 

11,520 ^ „ ui ♦ 

bearing stress = < allowable compressive stress 

!/4 xVixi 

< 66,000 


= 33,000 lb/in. 2 (safety factor of 2) 

€ = 2.8 inches (70 mm) 

We take this value to ensure safety against both shearing and compressing loads. We 
should, however, round it, so the length of the key is to be 3 inches (75 mm). 

Hub. Use a round seamless tube having a 2.25-inch outer diameter and 9/16-inch 
wall. A suitable material is AISI 1020 CD steel. Again, the length of the hub must not 
be less than 2.8 inches to keep the bearing stress below the allowable value. Take it as 
2.875 inches. 

Rim. Use a round seamless tube having a 10-inch outer diameter and 1/4-inch wall. 
Again, a suitable material is AISI 1020 CD steel because of its availability and ability 
to withstand the rubbing effect of the moving belt. 

Spokes. The positioning and welding of four or five spokes would create a serious 
problem and necessitates the use of a complicated welding fixture. Therefore, the 
spokes are to be replaced in the design by a web. Use a 5/16-inch flat plate, machined 
to have an outer diameter of 9.43 inches and an inner diameter 2.31 inches. An appro- 
priate material is AISI 1020 HR steel. Because weight can be a factor, it is good prac- 
tice to provide six equally spaced holes in the web by machining. These can also serve 
as an aid in the handling and positioning of the web during welding. 

Welding. Use conventional arc welding; an E7014 electrode (allowable tensile 
stress = 21,000 lb/in. 2 ) can be used. A fillet weld with a leg of 1/4 inch is adopted. The 
force is given by 

p = torque = 7200 = ^ pounds 
radius 1.125 inches 


H £ 

"' o 

■ .c 

t « 

UJ | 

O to 

— ■£ 



Chapter 4 Design Projects 137 

The required length of the weld is 

l Ax 0.707x0.57x21,000 

= 3.03 inches (75 mm) 

Use 2.00 inches (50 mm) of weld on each side of the hub. 

The circumference of the hub equals n times 2.25, or 7.06 inches. Space four 
welds, each 0.5 inch (12.5 mm) in length, equally, 90° apart around the circumference 
of the hub. Welds on both sides of the web should be staggered. Adopt the same welds 
at the rim. They should be safe because the shearing force is much lower (the radius is 
larger than that of the hub). 

Once all the dimensions and details are known, we are in a position to construct 
the pulley as shown in the workshop drawing in Figure 4.51. 

jgn Projects 

1. Design a table for the machine shop. The table should be 4 feet (1200 mm) in 
height, with a surface area of 3 by 3 feet (900 by 900 mm), and should be able to 
carry a load of half a ton. Because only two tables are required, the design should 
involve the use of steel angles and a plate that are to be joined together by welding. 

2. Design a tank for compressed air. It has a capacity of 100 cubic feet (2.837 m 3 ) and 
can withstand an internal pressure of 40 atmospheres (ata). The number of tanks re- 
quired is 50, and the tanks are going to be placed in a humid environment. 

3. Design a compressed-air reservoir (tank) that is to be subjected to an extremely cor- 
rosive environment. The capacity of the tank is 30 cubic feet (0.85 m 3 ), and the 
maximum gage pressure is 70 ata, but the pressure is pulsating from zero to the 
maximum value about once every 5 minutes. The number of tanks required is 100. 

4. A straight-toothed spur-gear wheel transmits a torque of 1200 lb ft (1632 Nm) to a 
2-inch-diameter (50 mm) steel shaft (AISI 1045 CD steel). The pitch diameter of 
the gear is 8 inches (200 mm), its width is 3 inches (75 mm), and the base diame- 
ter is 7.5 inches (187.5 mm). Make a detailed design for the gear's blank (i.e., be- 
fore the teeth are cut). 

5. A mobile winch (little crane) can be moved on casters. It has a capacity of lifting 
1 ton for 3 feet (0.9 m) about ground. The lifting arm can be extended, and the 
winch can then lift 1/2 ton for up to 6 feet (1.8 m). Knowing that the production 
volume is 4000 units and that casters and hydraulic pressure cylinders are to be pur- 
chased from vendors, provide a detailed design and include full specifications of 
the parts to be purchased. 

138 4 Joining of Metals 

6. The lifting arm for a crane is 60 feet (about 20 m), and its lifting capacity is 1 ton. 
It is to be used in construction work and to be subjected to humidity, dirt, and so 
on. Provide a detailed design for this arm using steel angles that are to be welded 

7. Design a frame for a hydraulic press for fabrication by welding. The height of the 
cross arm is 12 feet (about 4 m). The cross arm is mounted (by welding) on two 
vertical columns that are, in turn, welded to the base. The press can produce a max- 
imum load of 200 tons by means of a hydraulic cylinder attached to the cross arm 
(below it), and the stroke is 12 inches (300 mm). 

TIP: The energy absorbed when the frame deforms should not exceed 2 per- 
cent of the total energy output of the press. 

Chapter 5 

Metal Forming 




Metal forming processes have gained significant attention since World War II as 
a result of the rapid increase in the cost of raw materials. Whereas machining 
processes involve the removal of portions of the stock material (in the form of 
chips) in order to achieve the required final shape, metal forming processes 
are based upon the plastic deformation and flow of the billet material in its 
solid state so as to take the desired shape. Consequently, metal forming 
processes render themselves more efficient with respect to raw material uti- 
lization than machining processes, which always result in an appreciable ma- 
terial waste. 

In fact, although metal forming techniques were employed in manufacturing 
only semifinished products (like sheets, slabs, and rods) in the past, finished 
products that require no further machining can be produced today by these 
techniques. This was brought about by the recent developments in working 
methods, as well as by the construction features of the forming machines em- 
ployed. Among the advantages of these up-to-date forming techniques are high 
productivity and very low material waste. Therefore, more designers tend to 
modify the construction of the products manufactured by other processes to 
use forming. Also, bearing in mind that metal forming methods are still being 
used for producing semifinished products, it is evident that the vast majority of 
all metal products are subjected to forming, at least at one stage during their 
production. This latter fact clearly manifests the importance of the metal form- 
ing methods. 


140 5 Metal Forming 

Generally, metal forming involves both billet and sheet metal forming. How- 
ever, it has been a well-accepted convention to divide those processes into two 
main groups: bulk (or massive) forming and sheet metal working. In this chap- 
ter, only bulk forming processes (e.g., forging, cold forming, and rolling) are 
covered; Chapter 6 deals with the working of sheet metal. 



Factors Affecting Plastic Deformation 

During any forming process, the material plastically flows while the total volume of 
the workpiece remains substantially constant. However, there are some marked 
changes that take place on a microscopic scale within the grains and the crystal lattice 
of the metal, resulting in a corresponding change in the properties of the material. This 
latter change can be explained in view of the dislocation theory, which states that the 
plastic deformation and flow of metal are caused by movement and transfer of dislo- 
cations (defects in the crystal lattice) through the material with the final outcome of ei- 
ther piling up or annihilating them. Following are some factors that affect plastic 
deformation by influencing the course of dislocations. 

Impurities and alloying additives. It is well known that pure metals possess higher 
plasticity than their alloys. The reason is that the presence of structural components 
and chemical compounds impedes the transfer and migration of dislocations, resulting 
in lower plasticity. 

Temperature at which deformation takes place. As a rule, the plasticity of a metal 
increases with temperature, whereas its resistance to deformation decreases. The 
higher the temperature, the higher the plasticity and the lower the yield point. More- 
over, no work-hardening occurs at temperatures above the recrystallization tempera- 
ture. This should be expected because recrystallization denotes the formation and 
growth of new grains of metal from the fragments of the deformed grains, together 
with restoring any distortion in the crystal lattice. Consequently, strength values drop 
to the level of a nonwork-hardened state, whereas plasticity approaches that of the 
metal before deformation. In fact, a forming process is termed hot if the tempera- 
ture at which deformation takes place is higher than the recrystallization temperature. 
Lead that is formed at room temperature in summer actually undergoes hot forming 
because the recrystallization temperature for lead is 39.2°F (4°C). When deforma- 
tion occurs at a temperature below the recrystallization temperature of the metal, the 
process is termed cold forming. Cold forming processes are always accompanied by 
work-hardening due to the piling up of dislocations. As a result, strength and hardness 
increase while both ductility and notch toughness decrease. These changes can be 
removed by heat treatment (annealing). On the other hand, when hot forming a 
metal, the initial dendritic structure (the primary structure after casting) disintegrates 
and deforms, and its crystals elongate in the direction of the metal flow. The insoluble 

5.1 Plastic Deformation 


impurities like nonmetallic inclusions (around the original grain boundaries) are drawn 
and squeezed between the elongated grains. This texture of flow lines is usually re- 
ferred to as the fibrous macrostructure. This fibrous macrostructure is permanent and 
cannot be removed by heat treatment or further working. As a result, there is always 
anisotropy of mechanical properties; strength and toughness are better in the longitu- 
dinal direction of fibers. Also, during hot forming, any voids or cracks around grain 
boundaries are closed, and the metal welds together, which, in turn, results in im- 
provements in the mechanical properties of the metal. 

Rate of deformation. It can generally be stated that the rate of deformation (strain 
rate) in metal working adversely affects the plasticity of the metal (i.e., an increase in 
the deformation rate is accompanied by a decrease in plasticity). Because it takes the 
process of recrystallization some time to be completed, that process will not have 
enough time for completion when deformation occurs at high strain rates. Therefore, 
greater resistance of the metal to deformation should be expected. This does not mean 
that the metal becomes brittle. 

State of stress. A state of stress at a point can be simply described by the magni- 
tudes and directions of the principal stresses (a stress is a force per unit area) acting on 
planes that include the point in question. The state of stress is, in fact, a precise and 
scientific expression for the magnitudes and the directions of the external forces acting 
on the metal. All possible states of stress can be reduced to only nine main systems, as 
shown in Figure 5.1. These nine cases can, in turn, be divided into three groups. The 


The nine main systems 
of the state of stress 









? — 




' i 


.o 2 Plane 




»_ II., On -*- 

a 7 o 2 - 



Metal Forming 

first group includes two systems that are characterized by the absence of stress (forces) 
along two directions, and the stress system is therefore called uniaxial. This is the case 
when stretching sheet metal having a length that considerably exceeds its width. In 
each of the three systems included in the second group, it is clear that a stress along 
only one of the directions is absent. Because the other two directions (stresses) form a 
plane, each of these systems is referred to as a plane-stress state. It may approximately 
be represented by stretching of a thin sheet in two or more directions. The remaining 
group indicates the state of stress of a body, where there are stresses acting along all 
three directions in space, yielding the term triaxial. In fact, most of the bulk forming 
operations (forging, rolling, and wire drawing) cause states of stress that belong to this 
latter group. 

Load and Energy Requirement 

The force required for deforming a given metal (at any unchanged desired temperature 
and at usual strain-rate levels) is dependent upon the degree of deformation, which is 
the absolute value of the natural logarithm of the ratio of the final length of the billet 
to its original length. On the other hand, the energy consumed throughout the forming 
process is equivalent to the area under the load-deformation curve for that forming 
process. Therefore, that energy can be calculated if the relationship between the load 
and the deformation is known. Figure 5.2 shows the degree of deformation and the en- 
ergy consumed in an upsetting operation. It must be noted that in both hot and cold 
working, there is an upper limit for the degree of deformation (especially in cold work- 
ing) above which cracks and discontinuities in the workpiece initiate. 

Preheating the Metal for Hot Forming 

Before being subjected to hot forming processes, ingots (or billets) should be uni- 
formly heated throughout their cross sections, without overheating or burning the 
metal at the surface. This is particularly important when forming steels. Attention must 
also be given to the problems of decarburization and formation of scale in order to 
bring them to a minimum. The thermal gradient is another important factor that affects 


The degree of 
deformation and the 
energy consumed in 


. I — . — 

I Original 



5.1 Plastic Deformation 143 

the soundness of the deformed part. If the temperature gradient is high, thermal 
stresses may initiate and can cause internal cracks. This usually happens when a por- 
tion of the metal is above the critical temperature of metal (AC] or AC 3 ) while the rest 
of the billet is not. The larger the cross section of the billet and the lower its coefficient 
of thermal conductivity, the steeper the temperature gradient will be, and the more li- 
able to internal cracking during heating the billet becomes. In the latter case, the rate 
of heating should be kept fairly low (about 2 hours per inch of section of the billet) in 
order not to allow a great difference to occur between the temperatures at the surface 
and the core of the billet. The metal must then be "soaked" at the maximum tempera- 
ture for a period of time long enough to ensure uniformity of temperature. 

The maximum temperature to which the billet is heated before forming differs 
for different metals. There is usually an optimum range of temperatures within which 
satisfactory forming is obtained because of increased plasticity and decreased resis- 
tance to deformation. Nevertheless, any further increase in temperature above that 
range may, on the contrary, result in a defective product. Burned metal and coarse 
grain structure are some of the defects encountered when a metal is excessively 

The ingots may be heated in soaking pits, forge hearths, chamber furnaces, or 
car-bottom furnaces, which are all heated with gas. Rotary hearth furnaces represent 
another type of heating furnace. In mass production or automated lines, small ob- 
jects (billets) are heated using electric current and the phenomenon of induction. 
This induction-heating method is quick and keeps the surfaces of the billets clean, 
and temperatures can be accurately controlled. Moreover, physical equipment re- 
quires limited floor space and can be fully automated. 

Friction and Lubrication 
in Working of Metals 

Friction plays an important role in all metal forming processes and is generally con- 
sidered to be undesirable because it has various harmful effects on the forming 
processes, on the properties of products, and on the tool life. During the deformation 
of a metal, friction occurs at the contact surface between the flowing metal and the tool 
profile. Consequently, the flow of the metal is not homogeneous, which leads to the 
initiation of residual stresses, with the final outcome being an unsound product with in- 
ferior surface quality. Also, friction increases the pressure acting on the forming tool 
(as well as the power and energy consumed) and thus results in greater wear of the 

Friction in metal forming is drastically different from the conventional Columb's 
friction because extremely high pressure between the mating bodies (tool and work- 
piece) is involved. Recent theories on friction in metal forming indicate that it is actu- 
ally the resistance to shear of a layer, where intensive shear stress is generated as a 
result of relative displacement between two bodies. When these bodies have direct 
metal-to-metal contact, slipping and shear flow occur in a layer adjacent to the contact 
interface. But, if a surface of contact is coated with a material having low shear resis- 
tance (a lubricant that can be solid or liquid), slipping takes place through that layer of 


5 Metal Forming 

lubricant and, therefore, has low resistance. This discussion indicates clearly that the 
magnitude of the friction force is determined by the mechanical properties (yield point 
in shear) of the layer where actual slipping occurs. Hence, it is evident that a metal 
having a low yield point in shear, such as lead, can be used as a lubricant when form- 
ing metals having relatively high yield strength in compression. Figure 5.3 shows the 
shear layer in three different cases: solid lubrication, dry sticking friction, and hydro- 
dynamic (liquid) lubrication. 

In order to reduce friction and thus eliminate its harmful effects, lubricants are ap- 
plied to the tool-workpiece interface in metal forming processes. The gains include 
lower load and energy requirement, prevention of galling or sticking of the workpiece 
metal onto the tool, better surface finish of products, and longer tool life. An important 
consideration when selecting a lubricant is its activity (i.e., its ability to adhere 
strongly to the surface of the metal). The activity of a lubricant can, however, be en- 
hanced by adding material with high capability of adsorption, such as fat acids. Among 
other factors to be considered are thermal stability, absence of poisonous fumes, and 
complete burning during heat treatment of the products. 

In cold forming processes, vegetable and mineral oils as well as aqueous emul- 
sions are employed as lubricants. These have the advantage of acting as coolants, elim- 
inating excessive heat and thus reducing the temperature of the tool. Solid polymers, 
waxes, and solid soaps (sodium or calcium stearates) are also widely used in cold- 
metal working. 

For relatively high temperature applications, chlorinated organic compounds and 
sulfur compounds are used. Solid lubricants like molybdenum disulfide and graphite 
possess low-friction properties up to elevated temperatures and are, therefore, used as 
solid lubricants in hot forming. Graphite is sometimes dispersed in grease, especially 
in hot forging ferrous materials. Lately, use has been made of molten glass as a lubri- 
cant when alloy steels and special alloys are hot formed. The glass is added in the form 
of powder between the die and a hot billet. The advantages of molten glass include low 
friction, excellent surface finish, and improved tool life. 


The shear layer in three 

different cases Soft 



Dry sticking 



5.2 Rolling 145 

Cold Forming Versus Hot Forming 

Cold forming has its own set of advantages and disadvantages, as does hot form- 
ing, and, therefore, each renders itself appropriate for a certain field of applications. 
For instance, cold forming will enhance the strength of the workpiece metal, im- 
prove the quality of the surface, and provide good dimensional accuracy, but the 
plastic properties of the metal (elongation percentage and reduction-in-area percent- 
age) and the impact strength drop. Therefore, the final properties of cold-formed 
products are obtained as required by adjusting the degree of deformation and the 
parameters of the postheating treatment process. Because the loads involved in cold 
forming are high, this technique is generally employed in the manufacture of small 
parts of soft, ductile metals, such as low-carbon steel. Also, large quantities must be 
produced to justify the high cost of tooling involved. Nevertheless, if the products 
are to be further processed by machining, the increased hardness caused by cold 
working is a real advantage because it results in better machinability. Therefore, 
cold-rolled plates and cold-drawn bars are more suitable for machining purposes 
than hot-formed ones. 

On the other hand, the yield strength of a metal drops significantly at elevated 
temperatures, and no work-hardening occurs. Consequently, hot forming processes 
are used when high degrees of deformation are required and/or when forming large 
ingots or billets because the loads and energies needed are far lower than those re- 
quired in cold forming. Moreover, hot forming refines the grain structure, thus pro- 
ducing softer and more ductile parts suitable for further processing by cold forming 
processes. However, high temperatures affect the surface quality of products, giving 
oxidation and scales. 

Decarburization may also occur in steels, especially when hot forming 
high-carbon steel. The scales, oxides, and decarburized layers must be removed by 
one or more machining processes. This slows down the production, adds machin- 
ing costs, and yields waste material, resulting in lower efficiency of material uti- 
lization. A further limitation of hot forming is reduced tool life due to the 
softening of tool surfaces at elevated temperatures and the rubbing action of the 
hot metal while flowing. This actually subjects the tools to thermal fatigue, which 
shortens their life. 

Hot rolling is the most widely used metal forming process because it is employed to 
convert metal ingots to simple stock members called blooms and slabs. This process 
refines the structure of the cast ingot, improves its mechanical properties, and elimi- 
nates the hidden internal defects. The process is termed primary rolling and is fol- 
lowed by further hot rolling into plates, sheets, rods, and structural shapes. Some of 
these may be subjected to cold rolling to enhance their strength, obtain good surface 
finish, and ensure closer dimensional tolerances. Figure 5.4 illustrates the sequence of 
operations involved in manufacturing rolled products. 


Sequence of operations involved in manufacturing rolled products 

Finished products 


5.2 Rolling 


Principles of Rolling 

The process of rolling consists of passing the metal through a gap between rolls rotat- 
ing in opposite directions. That gap is smaller than the thickness of the part being 
worked. Therefore, the rolls compress the metal while simultaneously shifting it for- 
ward because of the friction at the roll-metal interfaces. When the workpiece com- 
pletely passes through the gap between the rolls, it is considered fully worked. As a 
result, the thickness of the work decreases while its length and width increase. How- 
ever, the increase in width is insignificant and is usually neglected. As can be seen in 
Figure 5.5, which shows the rolling of a plate, the decrease in thickness is called draft, 
whereas the increase in length and the increase in width are termed absolute elonga- 
tion and absolute spread, respectively. Two other terms are the relative draft and the 
coefficient of elongation, which can be given as follows: 

relative draft e = 

Ah x 100 K- h{ 

x 100 


coefficient of elongation r\ = — 
But because the volume of the work is constant, it follows that 


h n x b c 




h f x b t 

Equation 5.3 indicates that the coefficient of elongation is adversely proportional to the 
ratio of the final to the original cross-sectional areas of the work. 

As can be seen in Figure 5.6, the metal is deformed in the shaded area, or defor- 
mation zone. The metal remains unstrained before this area and does not undergo any 
further deformation after it. It can also be seen that the metal undergoing deformation 
is in contact with each of the rolls along the arc AB, which is called the arc of contact. 
It corresponds to a central angle, a, that is, in turn, called the angle of contact, or angle 
of bite. From the geometry of the drawing and by employing simple trigonometry, it 
can be shown that 

cos a = 1 

K - hf __ x _ Ah 
2R 2R 



Simple rolling of a plate 





5 Metal Forming 


The deformation zone, 
state of stress, and 
angle of contact in 

Equation 5.4 gives the relationship between the geometrical parameters of the rolling 
process, the angle of contact, the draft, and the radius of the rolls. Note that in order to 
ensure that the metal will be shifted by friction, the angle of contact must be less than 
fi, the angle of friction, where tan P = u, (the coefficient of friction between roll surface 
and metal). In fact, the maximum permissible value for the angle of contact depends 
upon other factors, such as the material of the rolls, the work being rolled, and the 
rolling temperature and speed. Table 5.1 indicates the recommended maximum angle 
of contact for different rolling processes. 

Load and Power Requirement 

As can also be seen in Figure 5.6, the main stress system in the deformation zone in a 
rolling process is triaxial compression, with the maximum (principal) stress acting nor- 
mal to the direction of rolling. The deformed metal is exerting an equal counterforce 
on each of the rolls to satisfy the equilibrium conditions. Therefore, this force normal 
to the direction of rolling is important when doing the design calculations for the rolls 
as well as the mill body. It is also important in determining the power consumption in 

TABLE 5.1 

Maximum allowable 
angle of contact for 

Rolling Process 

Maximum Allowable 
Angle of Contact 

Rolling of blooms and heavy sections 
Hot rolling of sheets and strips 
Cold rolling of lubricated sheets 

24°-30 c 

15°-20 e 

2°-10 c 

5.2 Rolling 149 

a rolling process. Unfortunately, the exact determination of that rolling load and power 
consumption is complicated and requires knowledge of theory of plasticity as well as 
calculus. Nevertheless, a first approximation of the roll load can be given by the fol- 
lowing simple equation: 

F=Y xbx V«xM < 5 - 5 ) 

where Y is the average (plane-strain) yield stress assuming no spread and is equal to 
\.\5Y, where Y is the mean yield stress of the metal. Therefore, Equation 5.5 should 
take the form 

F= l.\5YxbxVRxAh (5-6) 

Equation 5.6 neglects the effect of friction at the roll-work interface and, therefore, 
gives lower estimates of the load. Based on experiments carried out on a wide range 
of rolling mills, this equation can be modified to account for friction by multiplying by 
a factor of 1 .2. The modified equation is 

F= 1.2 x 1.157x6 xV/TxA/i (5.7) 

The power consumed in the process cannot be obtained easily; however, a rough esti- 
mate in low-friction conditions is given by 

Y xbxRx Ahxoi , ce , 

hp = _ (5.8, 

where co is the angular velocity of rolls in radians per second, and Y, b, R and Ah are 
all in English units. 

Rolling Mills 

A rolling mill includes one or more roll stands and a main drive motor, reducing gear, 
stand pinion, flywheel, and coupling gear between the units. The roll stand is the main 
part of the mill, where the rolling process is actually performed. It basically consists of 
housings in which antifriction bearings that are used for carrying (mounting) the rolls 
are fitted. Moreover, there is a screw-down mechanism to control the gap between the 
rolls and thus the required thickness of the product. 

Depending upon the profile of the rolled product, the body of the roll may be ei- 
ther smooth for rolling sheets (plates or strips) or grooved for manufacturing shapes 
such as structural members. A roll consists of a body, two necks (one on each side), 
and two wobblers (see Figure 5.5). The body is the part that contacts and deforms the 
metal of the workpiece. The necks rotate in bearings that act as supports, while the 
wobblers serve to couple the roll to the drive. Rolls are usually made from high- 
quality steel and sometimes from high-grade cast iron to withstand the very severe ser- 
vice conditions to which the rolls are subjected during the rolling process, such as 
combined bending and torque, friction and wear, and thermal effects. Gray cast-iron 
rolls are employed in roughing passes when hot rolling steel. Cast- or forged-steel rolls 
are used in blooming, slabbing, and section mills as well as in cold-rolling mills. 
Forged rolls are stronger and tougher than cast rolls. Alloy-steel rolls made of chrome- 
nickel or chrome-molybdenum steels are used in sheet mills. 


5 Metal Forming 

Classification of Rolling Mills 

Rolling mills are classified according to the number and arrangement of the rolls in a 
stand. Following are the five main types of rolling mills, as shown in Figure 5.7a 
through e. 

Two-high rolling mills. Two-high rolling mills, the simplest design, have a two-high 
stand with two horizontal rolls. This type of mill can be nonreversing (unidirectional), 
where the rolls have a constant direction of rotation, or reversing, where the rotation 
and direction of metal passage can be reversed. 

Three-high rolling mills. Three-high rolling mills have a three-high stand with three 
rolls arranged in a single vertical plane. This type of mill has a constant direction of 
rotation, and it is not required to reverse that direction. 

Four-high rolling mills. In sheet rolling, the rolls should be designed as small as pos- 
sible in order to reduce the rolling force F of the metal on the rolls and the power re- 
quirement. If such small-diameter rolls are used alone, they will bend and result in 
nonuniform thickness distribution along the width of the sheet, as shown in Figure 5.8. 
For this reason, another two backup rolls are used to minimize bending and increase the 
rigidity of the system. The four rolls are arranged above one another in a vertical plane. 
Also, the backup rolls always have larger diameters than those of the working rolls. 

Multihigh rolling mills (Sendzimir mills). Multihigh rolling mills are used particularly 
in the manufacture of very thin sheets, those with a thickness down to 0.0005 inch 
(0.01 mm) and a width up to 80 inches (2000 mm), into coils. In this case, the work- 
ing rolls must have very small diameters (to reduce load and power consumption, as 
explained before), usually in the range of 3/8 inch (10 mm) up to 1.25 inches (30 mm). 


The five main types of 
rolling mills: (a) two- 
high rolling mill; (b) 
three-high rolling mill; 

(c) four-high rolling mill; 

(d) multihigh rolling mill; 

(e) universal rolling mill 




3!^ — ^y}. 



5.2 Rolling 151 

FIGURE 5.8 Original Distorted 

shape roll 

..— y— -y_. 

Rolling thin sheets with 
small-diameter rolls 

Cross section of 
the sheet 

Such small-diameter working rolls make a drive practically impossible. They are, 
therefore, driven by friction through an intermediate row of driving rolls that are, in 
turn, supported by a row of backup rolls. This arrangement involves a cluster of either 
12 or 20 rolls, resulting in exceptional rigidity of the whole roll system and almost 
complete absence of working-roll deflections. An equivalent system that is sometimes 
used is the planetary rolling mill, in which a group of small-diameter working rolls ro- 
tate around a large, idle supporting roll on each side of the work. 

Universal rolling mills. Universal rolling mills are used for producing blooms from in- 
gots and for rolling wide-flange H beams (Gray's beams). In this type of mill, there are 
vertical rolls in addition to the horizontal ones. The vertical rolls of universal mills (for 
producing structural shapes) are idle and are arranged between the bearing chocks of 
the horizontal rolls in the vertical plane. 

The Range of Rolled Products 

The range of rolled products is standardized in each country in the sense that the 
shape, dimensions, tolerances, properties, and the like are given in a standard specifi- 
cations handbook that differs from country to country. The whole range of rolled prod- 
ucts can generally be divided into the following four groups. 

Structural shapes or sections. The first group includes general-purpose sections like 
round and square bars; angles; channel, H, and I beams; and special sections (with in- 
tricate shapes) like rails and special shapes used in construction work and industry. 
Figure 5.9 shows a variety of sections that belong to this group. These products are 
rolled in either rail mills or section mills, where the body of each roll has grooves 
called passes that are made in the bodies of the upper and lower rolls in such a man- 
ner as to lie in the same vertical plane. They are used to impart the required shape to 
the work. This process is carried out gradually (i.e., the stock is partly deformed at 
each stand, or pass, in succession). The skill of a rolling engineer is to plan and con- 
struct the details of a system of successive passes that ensures the adequate rolling of 
blanks into the desired shape. This operation is called roll pass design. Figure 5.10a il- 
lustrates the roll passes for producing rails; Figure 5.10b, those for producing an I 


5 Metal Forming 


Some structural shapes 
or sections produced by 










Equal-sided angle 








I beam 



/^»77> Channel beam ^777^77* 


Plates and sheets. Plates and sheets are produced in plate and sheet mills for the hot 
rolling of metal and in cold reduction mills for the production of cold-rolled coils, 
where multihigh rolling mills are employed, as previously mentioned. This group of 
products is classified according to thickness. A flat product with a width ranging from 
5/32 inch (4 mm) up to 4 inches (100 mm) is called a plate, whereas wider and thin- 
ner flat stocks are called sheets. 

Special-purpose rolled shapes. This group includes special shapes, one-piece rolled 
wheels, rings, balls, ribbed tubes, and die-rolled sections in which the cross section 
of the bar varies periodically along its length. These kinds of bars are used in the 
machine-building industry and in the construction industry for reinforcing concrete 
beams and columns. Figure 5.11a shows the sequence of operations in manufacturing 
a rolled wheel for railway cars; Figure 5.11b, the wheel during the final stage in the 
rolling mill. 

Seamless tubes. The process of manufacturing seamless tubes involves two steps: 

1. Piercing an ingot or a roughened-down round blank to form a thick-walled shell 

2. Rolling the obtained shell into a hollow thin-walled tube having the desired diam- 
eter and wall thickness 

In the first step, the solid blank is center-drilled at one end, heated to the appropriate 
temperature, and then placed in the piercing mill and forced into contact with the 
working rolls. There are several types of piercing mills, but the commonly used one 
has barrel-shaped rolls. As Figure 5.12 shows, the axes of the two rolls are skew lines, 
each deviating with a small angle from the direction of the blank axis. Also, the two 
rol)s rotate in the same direction, forcing the blank to rotate and proceed against a 
mandrel. A hole is formed and becomes larger; finally, a rough tube is obtained. The 
milling stand is provided with side rollers for guiding the blank and the formed rough 
tube during this operation. In the second step, the hollow shell (rough tube) is usually 
forced over another mandrel, and the combination is longitudinally rolled at their hot 

5.2 Rolling 


FIGURE 5.10 

Roll passes: (a) for producing rails; (b) for producing an I beam 



A few 




state between grooved rolls. Mills of different types are used, including continuous, 
automatic, and pilger mills. Finally, a sizing operation may be performed, between siz- 
ing rolls and without the use of a mandrel, at room temperature in order to improve the 
properties and finish of the tubes. 

Lubrication in Rolling Processes 

Friction plays a very important role in a rolling process and has some beneficial ef- 
fects, provided that it is not excessive. In fact, it is responsible for shifting the work 
between the rolls and should not, therefore, be eliminated or reduced below an appro- 
priate level. This is an important point to be taken into account when choosing a lu- 
bricant for a rolling process. 


Metal Forming 

FIGURE 5.11 

The production of a 
railway car wheel: (a) 
sequence of stages; (b) 
wheel in final stage in 



In the cold rolling of steel, fluid lubricants of low viscosity are employed, but 
paraffin is suitable for nonferrous materials like aluminum or copper alloys to avoid 
staining during subsequent heat treatment. On the other hand, hot rolling is often car- 
ried out without lubricants but with a flood of water to generate steam and break up 
the scales that are formed. Sometimes, an emulsion of graphite or graphited grease is 

Defects in Rolled Products 

A variety of defects in the products arise during rolling processes. A particular defect 
is usually associated with a particular process and does not arise in other processes. 
Following are some of the common defects in rolled products. 

Edge cracking. Edge cracking occurs in rolled ingots, slabs, or plates and is believed 
to be caused by either limited ductility of the work metal or uneven deformation, es- 
pecially at the edges. 

FIGURE 5.12 

The production of 
seamless tubes by 

5.3 Metal Drawing 155 

FIGURE 5.13 

Alligatoring when rolling 

aluminum slabs 

Arc of \ / \ 

contact ~~~-V-^^ ^N- — \ 


Alligatoring. Figure 5.13 shows the defect of alligatoring, which is less common 
than it used to be. It usually occurs in the rolling of slabs (particularly aluminum al- 
loys), where the workpiece splits along a horizontal plane on exit, with the top and 
bottom parts following the rotation of their respective rolls. This defect always oc- 
curs when the ratio of slab thickness to the length of contact falls within the range 
1.4 to 1.7. 

Folds. Folds are defects occurring during plate rolling when the reduction per pass is 
too small. 

Laminations. Laminations associated with cracking may develop when the reduction 
in thickness is excessive. 


Drawing is basically a forming process that involves pulling a slender semifinished 
product (like wire, bar stock, or tube) through a hole of a drawing die. The dimen- 
sions of that hole are smaller than the dimensions of the original material. Metals 
are usually drawn in their cold state, and the required shape may be achieved in a 
single drawing operation or through several successive drawing operations, in which 
case the diameters of the holes are successively decreasing. Sometimes, annealing 
is carried out between the drawing operations to relieve the metal from work- 
hardening. Accurate dimensions, good surface quality, increased strength and hard- 
ness, and the possibility of producing very small sections are some advantages of 
the drawing process. The drawing process has, therefore, wide industrial application 
and is used for manufacturing thin wires, thin-walled tubes, and components with 
sections that cannot be made except by machining. It is also used for sizing hot- 
rolled sections. 


Metal Forming 

Preparing the Metal for Drawing 

Before being subjected to the drawing process, metal blanks (wires, rods, or tubes) are 
heat treated and then cleaned of scales that result from that operation. Descaling is usu- 
ally done by pickling the heat-treated metal in acid solutions. Steels are pickled in ei- 
ther sulfuric or hydrochloric acid or a mixture of both; copper and brass blanks are 
treated in sulfuric acid, whereas nickel and its alloys are cleaned in a mixture of sul- 
furic acid and potassium bichromate. After pickling, the metal is washed to remove 
any traces of acid or slag from its surface. The final operation before drawing is dry- 
ing the washed blanks at a temperature above 212°F (100°C). This eliminates the 
moisture and a great deal of the hydrogen dissolved in the metal, thus helping to avoid 
pickling brittleness. 

If steel is to be subjected to several successive drawing passes, its surface should 
then be conditioned for receiving and retaining the drawing lubricant. Conditioning is 
performed directly after pickling and can take the form of sulling, coppering, phos- 
phating, or liming. In sulling, the steel rod is given a thin coat of iron hydroxide, which 
combines with lime and serves as a carrier for the lubricant. Phosphating involves ap- 
plying a film of iron, manganese, or zinc phosphates to which lubricants stick very 
well. Liming neutralizes the remaining acid and forms a vehicle for the lubricant. Cop- 
pering is used for severe conditions and is achieved by immersing the steel rods (or 
wires) in a solution of vitriol. All conditioning operations are followed by drying at a 
temperature of about 650°F (300°C) in special chambers. 

Wire Drawing 

Drawing dies. A die is a common term for two parts: the die body and the die holder. 
Die bodies are made of cemented carbides or hardened tool steel, whereas die holders 
are made of good-quality tool steel that possesses high toughness. The constructional 
details of a die are shown in Figure 5.14. It can be seen from the figure that the die 
opening involves four zones: entry, working zone, die bearing, and exit. The entry zone 
allows the lubricant to reach the working zone easily and also protects the wire (or rod) 
against scoring by sharp edges. The working zone is conical in shape and has an apex 

FIGURE 5.14 

The constructional 
details of a drawing die 

Q *~ Drawing force 

5.3 Metal Drawing 


angle that ranges between 6° and 24°, depending upon the type of work and the metal 
being drawn. The die bearing, sometimes called the land, is a short cylindrical zone in 
which a sizing operation is performed to ensure accuracy of the shape and dimensions 
of the end product. The exit zone provides back relief to avoid scoring of the drawn 
wire (or rod). In a wire-drawing operation, the end of the wire is pointed by swaging 
and then fed freely into the die hole so that it appears behind the die. This pointed end 
is gripped by the jaws of a carriage that pull the wire through the die opening, where 
it undergoes reduction in cross-sectional area and elongation in length. 

Draw benches. A wire-drawing operation usually involves the use of multidie draw 
benches, where the wire passes through a series of draw plates. First, the wire leaves the 
coil and passes through the first drawing die. Then, it is wound two or three turns around 
a capstan (drum) before it enters the next drawing die. A typical draw bench of this type 
with six draw plates is shown in Figure 5.15. In practice, a bench may include from 2 up 
to 22 draw plates, and the wire leaving the last die may attain a velocity of 9800 feet per 
minute (50 m/s). The capstan drives are designed to provide not only forward pull after 
each pass but also backward pull to the wire before it enters the next drawing die. 

Lubrication. Lubrication reduces the required drawing force and the energy con- 
sumed during the process, increases the service life of the die, and allows a smoother 
wire surface to be obtained. Various kinds of soap are used as lubricants in wire- 
drawing processes. Examples are sodium soap or calcium stearate, which is picked up 
by the wire from a soap box adjacent to the die. Although they are difficult to apply 
and remove, polymers are also used as solid lubricants, especially in severe conditions, 
as in the case of drawing hard alloys or titanium. Various kinds of mineral and veg- 
etable oils containing fatty or chlorinated additives are also used as drawing lubricants. 

Mechanics of wire drawing. The state of stress during the wire-drawing process (see 
Figure 5.14) involves compressive forces along two of the directions and tension along 
the third one. An approximate but simple estimate of the drawing force can be given 
by the following equation: 

F = a { xYx£n(^ 


FIGURE 5.15 

A typical multidie draw 

Draw plates (dies) 

Two or three 
j capstan drums 
and draw plates 

wire coil 


158 5 Metal Forming 

where: a is the original area 
a f is the final area 
Y is the mean yield stress of the metal 

In Equation 5.9, the ratio aja f is called the coefficient of elongation, or simply the 
drawing ratio. In industrial practice, it is usually about 1.25 up to 1.3. Another conju- 
gate term that is used in drawing processes is the reduction, given by the following 

reduction r = a °~ a{ x 100 (5.10) 


The theoretically obtained maximum value for the reduction is 64 percent; however, it 
usually does not exceed about 40 percent in industry. 

Defects in wire drawing. Structural damage in the form of voids or cracks occurs in 
different forms in wire-drawing processes under certain conditions. Following are 
some of the defects encountered: 

1. Internal bursts in wire, taking the form of repeating internal cup and cone frac- 
tures (cuppy wire), usually occur when drawing heavily cold-worked copper 
under conditions of light draft and very large die angles. 

2. Similar centerline arrowhead fractures occur if the blank is a sheet and when the 
die angle and reduction produce severe tension on the centerline. 

3. Transverse surface cracking may occur as a result of longitudinal tension stresses 
in the surface layers. 

Tube Drawing 

Diameter and thickness of pipes can be reduced by drawing. Figure 5.16 illustrates the 
simplest type of tube drawing. The final tube thickness is affected by two contradict- 
ing factors. The longitudinal stress tends to make the wall thinner, whereas the cir- 
cumferential stress thickens it. If a large die angle is used, the thinning effect will 

The technique shown in Figure 5.17 of using a fixed plug reduces the tube diam- 
eter and controls its thickness. However, a disadvantage of this type of tube drawing is 
the limitation imposed on the length of the tube by the length of the mandrel. When 
tubes having longer length are to be drawn, a floating mandrel like that shown in Fig- 
ure 5.18 is then employed. Another method that has gained widespread application is 
using a removable mandrel like that shown in Figure 5.19. 


Extrusion involves forcing a billet that is enclosed in a container through an open- 
ing whose cross-sectional area and dimensions are smaller than those of the original 
billet. The cross section of the extruded metal will conform to that of the die open- 

FIGURE 5.16 

Simplest type of tube 

FIGURE 5.17 

Tube drawing using a 
fixed plug 

- — Pull 

FIGURE 5.18 

Tube drawing using a 
floating mandrel 




ing. Historically, extrusion was first used toward the end of the eighteenth century for 
producing lead pipes. It later gained widespread industrial application for processing 
nonferrous metals and alloys like copper, brass, aluminum, zinc, and magnesium. Re- 
cently, with the modern developments in extrusion techniques, lubricants, and tool- 
ing, other metals, such as steels, titanium, refractory metals, uranium, and thorium, 
can also be extruded successfully. The stock used for extrusion is mainly a cast ingot 
or a rolled billet. Any surface defects in the original billets must be removed by saw- 
ing, shearing, turning, or any other appropriate machining operation before the ex- 


Metal Forming 

FIGURE 5.19 

Tube drawing using a 
removable mandrel 

trusion process is performed. Extrusion carried out when the billets are at their cold 
state is known as cold extrusion; when they are at elevated temperatures, it is known 
as hot extrusion. In this latter case, the container, the die, and the pressing plunger 
must be heated to a temperature of about 650°F (350°C) prior to each extrusion 

Types of Extrusion 

Direct extrusion. Direct extrusion is used in the manufacture of solid and hollow 
slender products and for structural shapes that cannot be obtained by any other metal 
forming process. Figure 5.20 illustrates the working principles of this method, and Fig- 
ure 5.21 shows the details of an extrusion die arrangement for producing channel sec- 
tions. As can be seen, during an extrusion process a billet is pushed out of the die by 
a plunger and then slides along the walls of the container as the operation proceeds. At 
the end of the stroke, a small piece of metal (stub-end scrap) remains unextruded in the 

FIGURE 5.20 

Principles of direct 
extrusion for producing 
solid objects 



5.4 Extrusion 


The extruded product is separated by shearing, and the stub-end is then ejected out 
of the container after the plunger is withdrawn. Also, the leading end of the extruded 
product does not undergo enough deformation. It is, therefore, poorly shaped and must 
be removed as well. Obviously, the efficiency of material utilization in this case is low, 
and the waste can amount to 10 or even 15 percent, as opposed to rolling, where the 
waste is only 1 to 3 percent. This makes the productivity of direct extrusion quite in- 
ferior to that of rolling. 

FIGURE 5.21 

Typical extrusion die 
arrangement for 
producing channel 



5 Metal Forming 

FIGURE 5.22 

Direct extrusion for 
producing hollow 

Figure 5.22 illustrates the technique used for producing hollow sections and tubes. 
As can be seen, a mandrel or a needle passes freely through a hole in the blank and the 
die opening. If the die opening is circular, an annular clearance between the die open- 
ing and the mandrel results. When the metal is extruded through the annular clearance, 
it forms a tube. A hole has to be pierced or drilled into the original blank before it is 

Based on this discussion, it is clear that the conventional extrusion process has the 
advantages of high-dimensional accuracy and the possibility of producing complex 
sections from materials having poor plasticity. On the other hand, its disadvantages in- 
clude low productivity, short tool life, and expensive tooling. Therefore, the process is 
usually employed for the manufacture of complex shapes with high-dimensional accu- 
racy, especially when the material of the product has a low plasticity. Figures 5.23 and 
5.24 show some extruded sections and parts, and Figure 5.25 shows some final prod- 
ucts assembled from extruded sections. 

FIGURE 5.23 

Some extruded 
sections (Courtesy of 
Midwest Aluminum, 
Inc., Kalamazoo, 

5.4 Extrusion 


FIGURE 5.24 

Some extruded parts 
(Courtesy of Midwest 
Aluminum, Inc., 
Kalamazoo, Michigan) 



Indirect extrusion. In indirect extrusion, the extrusion die is mounted on a hollow 
ram that is pushed into the container. Consequently, the die applies pressure to the bil- 
let, which undergoes plastic deformation. As shown in Figure 5.26, the metal flows out 
of the die opening in a direction opposite to the ram motion. There is almost no slid- 
ing motion between the billet and the container walls. This eliminates friction, and the 
extrusion load will be lower than that required in forward direct extrusion by about 30 
percent. Also, the amount of waste scrap is reduced to only 5 percent. Nevertheless, in- 
direct extrusion finds only limited application due to the complexity and the cost of 
tooling and press arrangement required. 

Another indirect extrusion method, usually called backward or reverse extrusion, 
used in manufacturing hollow sections is shown in Figure 5.27. In this case, the metal 
is extruded through the gap between the ram and the container. As in indirect extrusion 
for solid objects, the ram and the product travel in opposite directions. 


5 Metal Forming 

FIGURE 5.25 

Some products assembled from extruded sections (Courtesy of Midwest Aluminum, Inc. 
Kalamazoo, Michigan) 

Hydrostatic extrusion. A radical development that eliminates the disadvantages of 
cold extrusion (like higher loads) involves hydrostatic extrusion. Figure 5.28 illus- 
trates the basic principles of this process, where the billet is shaped to fit the die and 
surrounded by a high-pressure hydraulic fluid in a container. When the plunger is 
pressed, it increases the pressure inside the container, and the resulting high pressure 
forces the billet to flow through the die. Friction between the billet and the container 
is thus eliminated, whereas friction between the billet and the die is markedly re- 

FIGURE 5.26 

Indirect extrusion for 
producing solid objects 

5.4 Extrusion 


FIGURE 5.27 

Indirect (backward) 
extrusion for producing 
hollow objects 

FIGURE 5.28 

Principles of 
hydrostatic extrusion 



duced. Also, the buckling effect of longer billets is eliminated because virtually the 
entire length of the billet is subjected to hydrostatic pressure. This makes it possible 
to extrude very long billets. 

Impact extrusion. Impact extrusion involves striking a cold slug of soft metal (like 
aluminum) that is held in a shallow die cavity with a rapidly moving punch, thus caus- 
ing the metal to flow plastically around the punch or through the die opening. The slug 
itself is a closely controlled volume of metal that is lubricated and located in the die 
cavity. The press is then activated, and the high-speed punch strikes the slug. A fin- 
ished impacted product is extruded with each stroke of the press. These products are 
not necessarily cylindrical with a circular cross section. In fact, the range of shapes 
possible is very broad, including even irregular symmetrical shapes, as shown in Fig- 
ures 5.29 and 5.30. There are three types of the impact extrusion processes: forward, 
reverse, and combination (the names referring to the direction of motion of the de- 
forming metal relative to that of the punch). 

Figure 5.31 illustrates the basic principles of reverse impact extrusion. It is used 
for manufacturing hollow parts with forged bases and extruded sidewalls. The flowing 
metal is guided only initially; thereafter, it goes by its own inertia. This results in the 
elimination of friction and, therefore, an appreciable reduction in the load and energy 
required. A further advantage is the possibility of producing thinner walls. 

The principles of forward impact extrusion are illustrated in Figure 5.32. It is 
mainly employed in producing hollow or semihollow products with heavy flanges and 
multiple diameters formed on the inside and outside. Closer wall tolerances, larger 
slenderness ratios, better concentricities, and sound thinner sections are among the ad- 
vantages of this process. 


5 Metal Forming 

FIGURE 5.29 

Some shapes produced 
by impact extrusion 
(Courtesy of Metal 
Impact Corporation, 
Rosemont, Illinois) 

Complex shapes can be produced by a combination of the two preceding 
processes, which are performed simultaneously in the same single stroke, as shown in 
Figure 5.33. Like the other impact extrusion methods, this process has the advantage 
of cleaner product surfaces, elimination of trimming or further machining operations, 
and higher strength of the parts obtained. 

Mechanics of Extrusion 

We can clearly see (from Figure 5.20) that an element of the deforming metal being 
extruded is subjected to a state of stress involving triaxial compression. This all-around 
high pressure results in a marked improvement in the plasticity of the metal. Conse- 
quently, extrusion can be employed when working metal having poor plasticity, as op- 
posed to rolling or wire drawing, where only ductile metals can be formed (worked). 

5.4 Extrusion 


FIGURE 5.30 

Some components 
produced by impacting 
(Courtesy of Metal 
Impact Corporation, 
Rosemont, Illinois) 

Load requirement. For the sake of simplicity, it is sometimes assumed that the 
processes involve ideal deformation without any friction. The extrusion pressure can 
then be given by the following equation: 

Pextrus.on = Y X €n^ =YX (,lR 


FIGURE 5.31 

Principles of reverse 
impact extrusion 




5 Metal Forming 

FIGURE 5.32 

Principles of forward 
impact extrusion 

FIGURE 5.33 

Combination impacting 








where: a Q is the original cross-sectional area 

a f is the final cross-sectional area after extrusion 

R is the extrusion ratio 

Y is the mean yield stress of the metal 

The extrusion load is, therefore, 

F — p x a 


These equations are used to give only rough estimates because actual extrusion 
processes involve friction and the lack of homogeneous deformation of the metal, as 
will be seen later. Therefore, research workers developed several empirical formulas to 
give the extrusion pressure as a function of the extrusion ratio and the mechanical 
properties of the metal. A convenient formula was proposed by W. Johnson (the emi- 
nent British researcher in the area of metal forming) as follows: 

= Y 0.8+ l.5€n 


1 -r 


5.4 Extrusion 


In Equation 5.13, r is the reduction given by 
reduction r = — ^ — - 


Metal flow and deformation. To study metal flow, let us consider extruding a split bil- 
let involving two identical halves, with a rectangular grid engraved on the meridional 
plane of each half. The separation surface is covered with lanolin or a similar appro- 
priate material to prevent welding or sticking of the two halves during the process. 
After extruding the split billet, the two halves are separated, and the distortion of the 
grid can be investigated. Figure 5.34 shows the grid after extrusion. We can see that 
the units of the grid, which were originally square in shape, became parallelograms, 
trapezoids, and other shapes. The following can also be observed: 

1. The velocity of the core is greater than that of the outer layers. 

2. The outer layers are deformed to a larger degree than the core. 

3. The leading end of the extruded part is almost undeformed. 

4. The metal adjacent to the die does not flow easily, leading to the initiation of 
zones where little deformation occurs. These zones are called dead-metal zones. 

In fact, the preceding method for studying the metal flow is usually used with models 
made of wax, plasticine, and lead to predict any defect that may occur during the ac- 
tual process so that appropriate precautions can be taken in advance. 

Lubrication in Extrusion 

Friction at the billet-die and billet-container interfaces increases the load and the 
power requirement and reduces the service life of the tooling. For these reasons, lubri- 
cants are applied to the die and container walls. 

As in wire drawing, soaps and various oils containing chlorinated additives or 
graphite are used as lubricants in cold extrusion of most metals, whereas lanolin is usu- 
ally used for the softer ones. For hot extrusion of mild steel, graphite is an adequate lu- 
bricant. It is not, however, recommended for high-temperature extrusions, such as 
extruding molybdenum at 3250°F (1800°C); in this case, glass is the most successful 

Defects in Extruded Products 

Defects in extruded parts usually fall into one of three main categories: surface or in- 
ternal cracking, sinking (piping), and skin-inclusion defects. Cracking is caused by 
secondary tensile stresses acting within a material having low plasticity. Cracking can 
occur on the surface of a relatively brittle material during the extrusion process, and it 

FIGURE 5.34 

Distorted grid indicating 
metal flow in extrusion 


Metal Forming 

may also occur in the form of fire-tree or central bursts when extruding materials like 
bismuth, magnesium, 60/40 brass, steel, and brittle aluminum alloys. Piping involves 
sinking of the material at the rear of the stub-end. This defect is usually encountered 
toward the end of the extrusion stroke, especially when the original billets are rela- 
tively short. Skin-inclusion defects may take different forms, depending upon the de- 
gree of lubrication and the hardness of the surface layer of the original stock. When 
extruding lubricated billets of high-copper alloys, the surface skin will slide over the 
container wall and then penetrate the billet, as illustrated in Figure 5.35, where the 
three different extrusion defects are sketched. 

FIGURE 5.35 

Three different defects 
occurring in an 
extrusion process 





Design Considerations 

Conventional extrusions. When making parts that have constant cross sections, the 
extrusion process is usually more economical and faster than machining, casting, or 
fabricating the shapes by welding (or riveting). Also, the designer of the extruded sec- 
tion is relatively free to put the metal where he or she wants. Nevertheless, there are 
some design guidelines that must be taken into consideration when designing an ex- 
truded section: 

1. The circle size (i.e., the diameter of the smallest circle that will enclose the ex- 
trusion cross section) can be as large as 31 inches (775 mm) when extruding light 

2. Solid shapes are the easiest to extrude. Semihollow and hollow shapes are more 
difficult to extrude, especially if they have thin walls or include abrupt changes in 
wall thickness. 

3. Wall thicknesses must be kept uniform. If not, all transitions must be streamlined 
by generous radii at the thick-thin junctions. 

4. Sharp corners at the root of a die tongue should be avoided when extruding semi- 
hollow sections. 

5. A complicated section should be broken into a group of simpler sections that are as- 
sembled after the separate extrusion processes. In such a case, the sections should 
be designed to simplify assembly; for example, they should fit, hook, or snap to- 
gether. Screw slots or slots to receive other tightening material, such as plastic, may 
also be provided. 

5.4 Extrusion 


Figure 5.36 illustrates some recommended designs for assembling extruded alu- 
minum sections. Figure 5.37 illustrates and summarizes some recommended designs as 
well as those to be avoided as general guidelines for beginning designers. 

Aluminum impact extrusions. In order to accomplish good designs of aluminum im- 
pact extrusions, all factors associated with and affecting the process must be taken into 
account. Examples are alloy selection, tool design, lubrication, and, of course, the gen- 
eral consideration of mechanical design. Following are some basic guidelines and de- 
sign examples: 

1. Use alloys that apply in the desired case and have the lowest strength. 

2. An impact extrusion should be symmetrical around the punch. 

3. Threads, cuts, projections, and the like are made by subjecting the impact extru- 
sions to further processing. 

4. For reverse extrusions, the ratio of maximum length to internal diameter must not 
exceed 8 to avoid failure of long punches. 

5. A small outer-corner radius must be provided for a reverse extrusion, but the 
inner-corner radius must be kept as small as possible (see Figure 5.38a). 

6. The thickness of the bottom near the wall must be 15 percent greater than the 
thickness of the wall itself to prevent shear failure (see Figure 5.38b). 

7. The inside bottom should not be completely flat. To avoid the possibility of the punch 
skidding on the billet, only 80 percent of it at most can be flat (see Figure 5.38c). 

FIGURE 5.36 

Some recommended 
designs for assembling 
extruded aluminum 
sections (Courtesy of 
the Aluminum 
Association, Inc., 
Washington, D.C.) 


Lap joints 
H 2 


Held by 

self threading 


Lap-lock joints 
V\ 5 

Side entry Edge entry Dovetail 
sliding fit 

Cylindrical sliding fits 
H 8 9 


Adjustable As adapted to 
stair riser 


Metal Forming 

FIGURE 5.37 

Some design 
considerations for 
conventional extrusions 
(Courtesy of the 
Aluminum Association, 
Inc., Washington, D.C.) 





sssssssssgssg g 


^^ ^^ 

m\ \s\ \\ s\ \s \ \\ w 

JWs> g^^i 


•- c 



fcmmmvMs^ a 

^^ ^^ 





^^vvvv ^\\\\v\v^ 


When designing, visualize the die and tongue that will be 
necessary to produce a semi-hollow shape. By keeping 
the void symmetrical you lessen the chances that the die 
tongue may break. 

The preceding cross section has been further improved. 
The die tongue is now less likely to snap off. 


Further improvement results if outline can be changed to 
reduce area enclosed. Reduced area means less pressure 
on the tongue; easier extrusion. 


Hollow and multi-hollow extruded shapes are usually 
much more costly than the simple solid shape. Also less 
metal has been used. 


Metal dimensions are more easily held than gap or angle 

dimensions. Web also allows thinner wall sections in this 


The hollow condition of the "redesigned" part can be 

avoided by making the component in two pieces as shown 

by the dotted line. 


Transitions should be streamlined by a generous radius 

at any thick-thin junction. 


The preceding shape can be further improved by 

maintaining uniform wall thickness. 

In addition to using more metal, thick-thin junctions 

giv rise to distortion, die breakage or surface defects 

on the extrusion. 

Wide, thin sections can be hard to straighten after 
extrusion. Ribs help prevent twisting. 

5.4 Extrusion 


FIGURE 5.38 

Some design 
considerations for 
impact extrusions: (a) 
corner radii for reverse 
extrusion; (b) thickness 
of the bottom near the 
wall; (c) inside bottom; 
(d) ribs; (e) multiple- 
diameter parts 
(Courtesy of the 
Aluminum Association, 
Inc., Washington, D.C.) 

Preferably small 
as possible 

ftfr77777 f 

Appro x. "^ 
15° 3_ 


r- as 
small as 
*^ possible / J 



Bottom = 1.151V 

[*— 0.8D-*] 








8. External and internal bosses are permitted, provided that they are coaxial with the 
part. However, the diameter of the internal boss should not be more than 1/4 of the 
internal diameter of the shell. 

9. Longitudinal ribs, whether external or internal, on the full length of the impact ex- 
trusion are permitted. They should preferably be located in a symmetrical distrib- 
ution. However, the height of each rib must not exceed double the thickness of the 
wall of the shell (see Figure 5.38d). The main function of ribs is to provide stiff- 
ness to the walls of shells. They are also sometimes used for other reasons, such 
as to provide locations for drilling and tapping (to assemble the part), to enhance 
cooling by radiation, and to provide an appropriate gripping surface. 


Metal Forming 

10. An impact extrusion can have a wall with varying thickness along its length (i.e., 
it can be a multiple-diameter part). However, internal steps near the top of the 
product should be avoided because they cause excessive loading and wear of the 
punch (see Figure 5.38e). 

11. Remember that it is sometimes impossible to obtain the desired shape directly by 
impacting. However, an impact extrusion can be considered as an intermediate prod- 
uct that can be subjected to further working or secondary operations like machining, 
flange upsetting, nosing and necking, or ironing (see Figure 5.39a through c). 

Again, in addition to the preceding guidelines, general rules of mechanical design 
as well as common engineering sense are necessary for obtaining a successful design 
for the desired product. It would also be beneficial for the beginner to look at various 
designs of similar parts and to consult with experienced people before starting the de- 
sign process. Given in Figure 5.40 are sketches reflecting good design practice for 
some impact-extruded tubular parts and shells. 

FIGURE 5.39 

Some secondary 
operations after impact 
extruding: (a) flange 
upsetting; (b) nosing; 
(c) ironing (Courtesy of 
the Aluminum 
Association, Inc., 
Washington, D.C.) 



5.4 Extrusion 


FIGURE 5.40 

Sketches reflecting 
good design practice 
for some impact- 
extruded tubular parts 
and shells (Courtesy of 
Aluminum Association, 
Inc., Washington, D.C.) 

Tubular Parts 

<^yyyyyyyyyyyy ^A^yyyyM ^^^_ 


Flanged tube with open end. 


Cup and tube assembly, 
extruded as single piece. 

> /SSS/SSSSS///////////// 77Z. 

'/ss;;;s>;;;;s s/////w/777; 

Flange end closed. 

W ////////// //// && 


Flange with multiple 
step-down diameters. 

Flanged tube with 
multiple diameters. 

Partially closed end tube 
with heavy flange. 

r ysss;ss;y>y's;/s;;;s>;s;;s ss//sS;/;, 

^sssssssssss s//////;;/;. 



Combination impact with the 
flange at the midpoint. Such an 
impact also serves as a transition 
from one diameter to another. 
Wall thicknesses can also be varied. 


Uniform wall thickness 
with flanged end open. 

Outside longitudinal ribs can be 
spaced equally or in symmetrical 
patterns. Ribs may be extended 
to become cooling fins. 

Short recessed ribs in bottom can be 
used for tool insertions, drive, etc. 

>///>// //>//>//*/?, 

An external boss can be combined 
with an internal center tube. 

,', 1 J > > > 1 > 1 > > / S J J / l-7-TT s 

r-T-j i >/>>>! ///>// m 

An integral center tube can be 
formed so that assembly and 
machining are not needed. 

Inside bosses can be produced 
as integral parts of the closed 
end. The side wall can have 
longitudinal internal ribs. 

Combination impact. 


Metal Forming 


The term forging is used to define the plastic deformation of metals at elevated tem- 
peratures into predetermined shapes using compressive forces that are exerted through 
dies by means of a hammer, a press, or an upsetting machine. Like other metal form- 
ing processes, forging refines the microstructure of the metal, eliminates the hidden de- 
fects such as hair cracks and voids, and rearranges the fibrous macrostructure to 
conform with the metal flow. It is mainly the latter factor that gives forging its merits 
and advantages over casting and machining. By successful design of the dies, the 
metal flow during the process can be employed to promote the alignment of the fibers 
with the anticipated direction of maximum stress. A typical example is shown in Fig- 
ure 5.41, which illustrates the fibrous macrostructure in two different crankshafts pro- 
duced by machining from a bar stock and by forging. As can be seen, the direction 
of the fibers in the second case is more favorable because the stresses in the webs 
when the crankshaft is in service coincide with the direction of fibers where the 
strength is maximum. 

A large variety of materials can be worked by forging. These include low-carbon 
steels, aluminum, magnesium, and copper alloys, as well as many of the alloy steels 
and stainless steels. Each metal or alloy has its own plastic forging temperature range. 
Some alloys can be forged in a wide temperature range, whereas others have narrow 
ranges, depending upon the constituents and the chemical composition. Usually, the 
forging temperatures recommended for nonferrous alloys and metals are much lower 
than those required for ferrous materials. Table 5.2 indicates the range of forging tem- 
peratures for the commonly used alloys. 

Forged parts vary widely in size ranging from a few pounds (less than a kilogram) 
up to 300 tons (3 MN) and can be classified into small, medium, and heavy forgings. 

FIGURE 5.41 

The fibrous 
macrostructure in two 
crankshafts produced 
by machining and by 

Produced by machining 
from a bar stock 

Produced by forging 

TABLE 5.2 

Forging temperature 
range for different 


Forging Temperature 

Low-carbon steel 

1450-2550°F (800-1400°C) 


645-900°F (340-480°C) 


645-800°F (340-430°C) 


800-1900°F (430-1040°C) 


1100-1700°F (590-930°C) 

5.5 Forging 177 

Small forgings are illustrated by small tools such as chisels and tools used in cutting 
and carving wood. Medium forgings include railway-car axles, connecting rods, small 
crankshafts, levers, and hooks. Among the heavier forgings are shafts of power-plant 
generators, turbines, and ships, as well as columns of presses and rolls for rolling 
mills. Small and medium forgings are forged from rolled sections (bar stocks and 
slabs) and blooms, whereas heavier parts are worked from ingots. 

All forging processes fall under two main types: open-die forging processes, in 
which the metal is worked between two flat dies, and closed-die forging processes, in 
which the metal is formed while being confined in a closed impression of a die set. 

Open-Die Forging 

Open-die forging is sometimes referred to as smith forging and is actually a develop- 
ment or a modern version of a very old type of forging, blacksmithing, that was prac- 
ticed by armor makers and crafts people. Blacksmithing required hand tools and was 
carried out by striking the heated part repeatedly by a hammer on an anvil until the de- 
sired shape was finally obtained. Nowadays, blacksmith forging is used only when low 
production of light forgings is required, which is mainly in repair shops. Complicated 
shapes having close tolerances cannot be produced economically by this process. 

The modern version of blacksmithing, open-die forging, involves the substitution 
of a power-actuated hammer or press for the arm, hand hammer, and anvil of the smith. 
This process is used for producing heavy forgings weighing up to more than 300 tons, 
as well as for producing small batches of medium forgings with irregular shapes that 
cannot be produced by modern closed-die forging. The skill of the operator plays an 
important role in achieving the desired shape of the part by manipulating the heated 
metal during the period between successive working strokes. Accordingly, the shape 
obtained is just an approximation of the required one, and subsequent machining is al- 
ways used in order to produce the part that accurately conforms to the blueprint pro- 
vided by the designer. 

Open-die forging operations. A smith-forging process usually consists of a group of 
different operations. Among the operations employed in smith forging are upsetting, 
drawing out, fullering, cutting off, and piercing. The force and energy required differ 
considerably from one operation to another, depending upon the degree of "confine- 
ment" of the metal being worked. Following is a brief description of some of these 

1. Upsetting. Upsetting involves squeezing the billet between two flat surfaces, thus 
reducing its height due to the increase in the cross-sectional area. As can be seen in 
Figure 5.42a, the state of stress is uniaxial compression. In practice, however, the 
billets' surfaces in contact with the die are subjected to substantial friction forces 
that impede the flow of the neighboring layers of metal. This finally results in a het- 
erogeneous deformation and in barreling of the deformed billet. To obtain uniform 
deformation, the billet-die interfaces must be adequately lubricated. 

2. Drawing out. In drawing out, the workpiece is successively forged along its length 
between two dies having limited width. This results in reducing the cross-sectional 


5 Metal Forming 

FIGURE 5.42 

Various smith-forging 
operations: (a) 
upsetting; (b) drawing 
out; (c) piercing a short 
billet; (d) piercing a 
long billet; (e) cutting 
off; (f) bending 




I — . — 1 L — [ — __l I _ _j _ - 1 


r >< 




area of the workpiece while increasing its length, as shown in Figure 5.42b. This 
operation can be performed by starting either at the middle or at the end of the 
workpiece. A large reduction in the cross-sectional area can be achieved by reduc- 
ing the feed of the workpiece. The bite (i.e., the length of feed before the working 
stroke) ranges between 40 and 75 percent of the width of the forging die. 

Piercing operation. A piercing operation is performed in order to obtain blind or 
through holes in the billet. A through hole can be pierced directly in a short billet 
in a single stroke by employing a punch and a supporting ring, as shown in Fig- 
ure 5.42c. On the other hand, billets with large height-to-diameter ratios are 
pierced while located directly on the die with the help of a piercer and possibly 
an extension piece as well, as shown in Figure 5.42d. In this latter case, the di- 
ameter of the piercer must not exceed 50 percent of that of the billet. For larger 
holes, hollow punches are employed. Also, holes can be enlarged by tapered 

5.5 Forging 


4. Cutting off. Cutting off involves cutting the workpiece into separate parts using a 
forge cutter or a suitable chisel. This is usually done in two stages, as can be seen 
in Figure 5.42e. 

5. Bending. In bending, thinning of the metal occurs on the convex side at the point 
of localized bending (where bending actually takes place). It is, therefore, recom- 
mended to upset the metal at this location before bending is performed, as shown 
in Figure 5.42f, in order to obtain a quality bend. 

Examples of open-die forged parts. As mentioned before, a part may require a series 
of operations so that it can be given the desired shape by smith forging. Following are 
some examples of smith-forged industrial components, together with the steps in- 
volved in the manufacture of each part: 

1. Large motor shaft. First, 24-inch-square (60 cm) steel ingots are rolled into 
square blooms, each having a 12-inch (30-cm) side. The blooms are then heated 
and hammered successively across the corners until the workpiece is finally 
rounded to a diameter of 10 inches (25 cm). These steps are illustrated in Fig- 
ure 5.43. 

2. Flange coupling. The sequence of operations is illustrated in Figure 5.44. There are 
two operations or stages involved, upsetting and heading. In heading, the flow of 
metal of most of the billet is restricted by using a ring-shaped tool. This process al- 
lows excellent grain flow to be obtained, which is particularly advantageous in car- 
rying tangential loads. 

FIGURE 5.43 

The production of a 
large motor shaft by 
smith forging 

Rotate - 


C? a 

l^»^^ J L^ — >. 1 L/ — ■ 1 

FIGURE 5.44 

The production of 
flange coupling by 
smith forging 



m m 

J 1 


Metal Forming 

3. Rings. A billet is first upset and is then subjected to a piercing operation. This is 
followed by an expanding operation using a mandrel to reduce the thickness of 
the ring and increase its diameter as required. Larger rings are usually expanded 
on a saddle. The steps involved in the process of ring forging are illustrated in 
Figure 5.45. 

Equipment for smith forging. Smaller billets are usually smith-forged using pneu- 
matic-power hammers. Larger components are worked in steam-power hammers (or 
large pneumatic hammers), whereas very large and heavy parts are produced by em- 
ploying hydraulic presses. Following is a brief description of smith-forging equipment: 

1. Steam-power hammers. A steam-power hammer consists mainly of the moving 
parts (including the ram, the rod, and the piston); a lifting and propelling device, 
which is a double-acting high-pressure steam cylinder; the housing or frame, which 
can be either an arch or an open type; and the anvil. Figure 5.46 illustrates the 
working principles. First, the piston and the other moving parts are raised by ad- 
mitting steam into the lower side of the cylinder (under the piston) through the 

FIGURE 5.45 

The production of large 
rings by smith forging 





FIGURE 5.46 

The working principles 
of a steam-power 



5.5 Forging 181 

sliding valve. When a blow is required, the lever is actuated; the sliding valve is ac- 
cordingly shifted to admit steam to the upper side of the cylinder (above the piston) 
and exhaust the steam that was in the lower side, thus pushing the moving parts 
downward at a high speed. In steam-power hammers, the velocity of impact can be 
as high as 25 feet per second (3 m/s), whereas the mass of the moving parts can be 
up to 11,000 slugs (5000 kg). The amount of energy delivered per blow is, there- 
fore, extremely large and can be expressed by the equation: 

E = '/2 mV 2 (5.15) 

where: E is the energy 

m is the mass of the moving parts 
V is the impact velocity 

Nevertheless, not all of that energy is consumed in the deformation of the work- 
piece. The moving parts rebound after impact, and the anvil will try to move in the 
opposite direction, thus consuming or actually wasting a fraction of the blow en- 
ergy. The ratio between the energy absorbed in deforming the metal to that deliv- 
ered by the blow is called the efficiency of a hammer and can be given by the 
following equation: 

M 7 

T\=— J —{\-K 2 ) (5.16) 

M + m 

where: M is the mass of the anvil 

A' is a factor that depends upon the elasticity of the billet 

The harder and more elastic the billet is, the higher that factor will be, and the 
lower the efficiency becomes. In addition, the hammer efficiency depends upon the 
ratio MI{M + m), or actually the ratio between the masses of the anvil and the mov- 
ing parts, which is taken in practice between 15 and 20. On the other hand, the 
value of K ranges between 0.05 and 0.25. 

2. Pneumatic-power hammers. There are two kinds of pneumatic-power hammers. 
The first kind includes small hammers in which the air compressor is built in; they 
usually have open frames because their capacity is limited. The second kind of 
pneumatic hammer is generally similar to a steam-power hammer in construction 
and operation, the only difference being that steam is replaced by compressed air (7 
to 8 times the atmospheric pressure). As is the case with steam, this necessitates 
separate installation for providing compressed air. Pneumatic hammers do not have 
some of the disadvantages of steam hammers, such as dripping of water resulting 
from condensation of leakage steam onto the hot billet. This may result in cracking 
of the part, especially when forging steel. 

3. Hydraulic presses. Heavy forgings are worked in hydraulic presses. The press in- 
stallation is composed of the press itself and the hydraulic drive. Presses capable of 
providing a force of 75,000 tons (750 MN) are quite common. Still, hydraulic 
presses that are commonly used in the forging industry have capacities ranging be- 
tween 1000 tons (10 MN) and 10,000 tons (100 MN). These presses can success- 

182 5 Metal Forming 

fully handle forgings weighing between 8 and 250 tons. The large-capacity presses 
require extremely high oil pressure in the hydraulic cylinders (200 to 300 times the 
atmospheric pressure). Because no pump can deliver an adequate oil discharge at 
that pressure level, this process is usually overcome by employing accumulators 
and intensifies that magnify the oil pressure delivered by the pump by a factor of 
40 or even 60. 

Planning the production of a smith-forged part. Before actually smith forging a part, 
all the details of the process must be thoroughly planned. This involves preparation 
of the design details, calculation of the dimensions and the weight of the stock and 
of the product, choosing the forging operations as well as their sequence, choosing 
tools and devices that will be used, and thinking about the details of the heating and 
cooling cycles. 

The first step in the design process is to draw the finished part and then obtain the 
drawing of the forging by adding a machining as well as a forging allowance all 
around. The machining allowance is the increase in any dimension to provide excess 
metal that is removed by machining. This subsequent machining is required to remove 
scales and the chilled, defected surface layers. The forging allowance is added mainly 
to simplify the shape of the as-forged part. It is always recommended to make the 
shape of a forging symmetrical and confined by plane and cylindrical surfaces. At this 
stage, a suitable tolerance is assigned to each dimension to bring the design process to 
an end. 

The next step is to choose the appropriate equipment. Two factors affect the deci- 
sion: the size of the forging and the rate of deformation (strain rate). Usually, forgings 
weighing 2 tons or more are forged in hydraulic presses. Also, small forging made of 
high-alloy steels and some nonferrous alloys must be forged on a press because they 
are sensitive to high strain rates that arise when using power-hammer forging. At this 
point, the manufacturing engineer is in a position to decide upon operations, tools, de- 
vices, and the like needed to accomplish the desired task. 

Closed-Die Forging 

Closed-die forging involves shaping the hot forging stock in counterpart cavities or im- 
pressions that have been machined into two mating halves of a die set. Under impact 
(or squeezing), the hot metal plastically flows to fill the die cavity. Because the flow of 
metal is restricted by the shape of the impressions, the forged part accurately conforms 
to the shape of the cavity, provided that complete filling of the cavity is achieved. 
Among the various advantages of closed-die forging are the greater consistency of 
product attributes than in casting, the close tolerances and good surface finish with 
minimum surplus material to be removed by machining, and the greater strength at 
lower unit weight compared with castings or fabricated parts. In fact, the cost of parts 
produced by machining (only) is usually two to three times the cost of closed-die forg- 
ings. Nevertheless, the high cost of forging dies (compared with patterns, for example) 
is the main shortcoming of this process, especially if intricate shapes are to be pro- 
duced. Therefore, the process is recommended for mass or large-lot production of steel 
and nonferrous components weighing up to about 900 pounds (350 kg). 

5.5 Forging 183 

Generally, there are two types of closed-die forging: conventional (or flash) die 
forging and flashless die forging. In conventional flash die forging, the volume of the 
slug has to be slightly larger than that of the die cavity. The surplus metal forms a flash 
(fin) around the parting line. In flashless forging, no fin is formed, so the process con- 
sequently calls for accurate control of the volume of the slug. If the slug is smaller than 
the required final product, proper filling of the die cavity is not achieved. On the other 
hand, when the size of the slug is bigger than that of the desired forging, excessive 
load buildup will eventually result in the breaking of the tooling and/or equipment. Ac- 
cordingly, flashless-forging dies are fitted with load-limiting devices to keep the gen- 
erated load below a certain safe value in order to avoid breakage of the tooling. 

In addition to shaping the metal in die cavities, the manufacturing cycle for a die- 
forged part includes some other related operations, such as cutting or cropping the 
rolled stock into slugs or billets, adequately heating the slugs, forging the slugs, trim- 
ming the flash (in conventional forging), heat treating the forgings, descaling, and, fi- 
nally, inspecting or quality controlling. The forging specifications differ from one 
country to another; however, in order to ensure the product quality, one or more of the 
following acceptance tests must be passed: 

1. Chemical composition midway between the surface and the center 

2. Mechanical properties 

3. Corrosion tests 

4. Nondestructive tests like magnetic detection of surface or subsurface hair cracks 

5. Visual tests such as macroetch and macroexamination and sulfur painting for steel 

Closed-die forging processes can be carried out using drop forging hammers, me- 
chanical crank presses, and forging machines. Factors such as product shape and tol- 
erances, quantities required, and forged alloys play an important role in determining 
the best and most economical equipment to be employed in forging a desired product 
as each of the processes has its own advantages and limitations. Following is a brief 
description of the different techniques used in closed-die forging. 

Drop forging. In drop forging, a type of closed-die forging, the force generated by the 
hammer is caused by gravitational attraction resulting from the free fall of the ram. The 
ram may be lifted by a single-acting steam (or air) cylinder or by friction rollers that en- 
gage a board tightly fastened to the ram. In this latter type, called a board hammer, once 
the ram reaches a predetermined desired height, a lever is actuated, the rollers retract, 
and the board and ram fall freely to strike the workpiece. Figure 5.47 illustrates the 
working principles. Whether a board hammer or single-acting steam hammer is used, 
accurate matching of the two halves of the die (i.e., the impressions) must be ensured. 
Therefore, the hammers employed in drop forging are usually of the double-housing (or 
arch) type and are provided with adequate ram guidance. The desired alignment of the 
two halves of the die is then achieved by wedging the upper half of the die onto the ram 
and securing the lower half onto a bolster plate that is, in turn, tightly mounted on the 
anvil. Also, the ratio of the weights of the anvil and the moving parts can go as high as 
30 to 1 to ensure maximum efficiency and trouble-free impact. 


5 Metal Forming 

FIGURE 5.47 

The working principles 
of a board hammer 






Drop-forging dies can have one, two, or several impressions, depending upon the 
complexity of the required product. Simple shapes like gears, small flywheels, and 
straight levers are usually forged in dies with one or two impressions, whereas prod- 
ucts with intricate shapes are successively worked in multiple-impression dies, thus 
making it possible to preshape a forging before it is forged into its final form. Opera- 
tions like edging, drawing out, fullering, and bending are performed, each in its as- 
signed impression. Finally, the desired shape is imparted to the metal in a finishing 
impression that has exactly the same shape as the desired product; its dimensions are 
slightly larger because shrinkage due to cooling down must be taken into account. As 
can be seen in Figure 5.48, a gutter for flash is provided around the finishing impres- 
sions. When properly designed, the gutter provides resistance to the flow of metal into 

FIGURE 5.48 

A gutter providing a 
space for excess metal 


A forging in 

the finishing 


Upper die half 

Lower die half 


5.5 Forging 


it, thus preventing further flow from the impression and forcing the metal to fill all the 
details, such as corners (which are the most difficult portions to fill). 

The drop-forging process may involve several blows so that the desired final 
shape of the forged part can be obtained. Lubricants are applied to ensure easy flow of 
the metal within the cavity and to reduce friction and die wear. As many as four blows 
may be needed while the part is in the finishing impressions, and the part should be 
lifted slightly between successive blows to prevent overheating of the die. Finally, the 
gas pressure forces the part out of the die. The number of blows delivered when the 
part is in the different preshaping impressions is 1 Vi to 2 times the number of blows 
while the part is in the finishing impression. This sequence of drop-forging operations 
is shown when forging a connecting rod. As can be seen in Figure 5.49, the heated 
stock is first placed in the fullering impression and then hammered once or twice to ob- 
tain local spreading of the metal on the expanse of its cross section. The stock is then 
transferred to the edging impression, where the metal is redistributed along its length 
in order to properly fill the finishing die cavities (i.e., metal is "gathered" at certain 
predetermined points and reduced at some other ones). This is usually achieved 
through a series of blows, together with turnovers of the metal, as required. The next 
operation in this sequence is bending, which may or may not be needed, depending 
upon the design of the product. The stock is then worked in the semifinishing, or 
blocking, impression before it is finally forged into the desired shape in the finishing 
impression. We can see that the blocking operation contributes to reducing the tool 
wear in the finishing impression by giving the part its general shape. 

Press forging. Press forging, which is usually referred to as hot pressing, is carried 
out using mechanical (crank-type) or hydraulic presses. These exert force at relatively 
slow ram travel, resulting in steadily applied pressure instead of impacting pressure. 

FIGURE 5.49 

A multiple-impression 
die and the forging 
sequence for a 
connecting rod 

Initial forging stock 




Metal Forming 

FIGURE 5.50 

Flash and flashless hot 




Flash hot pressing 


Flashless hot pressing 

The nature of metal deformation during hot pressing is, therefore, substantially differ- 
ent from that of drop forging. Under impact loading, the energy is transmitted into only 
the surface layers of the workpiece, whereas, under squeezing (steadily applied pres- 
sure), deformation penetrates deeper so that the entire volume of the workpiece simul- 
taneously undergoes plastic deformation. Although multiple-impression dies are used, 
it is always the goal of a good designer to minimize the number of impressions in a 
die. It is also considered good industrial practice to use shaped blanks or preforms, 
thus enabling the part to be forged in only a single stroke. 

Hot pressing involves both flash as well as flashless forging. In both cases, the 
forged part is pushed out of the die cavity by means of an ejector, as is illustrated in 
Figure 5.50. Examples of some hot-pressed parts are shown in Figure 5.51, which also 
shows the sequence of operations, the production rate, the estimated die life, and the 
approximate production cost. 

A characterizing feature of hot pressing is the accurate matching of the two halves 
of a die due to the efficient guidance of the ram. Also, the number of working strokes 
per minute can be as high as 40 or even 50. There is also the possibility of automating 
the process through mechanization of blank feeding and of forging removal. It can, 
therefore, clearly be seen that hot pressing has higher productivity than drop forging 
and yields parts with greater accuracy in terms of tolerances within 0.010 to 0.020 inch 
(0.2 up to 0.5 mm), less draft, and fewer design limitations. Nevertheless, the initial 
capital cost is higher compared with drop forging because the cost of a crank press is 
always higher than that of an equivalent hammer and because the process is economi- 
cal only when the equipment is efficiently utilized. The difficulty of descaling the 
blanks is another shortcoming of this process. However, this disadvantage can be elim- 
inated by using hydraulic descaling (using a high-pressure water jet) or can be origi- 
nally avoided by using heating furnaces with inert atmosphere. 

Die forging in a horizontal forging machine. Although originally developed for head- 
ing operations, the purpose of this machine has been broadened to produce a variety of 
shapes. For instance, all axisymmetric parts such as rods with flanges (with through 

5.5 Forging 


FIGURE 5.51 

Examples of hot-pressed parts 

Break lever 


Die life 

Cost in cents per piece 



40,000 pieces 





Sizing ^^O*' 


Bearing race 


SAE-5 00 

Die life 

Cost in cents per piece 





Backward extrusion 


h« 1 — *-i 



1 ~l 28.88+° 5 mm 


(Gas equipment) 




Die life 


Cost in cents per piece 


Trimming ™^ 1 " 

and blind holes) and/or side projections are commonly produced on horizontal forging 
machines. A rolled stock is cut to length, heated in a heating unit, and automatically 
fed to the machine. As can be seen in Figure 5.52, the hot part is then held by station- 
ary grips (actually a split die) and upset by an upsetting ram or header. The process in- 
volves mainly upsetting and gathering where the blank is first upset; then metal flows 
to fill the die cavity, as opposed to drop forging, where it is spread or flattened. In the 
return stroke, the upsetting ram retracts, and the part is removed or transferred to the 
next impression of the horizontal forging machine. It is obvious that a part can be 
forged in one or several cavities, depending upon the complexity of its shape. 

The main advantage of this process is the high production rate (up to 5000 parts 
per hour) due to the fact that it can be fully automated. Further advantages include the 


Metal Forming 

FIGURE 5.52 

Die forging in a 
horizontal forging 

Grip die 



Grip die 

elimination of the flash and the forging draft and the high efficiency of material uti- 
lization because the process involves little or no waste. 

Recent Developments in Forging 

Warm forging, high-energy-rate forging, and forming of metals in their mushy state are 
among the important developments in forging technology. These newly developed 
processes are usually carried out to obtain intricate shapes or unique structures that 
cannot be obtained by conventional forging processes. Following is a brief description 
of each of these processes, together with their advantages and disadvantages. 

Warm forging. Warm forging involves forging of the metal at a temperature some- 
what below the recrystallization temperature. This process combines some advantages 
of both the hot and the cold forming processes while eliminating their shortcomings. 
On one hand, increased plasticity and lower load requirements are caused by the rela- 
tively high forging temperature. On the other hand, improved mechanical properties, 
less scaling, and longer die life are due to the lower temperatures used as compared 
with those used with hot forging. 

High-energy-rate forging. The conventional forging process takes some time, during 
which the hot metal cools down and its resistance to deformation increases. As this 
does not occur with high-energy-rate forging (HERF), where the whole process is per- 
formed within a few thousandths of a second, the hot metal does not have enough time 
to cool down and heat is not dissipated into the surroundings. Therefore, HERF is very 
successful when forging intricate shapes with thin sections. A special HERF machine 
must be used. In fact, the Petro-Forge machine was developed at the Mechanical En- 
gineering Department of Birmingham University in England for this reason, and a 
bulky machine with the name Dynapak was developed in the United States. In the first 
case, the machine consists mainly of an internal-combustion (IC) cylinder integrated 
into the structure of a high-speed press. The IC cylinder is provided with a sudden re- 
lease valve that allows the platen attached to the piston to be fired instantaneously 
when the combustion pressure reaches a preset level. The four stages of the working 

5.5 Forging 


FIGURE 5.53 

The working cycle of the 

cycle of the Petro-Forge are shown in Figure 5.53. In the case of the Dynapak, high- 
pressure nitrogen in a power cylinder is used to push the platen downward. Installa- 
tions to produce and keep high-pressure gas are, therefore, required in this case. 

Forging of alloys in their mushy state. Forging alloys in their mushy state involves 
plastically forming alloys in the temperature range above the solidus line. Because 
an alloy at that temperature consists partly of a liquid phase, a remarkable decrease 
in the required forging load is experienced. The process also has some other merits, 
such as the high processing rate and the high quality of products compared with 
castings. Moreover, the friction at the billet-container interface has been found to be 
almost negligible. Nevertheless, the process is still considered to be in its experi- 
mental stage because of the instability of alloys having low solid fractions. Recently, 
it was reported that progress has been made toward solving this problem at the In- 
stitute of Industrial Science, Tokyo University, where the instability was overcome 


Charging Working stroke 

Return stroke 

Oil sump 

<*> J 

il mist 


At the beginning of the firing cycle the ram/piston assembly 
(A) is held at the top of its stroke by low pressure air in the 
back pressure chamber (B) closing the combustion chamber 
porting by the seal (C), this being a cylindrical projection on 
the top face of the piston (A). The exhaust valve (D) is open 
and pressure in the combustion chamber (E) is atmospheric. 
Upon pressing the firing button the fuel injection phase 
starts; the exhaust valve (D) is closed and the gaseous fuel 
is admitted into the combustion chamber (E) via the gas 
valve (F). 

Working stroke 

As soon as the force due to the combustion pressure acting 
on the small area (I) on top of the seal (C) is sufficiently 
large to overcome the opposed force due to the low back 
pressure in the space (B) acting on the annular lower face 
of the piston, the piston (A) starts to move. As a result the 
porting between the combustion chamber (E) and the cylinder 
is opened and the gases are permitted to expand to act over 
the whole piston area. This results in a large force surge acting 
on the piston/ram assembly which is accelerated downwards 
to impinge on the workpiece. 


After closing the gas valve (F) the combustion chamber is 
charged by admitting compressed air through the inlet valve 
(G). As soon as charging is completed, the inlet valve (G) is 
closed and the air/gas mixture is ignited by the spark plug 
(H). This results in a seven to eightfold rise of the pressure 
in the combustion chamber (E). 

Return stroke 

During the working stroke the back pressure in space (B) is 
intensified and consequently acts as a return spring as soon as 
the forming operation is completed, thus rapidly separating 
the dies. The return of the ram/piston assembly to its initial 
position is completed by the opening of the exhaust valve (D) 
which permits gases to leave through the duct (J). The cycle 
of operation is normally completed in one second. 

190 5 Metal Forming 

by dispersing a very fine alumina powder. This also yielded improved mechanical 
properties of forgings. 


For the proper planning of a forging process, it is important to know the deformation be- 
havior of the metal to be forged with regard to the resistance to deformation and any an- 
ticipated adverse effects, such as cracking. For this reason, the term forgeability was 
introduced and can be defined as the tolerance of a metal for deformation without failure. 
Although there is no commonly accepted standard test, quantitative assessment of the 
forgeability of a metal (or an alloy) can be obtained through one of the following tests. 

Upsetting test. The upsetting test involves upsetting a series of cylindrical billets 
having the same dimensions to different degrees of deformation (reductions in height). 
The maximum limit of upsettability without failure or cracking (usually peripheral 
cracks) is taken as a measure of forgeability. 

Notched-bar upsetting test. The notched-bar upsetting test is basically similar to the 
first test, except that longitudinal notches or serrations are made prior to upsetting. It 
is believed that this test provides a more reliable index of forgeability. 

Hot-impact tensile test. A conventional impact-testing machine fitted with a tension- 
test attachment is employed. A hot bar of the metal to be studied is tested, and the im- 
pact tensile strength is taken as a measure of forgeability. This test is recommended 
when studying the forgeability of alloys that are sensitive to high strain rates. 

Hot twist test. The hot twist test involves twisting a round, hot bar and counting the 
number of twists until failure. The greater the number of twists, the better the forge- 
ability is considered to be. Using the same bar material, this test can be performed at 
different temperatures in order to obtain the forging temperature range in which the 
forgeability of a metal is maximum. 

Forgeability of Some Alloys 

It is obvious that the results of any of the preceding tests are affected by factors like 
the composition of an alloy, the presence of impurities, the grain size, and the number 
of phases present. These are added to the effect of temperature, which generally im- 
proves forgeability up to a certain limit, where other phases start to appear or where 
grain growth becomes excessive. At this point, any further increase in temperature is 
accompanied by a decrease in forgeability. Following is a list indicating the relative 
forgeability of some alloys in descending order (i.e., alloys with better forgeability are 
mentioned first): 

1. Aluminum alloys 

2. Magnesium alloys 

3. Copper alloys 

4. Plain-carbon steels 

5.5 Forging 191 

5. Low-alloy steels 

6. Martensitic stainless steel 

7. Austenitic stainless steel 

8. Nickel alloys 

9. Titanium alloys 

10. Iron-base superalloys 

11. Cobalt-base superalloys 

12. Molybdenum alloys 

13. Nickel-base superalloys 

14. Tungsten alloys 

15. Beryllium 

Lubrication in Forging 

In hot forging, the role of lubricants is not just limited to eliminating friction and en- 
suring easy flow of metal. A lubricant actually prevents the hot metal from sticking to 
the die and meanwhile prevents the surface layers of the hot metal from being chilled 
by the relatively cold die. Therefore, water spray, sawdust, or liners of relatively soft 
metals are sometimes employed to prevent adhesion. Mineral oil alone or mixed with 
graphite is also used, especially for aluminum and magnesium alloys. Graphite and/or 
molybdenum disulfide are widely used for plain-carbon steels, low-alloy steels, and 
copper alloys, whereas melting glass is used for difficult-to-forge alloys like alloy 
steel, nickel alloys, and titanium. 

Defects in Forged Products 

Various surface and body defects may be observed in forgings. The kind of defect de- 
pends upon many factors, such as the forging process, the forged metal, the tool de- 
sign, and the temperature at which the process is carried out. Cracking, folds, and 
improper sections are generally the defects observed in forged products. Following is 
a brief description of each defect and its causes. 

Cracking. Cracking is due to the initiation of tensile stresses during the forging 
process. Examples are hot tears, which are peripheral longitudinal cracks experienced 
in upsetting processes at high degrees of deformation, and center cavities, which occur 
in the primary forging of low-ductility steels. Thermal cracks may also initiate in cases 
when nonuniform temperature distribution prevails. 

Folds. In upsetting and heading processes, folding is a common defect that is obvi- 
ously caused by buckling. Folds may also be observed at the edges of parts produced 
by smith forging if the reduction per pass is too small. 

Improper sections. Improper sections include dead-metal zones, piping, and turbu- 
lent (i.e., irregular or violent) metal flow. They are basically related to and caused by 
poor tool design. 


5 Metal Forming 

Forging Die Materials 

During their service life, forging dies are subjected to severe conditions such as high 
temperatures, excessive pressures, and abrasion. A die material must, therefore, pos- 
sess adequate hardness at high temperatures as well as high toughness to be able to 
withstand the severe conditions. Special tool steels (hot-work steels including one or 
more of the following alloying additives: chromium, nickel, molybdenum, and vana- 
dium) are employed as die materials. Die blocks are annealed, machined to make the 
shanks, hardened, and tempered; then, impression cavities are sunk by toolmakers. 

Fundamentals of Closed-Die Forging Design 

The range of forged products with respect to size, shape, and properties is very wide 
indeed. For this reason, it is both advisable and advantageous for the product designer 
to consider forging in the early stages of planning the processes for manufacturing new 
products. The forging design is influenced not only by its function and the properties 
of the material being processed but also by the kind, capabilities, and shortcomings of 
the production equipment available in the manufacturing facilities. Therefore, it is im- 
possible to discuss in detail all considerations arising from the infinite combinations of 
the various factors. Nevertheless, some general guidelines apply in all cases and 
should be strictly adhered to if a sound forging is to be obtained. Following are some 
recommended forging design principles. 

Parting line. The plane of separation between the upper and lower halves of a closed 
die set is called the parting line. The parting line can be straight, whether horizontal or 
inclined, or can be irregular, including more than one plane. The parting line must be 
designated on all forging drawings as it affects the initial cost and wear of the forging 
die, the grain flow that, in turn, affects the mechanical properties of the forging, and, 
finally, the trimming procedure and/or subsequent machining operations on the fin- 
ished part. Following are some considerations for determining the shape and position 
of the parting line: 

1. The parting line should usually pass through the maximum periphery of the forging 
mainly because it is always easier to spread the metal laterally than to force it to fill 
deep, narrow die impressions (see Figure 5.54). 

FIGURE 5.54 

Recommended location 
of the parting line 
(Courtesy of the 
Aluminum Association, 
Inc., Washington, D.C.) 


Less desirable 

5.5 Forging 


FIGURE 5.55 

Flat-sided forging for 
simplifying the die 
construction (Courtesy 
of the Aluminum 
Association, Inc., 
Washington, D.C.) 

Plane surface formed 
/ by flat upper die 


Parting line 

Contour of forging 

formed by impression 

in bottom die 

FIGURE 5.56 

Using the parting line 
to promote the 
alignment of the fibrous 
(Courtesy of the 
Aluminum Association, 
Inc., Washington, D.C.) 

Grain structure is 

ruptured at the 

parting line 

Parting line 


These parting lines result in metal flow patterns 
that cause forging defects 

Most economical as all of 
the impression is in one die 

This parting line should not be — 
above the center of the bottom web 


Parting at the ends 

of ribs results in 

good grain structure 

Recommended - The flow lines are smooth at stressed sections 
with these parting lines 


Metal Forming 

2. It is always advantageous, whenever possible, to try to simplify the die construction 
if the design is to end up with flat-sided forgings (see Figure 5.55). This will 
markedly reduce the die cost because machining is limited to the lower die half. 
Also, the possibility of mismatch between die halves is eliminated. 

3. If an inclined parting line must exist, it is generally recommended to limit the in- 
clination so that it does not exceed 75°. The reason is that inclined flashes may cre- 
ate problems in trimming and subsequent machining. 

4. A parting line should be located so that it promotes alignment of the fibrous 
macrostructure to fulfill the strength requirement of a forging. Because excess 
metal flows out of the die cavity into the gutter as the process proceeds, mislocat- 
ing the parting line will probably result in irregularities, as can be seen in Figure 
5.56, which indicates the fibrous macrostructures resulting from different locations 
of the parting line. 

5. When the forging comprises a web enclosed by ribs, as illustrated in Figure 5.57, 
the parting line should preferably pass through the centerline of the web. It is also 
desirable, with respect to the alignment of fibers, to have the parting line either at 
the top or at the bottom surfaces. However, that desirable location usually creates 
manufacturing problems and is not used unless the direction of the fibrous 
macrostructure is critical. 

6. If an irregular parting line must exist, avoid side thrust of the die, which will cause 
the die halves to shift away from each other sideways, resulting in matching errors. 
Figure 5.58 illustrates the problem of side thrust accompanying irregular parting 
lines, together with two suggested solutions. 

Draft. Draft refers to the taper given to internal and external sides of a closed-die forg- 
ing and is expressed as an angle from the direction of the forging stroke. Draft is required 
on the vast majority of forgings to avoid production difficulties, to aid in achieving desired 
metal flow, and to allow easy removal of the forging from the die cavity. It is obvious that 

FIGURE 5.57 

Location of the parting 
line with respect to a 
web (Courtesy of the 
Aluminum Association, 
Inc.. Washington, D.C.) 

Parting line 

A— 1 

Section AA 

Section BB 

5.5 Forging 


FIGURE 5.58 

The problem of side 
thrust accompanying 
irregular parting lines 
and two suggested 
solutions (Courtesy of 
the Aluminum 
Association, Inc., 
Washington, D.C.) 

Die lock 

Impractical — Side thrust makes it difficult to hold the dies 
in match accurately 

Die lock 

Not recommended — Dies with counterlocks are expensive to 
build and troublesome to maintain 

Die lock 

— Upper die 


- Forging 

Bottom die 

Upper die 



Bottom die 

Upper die 


Bottom die 

Preferred — The best method is to incline the forging with 
respect to the forging plane 

the smaller the draft angle, the more difficult it is to remove the forging out of the die. For 
this reason, draft angles of less than 5° are not permitted if the part is to be produced by 
drop forging (remember that there is no ejector to push the part out). Standard draft angles 
are 7°, 5°, 3°, 1 °, and 0°. A draft angle of 3° is usually used for metal having good forge- 
ability, such as aluminum and magnesium, whereas 5° and 7° angles are used for steels, 
titanium, and the like. It is a recommended practice to use a constant draft all over the pe- 
riphery of the forging. It is also common to apply a smaller draft angle on the outside pe- 
riphery than on the inside one. This is justified in that the outer surface will shrink away 
from the surface of the die cavity as a result of the part's cooling down, thus facilitating 
the removal of the forging. Following are some useful examples and guidelines: 

1. When designing the product, try to make use of the natural draft inherent in some 
shapes, such as curved and conical surfaces (see Figure 5.59). 

2. In some cases, changing the orientation of the die cavity may result in natural draft, 
thus eliminating the need for any draft on the surfaces (see Figure 5.60). 


5 Metal Forming 

FIGURE 5.59 

Examples of the natural 

draft inherent in some 

Parting / 
line v 


FIGURE 5.60 

Examples of changing 
the orientation of the 
impression to provide 
natural draft 



FIGURE 5.61 

Methods for matching 
the contours of two die 
impressions having 
different depths: 
(a) increasing the 
dimension of the upper 
surface; (b) using a 
pad; (c) employing a 
matching draft 
(Courtesy of the 
Aluminum Association, 
Inc., Washington, D.C.) 

Sometimes, the cavity in one of the die halves (for instance, the upper) is shallower 
than that in the other half. This may create problems in matching the contours of 
the two die halves at the parting line. It is, therefore, recommended that one of the 
three methods illustrated in Figure 5.61a, b, or c be used. The first method involves 
keeping the draft the same as in the lower cavity but increasing the dimension of 
the upper surface of the cavity. This results in an increase in weight, and this solu- 


5° (Ref) 


Parting line 


«- Dim (applies to 
lower side only) 


Parting line 

!*- Dim (applies to 
both side) 




«t ; 

Parting line \ 


Dim (ar. 

)plies tc 

note that it applies to the 
original 5° intersection and 
not to the subsequent match 
draft intersection) 


5.5 Forging 


tion is limited to smaller cavities. The second method is based on keeping the draft 
constant in both halves by introducing a "pad" whose height varies between 0.06 
inch (1.5 mm) and 0.5 inch (12.5 mm), depending upon the size of the forging. The 
third method, which is more common, is to provide greater draft on the shallower 
die cavity; this is usually referred to as matching draft. 

Ribs. A rib is a thin part of the forging that is normal to (or slightly inclined to) the 
forging plane. It is obvious that optimized lighter weight of a forging calls for reduc- 
ing the thickness of long ribs. However, note that the narrower and longer the rib is, 
the higher the forging pressure is and the more difficult it is to obtain a sound rib. It is 
actually a common practice to keep the height-to-thickness ratio of a rib below 6, 
preferably at 4. The choice of a value for this ratio depends upon many factors, such 
as the kind of metal being processed and the forging geometry (i.e., the location of the 
rib, the location of the parting line, and the fillet radii). Figure 5.62 indicates the de- 
sirable rib design as well as limitations imposed on possible alternatives. 

Webs. A web is a thin part of the forging that is passing through or parallel to the 
forging plane (see Figure 5.63). Although it is always desirable to keep the thickness 
of a web at the minimum, there are practical limits for this. The minimum thickness 
of webs depends on the kind of material being worked (actually on its forging tem- 
perature range), the size of forging (expressed as the net area of metal at the parting 
line), and on the average width. Table 5.3 indicates recommended web thickness val- 
ues applicable to precision and conventional aluminum forgings. For blocking cavi- 
ties, the values given in Table 5.3 must be increased by 50 percent. Also, for steels 
and other metals having poorer forgeability than aluminum, it is advisable to increase 
the values for web thickness. Thin webs may cause unfilled sections, may warp in 
heat treatment, and may require additional straightening operations; they even cool 
faster than the rest of the forging after the forging process, resulting in shrinkage, 
possible tears, and distortion. 

FIGURE 5.62 


Recommended rib 




design (Courtesy of the 



Aluminum Association, 



Inc., Washington, D.C.) 





thin-ledged ribs 
small fillet radii 


Possible defect 
if a> b 


Metal Forming 

FIGURE 5.63 

The shape of a web in 
forging (Courtesy of the 
Aluminum Association, 
Inc., Washington. D.C.) 

Corner radii. There are two main factors that must be taken into consideration when 
selecting a small value for a corner radius. First, a small corner radius requires a sharp 
fillet in the die steel, which acts as a stress raiser; second, the smaller the corner radius, 
the higher the forging pressure required to fill the die cavity. In addition, some other 
factors affect the choice of the corner radius, such as the distance from the corner to 
the parting line and the forgeability of the metal being worked. The larger the distance 
from the parting line, the larger the corner radius should be. Also, whereas a corner ra- 
dius of 0.0625 inch (1.5 mm) is generally considered adequate for aluminum forging, 
a corner radius of at least 0.125 inch (3 mm) is used for titanium forgings of similar 
shape and size. In addition, the product designer should try to keep the corner radii as 
consistent as possible and avoid blending different values for a given shape in order to 
reduce the die cost (because there will be no need for many tool changes during die 
sinking). Corner radii at the end of high, thin ribs are critical. A rule of thumb states 

5.5 Forging 


TABLE 5.3 

Recommended size of 
minimum web 

Up to Average Width 
in. (m) 

Up to Cross-Sectional Area 
in. 2 (m 2 ) 

Web Thickness 
in. (mm) 

3 (0.075) 

10 (0.00625) 

0.09 (2.25) 

4 (0.1) 

30 (0.01875) 

0.12 (3) 

6 (0.15) 

60 (0.0375) 

0.16 (4) 

8 (0.2) 

100 (0.0625) 

0.19 (4.75) 

11 (0.275) 

200 (0.125) 

0.25 (6.25) 

14 (0.35) 

350 (0.21875) 

0.31 (7.75) 

18 (0.45) 

550 (0.34375) 

0.37 (9.25) 

22 (0.55) 

850 (0.53125) 

0.44 (11) 

26 (0.65) 

1200 (0.75) 

0.50 (12.5) 

34 (0.85) 

2000 (1.25) 

0.62 (15.5) 

41 (1.025) 

3000 (1.875) 

0.75 (18.75) 

47 (1.1175) 

4000 (2.50) 

1.25 (31.25) 

52 (1.3) 

5000 (3.125) 

2.00 (50) 

that it is always desirable to have the rib thickness equal to twice the value of the cor- 
ner radius. A thicker rib may have a flat edge with two corner radii, each equal to the 
recommended value. Figure 5.64 illustrates these recommendations regarding corner 
radii for ribs. 

Fillet radii. It is of supreme importance that the product designer allow generous radii 
for the fillets because abrupt diversion of the direction of metal flow can result in nu- 
merous defects in the product. Figure 5.65 indicates the step-by-step initiation of forg- 
ing defects and shows that small fillets result in separation of the metal from the die 
and initiation of voids. Although these can be filled at a later stage, laps and cold shuts 
will replace these voids. When the shape of the part to be forged is intricate (i.e., in- 
volving thin ribs and long, thin webs), the metal may preferentially flow into the gut- 
ter rather than into the die cavity. This results in a shear in the fibrous macrostructure 
and is referred to as flow-through. This latter defect can be avoided by using larger- 
than-normal fillets. 

FIGURE 5.64 

regarding corner radii 
for ribs (Courtesy of the 
Aluminum Association, 
Inc., Washington, D.C.) 



values of radii 

t = 2/? 



5 Metal Forming 

FIGURE 5.65 

Defects caused by 
employing smaller fillet 
radii (Courtesy of the 
Aluminum Association 
Inc., Washington, D.C.) 

Large fillets Forging stock 


Die motion 

Metal does not 

i sharp corner 

Metal reaches 
bottom of 

cavity before 
ing section 

These cold shuts 
flawed in the 

Punchout holes. Punchout holes are through holes in a thin web that are produced 
during, but not after, the forging process. Punchouts reduce the net projected area of 
the forging, thus reducing the forging load required. If properly located and designed, 
they can be of great assistance in producing forgings with thin webs. In addition to the 
manufacturing advantages of punchouts, they serve functional design purposes, such 
as reducing the mass of a forging and/or providing clearance. Following are some 
guidelines regarding the design of punchouts: 

1. Try to locate a punchout around the central area of a thin web, where the frictional 
force that impedes the metal flow is maximum. 

2. Whenever possible, use a gutter around the interior periphery of a punchout. This 
provides a successful means for the surplus metal to escape. 

3. A single large punchout is generally more advantageous than many smaller ones 
that have the same area. Accordingly, try to reduce the number of punchouts unless 
more are dictated by functional requirements. 

5.6 Cold Forming Processes 201 

4. Although punchouts generally aid in eliminating the problems associated with the 
heat treatment of forgings, it may prove beneficial to take the limitations imposed 
by heat treatment processes into account when designing the contour of a punchout 
(i.e., try to avoid irregular contours with sharp corners). 

Pockets and recesses. Pockets and recesses are used to save material, promote the 
desirable alignment of the fibrous macrostructure, and improve the mechanical proper- 
ties by reducing the thickness, thus achieving a higher degree of deformation. Follow- 
ing are some guidelines: 

1. Recesses should never be perpendicular to the direction of metal flow. 

2. Recesses are formed by punches or plugs in the dies. Therefore, the recess depth is 
restricted to the value of its diameter (or to the value of minimum transverse di- 
mension for noncircular recesses). 

3. Simple contours for the recesses, together with generous fillets, should be tried. 


Cold forming processes are employed mainly to obtain improved mechanical proper- 
ties, better surface finish, and closer tolerances. Several cold forming techniques have 
found wide industrial application. Among these are sizing, swaging, coining, and cold 
heading. Following is a brief description of each of them. 


Sizing (see Figure 5.66a) is a process in which the metal is squeezed in the forming di- 
rection but flows unrestricted in all transverse directions. This process is used primar- 
ily for straightening forged parts, improving the surface quality, and obtaining accurate 
dimensions. A sizing operation can ensure accuracy of dimensions within 0.004 up to 
0.010 inch (0.1 up to 0.25 mm). Meanwhile, the pressure generated on the tools can go 
up to 180,000 pounds per square inch (1300 MN/m 2 ). 


Swaging (see Figure 5.66b) involves imparting the required shape and accurate di- 
mensions to the entire forging (or most of it). Usually, swaging is carried out in a die 
where a flash is formed and subsequently removed by abrasive wheels or a trimming 
operation. Note that the flow of metal in the swaging process is more restricted than in 
sizing. Accordingly, higher forming pressures are experienced and can go up to 
250,000 pounds per square inch (1800 MN/m 2 ). 


Coining (see Figure 5.66c) is a process in which the part subjected to coining is com- 
pletely confined within the die cavity (by the die and the punch). The volume of the 
original forging must be very close to that of the finished part. Any tangible increase 
in that volume may result in excessive pressures and the breakage of tools. Still, com- 


Metal Forming 

FIGURE 5.66 

Cold forming 
processes: (a) sizing; 
(b) swaging; (c) coining 


mon pressures (even when no problems are encountered) are in the order of 320,000 
pounds per square inch (2200 MN/nV). For this reason, coining processes (also sizing 
and swaging) are carried out on special presses called knuckle presses. The main 
mechanism of a knuckle press is shown in Figure 5.67. It is characterized by the abil- 
ity to deliver a large force with a small stroke of the ram. 

Cold Heading 

Cold heading is used to manufacture bolts, rivets, nuts, nails, and similar parts with 
heads and collars. A group of typical products are illustrated in Figure 5.68. The main 
production equipment involves a multistage automatic cold header that operates on the 

5.6 Cold Forming Processes 


FIGURE 5.67 

The working principles 
of a knuckle press for 
cold forming processes 


same principle as a horizontal forging machine. Full automation and high productivity 
are among the advantages of this process. Products having accurate dimensions can be 
produced at a rate of 30 to 300 pieces per minute. Starting from coiled wires or rods 
made of plain-carbon steel and nonferrous metals with diameters ranging from 0.025 
to 1.6 inches (0.6 to 40 mm), blanks are processed at different stations. Feeding, trans- 
fer, and ejection of the products are also automated. Figure 5.69 illustrates the differ- 
ent stages involved in a simple cold heading operation. 

Lubrication in Cold Forming 

Lubricants employed in cold forming are similar to those used in heavy wire-drawing 
processes. Phosphating followed by soap dipping is successful with steels, whereas 
only soap is considered adequate for nonferrous metals. 

FIGURE 5.68 

Some products 
manufactured using an 
automatic cold header 


Metal Forming 

FIGURE 5.69 

Different stages of a 
simple cold heading 


3- € 


Review Questions 


1. Why have metal forming processes gained 
widespread industrial application since World 
War II? 

2. What are the two main groups of metal form- 
ing processes? 

3. List the different factors affecting the defor- 
mation process. Tell how each influences de- 

4. Why are cold forming processes always ac- 
companied by work-hardening, whereas hot 
forming processes are not? 

5. What is meant by the fibrous macro structure? 

6. Are the mechanical properties of a rolled sheet 
isotropic? Why? 

7. What is meant by the state of stress? List the 
three general types. 

8. List some advantages of hot forming. What are 
some disadvantages? 

9. List some advantages of cold forming. What 
are some disadvantages? 

10. What may happen when a large section of 
steel is heated at a rapid rate? Why? 

11. What should be avoided when heating large 
steel sections prior to hot forming? 

12. Where does friction occur in metal forming? 

13. What are the harmful effects of friction on the 
forming process? 

14. Is friction always harmful in all metal forming 

15. Can lead be used as a lubricant when forming 
copper? Why? 

16. When forming lead at room temperature, do 
you consider it cold forming? Why? 

17. Why are lubricants used in metal forming 
processes? List some useful effects. 

18. List some lubricants used in cold forming 

19. List some lubricants used in hot forming 

20. Which do you recommend for further process- 
ing by machining, a cold-worked part or a hot- 
worked part? 

21. Is hot rolling the most widely used metal 
forming process? Why? 

22. List some of the useful effects of hot rolling. 

23. Define rolling. 

24. What is the angle of contact? 

25. For heavier sections, would you recommend 
larger angles of contact in rolling? Why? 

26. What is the state of stress in rolling? 

27. List the different types of rolling mills. 

28. What are the different parts of a roll? What is 
the function of each? 

Explain why Sendzimir mills are used. 

What are universal mills used for? 

31. List three groups included in the range of 
rolled products. 


Chapter 5 Review Questions 


32. Explain, using sketches, how seamless tubes 
are manufactured. 

33. What is alligatoring? What causes it? 

34. Define wire drawing. 

35. Which mechanical property should the metal 
possess if it is to be used in a drawing 
process? Why? 

36. What is the state of stress in drawing? 

37. List some advantages of the drawing process. 

38. How is a metal prepared for a drawing process? 

39. What are the different zones in a drawing die? 

40. Mention the range of the apex angles (of con- 
ical shapes) used in drawing dies. 

41. What material do you recommend to be used 
in making drawing dies? 

42. Describe a draw bench. 

43. What kinds of lubricants are used in drawing 

44. What is the drawing ratio? 

45. Give an expression indicating the reduction 
achieved in a wire-drawing process. 

46. Why do internal bursts occur in wire-drawing 

47. What are arrowhead fractures and why do they 

48. What is the state of stress in tube drawing? 

49. Using sketches, illustrate the different tech- 
niques used in tube drawing. 

50. Define extrusion. 

51. Why can extrusion be used with metals having 
relatively poor plasticity? 

52. List some advantages of the extrusion process. 

53. What are the shortcomings and limitations of 
the extrusion process? 

54. Using sketches, differentiate between the di- 
rect and indirect extrusion techniques. 

55. Although indirect extrusion almost eliminates 
friction, it is not commonly used in industry. 

56. List the advantages of hydrostatic extrusion. 

57. Compare extrusion with rolling with respect to 
efficiency of material utilization. 

58. When is conventional direct extrusion recom- 
mended as a production process? 

59. Describe impact extrusion. 

60. Why is the leading end of an extruded section 
always sheared off? 

61. What are dead-metal zones? 

62. If hardness measurements are taken across the 
section (say, circular) of an extruded part, 
what locations will have higher hardness val- 
ues? Can you plot hardness versus distance 
from the center? 

63. What lubricants can be used in cold extrusion? 

64. What material do you recommend as a lubri- 
cant when hot extruding stainless steel? 

65. What defect may occur when extruding mag- 
nesium at low extrusion ratios? 

66. What is piping and why does it occur? 

67. In extrusion dies, what is meant by the circle 

68. List some considerations that must be taken 
into account when designing a section for ex- 

69. Why should a designer try to avoid sharp cor- 
ners at the root of a die tongue? Explain using 
neat sketches. 

70. As a product designer, you are given a very in- 
tricate section for production by extrusion. Is 
there any way around this problem without 
being forced to use a die with a very intricate 
construction? How? 

71. List some considerations for the design of im- 
pact extrusions. 

72. How can you avoid shear failure at the bottom 
of the wall of an impact extrusion? 

73. Does forging involve just imparting a certain 
shape to a billet? 


Metal Forming 




74. Is it just a matter of economy to produce a 
crankshaft by forging rather than by machin- 
ing from a solid stock? Why? 
Can a metal such as aluminum be forged at 
any temperature? Why? 
List the main types of forging processes. 
Which process is suited for the production of 
small batches of large parts? 

Give examples of parts produced by each type 

of forging process. Support your answer with 


What is the modern version of blacksmi thing? 

What are the different operations involved in 

that process? 

80. When do you recommend using a power-actu- 
ated hammer as a forging machine? Mention 
the type of forging process. 

81. For which type of forging is a drop hammer 

82. For which type of forging is a crank press 

Using sketches, illustrate the different stages 
in manufacturing a ring by forging. 

List the advantages that forging has over cast- 
ing when producing large numbers of small 
parts having relatively complex shapes. 

85. In the comparison of Question 84, what are 
the shortcomings of forging? Why don't they 
affect your decision in that particular case? 

86. List some of the specified acceptance tests to 
be performed on forgings. 

87. What is a board hammer used for? 

88. Is it true that a closed type of forging die can 
have only one impression? Explain why. 



89. What does hot pressing mean? 

90. What is the advantage of HERF? 

91. What are the advantages of warm forging? 

92. What is meant by a mushy state? 

93. Define forgeability. How can it be quantita- 
tively assessed? 

94. What is the most forgeable metal? 

95. What is the main role of lubricants in hot forg- 

96. As a product designer, how can you manipu- 
late the alignment of the fibrous macrostruc- 

97. List some guidelines regarding the location of 
the parting line between the upper and lower 
halves of a die set. 

98. What is meant by the term draft in forging? 

99. A die was designed to forge an aluminum part. 
Can the same design be used to forge a similar 
part made of titanium? Why? 

100. Explain the meaning of matching draft, using 

101. Differentiate between a web and a rib in a 

102. What is the difference between a corner radius 
and a fillet radius? Use sketches. 

103. What are punchout holes in a forging? 

104. List some advantages of including punchout 
holes in a forging design. 

105. Why are recesses sometimes included in a 
forging design? 

106. List the different cold forming processes and 
use sketches to illustrate how they differ. 

Chapter 5 Design Example 




1. In hot rolling, determine the load on each roll of 
a two-high rolling mill, given the following: 

Diameter of the 


Stock width: 

Initial thickness: 

Final thickness: 

Flow stress of 

rolled material: 

20 inches (500 mm) 
48 inches (1020 mm) 
0.08 inch (2 mm) 
0.04 inch (1 mm) 

14,200 lb/in. 2 (100 MN/m 2 ) 

In hot rolling low-carbon-steel plate 48 inches 
( 1 200 mm) in width, given the roll diameter as 20 
inches (500 mm), initial thickness as 1.5 inches 
(37.5 mm), final thickness as 0.4 inch (10 mm), 
and the flow stress of steel as 28,400 lb/in. 2 (200 
MN/m 2 ), calculate the number of rolling passes if 
the maximum load on the roll in each pass is not 
to exceed 225,000 pounds force (1.0 MN). 


3. Write a computer program to solve Problem 2, 
assuming that all the data are variables to be 
given for each design. 

4. Calculate the maximum achievable reduction in 
a single drawing of a lead wire. 

5. Estimate the largest possible extrusion ratio of 
2.0-inch (50-mm) aluminum bar having mean 
flow stress of 21,900 lb/in. 2 (150 MN/m 2 ) if the 
press available has a capacity of only 45,000 
pounds force (200 kN). 

6. Plot a curve indicating the efficiency of a drop 
hammer versus the ratio between the weights of 
the anvil and the moving parts if the value of K 
that represents the elasticity of the billet is taken 
as 0. 1 . What ratio do you suggest? Why should it 
not be justified to take large ratios? 

Design Example 


Design a simple wrench that measures 1/2 inch (12.5 mm) across bolt-head flats and 
is used for loosening nuts and bolts. The torque required to loosen (or tighten) a bolt 
(or a nut) is 1 lb ft (6.8 Nm). The production volume is 25,000 pieces per year. Forg- 
ing is recommended as a manufacturing process. 


Because the wrench is going to be short, it cannot be held by the full hand but prob- 
ably by only three fingers. The force that can be exerted is to be taken, therefore, as 
4 pounds. The arm of the lever is equal to (1 x 12)/4, or 3 inches (75mm). Add on 
allowance for the holding fingers. The shape of the wrench will be as shown in Fig- 
ure 5.70. 

Now, let us select the materials. A suitable material would be AISI 1045 CD steel 
to facilitate machining (sawing) of the stock material. Closed-die forging of the billets 


5 Metal Forming 

FIGURE 5.70 

A wrench manufactured 
by forging 

Section AA 

1875 inch 

0.6 inch 

R= 0.95 inch 

0.375 inch 


■Parting line 


0.25 inch 

is recommended, as well as employing drop-forging hammers. To facilitate withdrawal 
of the part, the cross section of the handle should be elliptical (see Figure 5.70). The 
parting line should coincide with the major axis of the ellipse. 
Let us check the stress due to bending: 

/ = -n a 3 b = - (7t)(0.375) 3 (0.1875) = 7.7 x 10" 
4 4 

where: a is half the major axis 
b is half the minor axis 

3 in. 4 

stress = 



5 x 12 x 0.375 

= 2922 lb/in. 

/ / 7.7 x 10" 3 

It is less than the allowable stress for 1045 CD steel, which is 

^°°° = 30,000 1M„.' 

In order to check the bearing stress, let us assume a shift of 0.25 inch between the 
forces acting on the faces of the nut to form a couple (this assumption can be verified 
if we draw the nut and the wrench to scale): 

each force = 


= 240 pounds 

Further assume that the bearing area is 0.375 by 0.25 inch. The bearing stress is, 


= 2560 lb/in/ 

0.375 x 0.25 
It is less than the allowable stress of the 1045 CD steel. 

Chapter 5 Design Projects 209 

The forged wrench finally has to be trimmed and then machined on the surfaces 
indicated in Figure 5.70. An allowance of 1/64 inch should be provided between the 
wrench open-head and the nut. Now, our design is complete and ready to be released 
to the workshop. 

ssign Projects 



1. A clock frame 3 by 5 inches (75 by 125 mm) is manufactured by machining an 
aluminum-alloy stock. Make a design and a preliminary feasibility study so that it 
can be produced by extrusion. Assume the production volume is 20,000 pieces per 

2. A motor frame that has a 6-inch (150-mm) internal diameter and that is 10 inches 
(250 mm) long is currently produced by casting. That process yields a high per- 
centage of rejects, and the production cost is relatively high. Knowing that the pro- 
duction volume is 20,000 pieces per year, redesign the part so that it will be lighter 
and can be easily produced by an appropriate metal forming operation that has a 
high efficiency of material utilization. 

3. A pulley transmits a torque of 600 lb ft (816 Nm) to a shaft that is IV4 inches (31 
mm) in diameter. It is to be driven by a flat belt that is 2 inches (50 mm) in width. 
Provide a detailed design for the pulley if the production volume is 10,000 pieces 
per year and the pulley is manufactured by forging. 

4. A connecting lever is to be manufactured by forging. The estimated production vol- 
ume is 50,000 pieces per year. The lever has two short bosses, each at one of its 
ends, and each has a vertical hole 3/4 inch (19 mm) in diameter. The horizontal dis- 
tance between the centers of the two holes is 12 inches (300 mm), and the vertical 
difference in levels is 3 inches (75 mm). The lever during its functioning is sub- 
jected to a bending moment of 200 lb ft (272 Nm). Make a detailed design for this 

5. If the lever in Problem 4 is to be used in a space vehicle, would you use the same 
material? What are the necessary design changes? Make a design appropriate for 
this new situation. 

6. Design a gear blank that transmits a torque of 200 lb ft (272 Nm) to a shaft that is 
3/4 inch (19 mm) in diameter. The pitch diameter of the gear is 8 inches (200 mm), 
and 40 teeth are to be cut in that blank by machining. Assume the production vol- 
ume is 10,000 pieces per year. 

7. A straight-toothed spur-gear wheel transmits a torque of 1200 lb ft (1632 Nm) to 
a steel shaft (AISI 1045 CD steel) that is 2 inches (50 mm) in diameter. The pitch 

210 5 Metal Forming 

diameter of the gear is 16 inches (400 mm), its width is 4 inches (100 mm), and 
the base diameter is 15 inches (375 mm). Make a complete design for this gear's 
blank (i.e., before teeth are cut) when it is to be manufactured by forging. Assume 
the production volume is 10,000 pieces per year. 

A shaft has a minimum diameter of 1 inch (25 mm) at both its ends, where it is 
to be mounted in two ball bearings. The total length of the shaft is 12 inches 
(300 mm). The shaft is to have a gear at its middle, with 40 teeth and a pitch- 
circle diameter of 1.9 inches (47.5 mm). The width of the gear is 2 inches (50 mm). 
Make a design for this assembly if the production volume is 50,000 per year. 

Chapter 6 

eet Metal 


The processes of sheet metal working have recently gained widespread indus- 
trial application. Their main advantages are their high productivity and the close 
tolerances and excellent surface finish of the products (which usually require 
no further machining). The range of products manufactured by these processes 
is vast, but, in general, all of these products have thin walls (relative to their 
surface area) and relatively intricate shapes. Sheets made from a variety of 
metals (e.g., low-carbon steel, high-ductility alloy steel, copper and some of its 
alloys, and aluminum and some of its alloys) can be successfully worked into 
useful products. Therefore, these processes are continually becoming more at- 
tractive to the automotive, aerospace, electrical, and consumer goods indus- 
tries. Products that had in the past always been manufactured by processes 
like casting and forging have been redesigned so that they can be produced by 
sheet metal working. Components like pulleys, connecting rods for sewing ma- 
chines, and even large gears are now within the range of sheet metal products. 

Sheet metals are usually worked while in their cold state. However, when 
processing thick sheets, which are at least 0.25 inch (6 mm) and are referred 
to as plates, thermal cutting is employed to obtain the required blank shape, 
and the blank is then hot-worked in a hydraulic or friction screw press. Thus, 
fabrication of boilers, tanks, ship hulls, and the like would certainly require hot 
working of thick plates. 

By far, the most commonly used operations in sheet metal working are 
those performed in a press. For this reason, they are usually referred to as 



6 Sheet Metal Working 

press working, or simply stamping, operations. Other techniques involve high- 
energy-rate forming (HERF), like using explosives or impulsive discharges of 
electrical energy to form the blank, and spinning of the sheet metal on a form 
mandrel. This chapter will describe each of the various operations employed in 
sheet metal working. 


All press working operations of sheet metals can be divided into two main groups: cut- 
ting operations and shape-forming operations. Cutting operations involve separating a 
part of the blank, whereas forming operations involve nondestructive plastic deforma- 
tion, which causes relative motion of parts of the blank with respect to each other. Cut- 
ting operations include shearing, cutoff, parting, blanking, punching, and notching. 
Shape-forming operations include various bending operations, deep drawing, emboss- 
ing, and stretch-forming. 

Cutting Operations 

The mechanics of separating the metal are the same in all sheet metal cutting operations. 
Therefore, the operations are identified according to the shape of the curve along which 
cutting takes place. When the sheet metal is cut along a straight line, the operation is 
called shearing and is usually performed using inclined blades or guillotine shears in 
order to reduce the force required (see Figure 6.1). Cutting takes place gradually, not all 
at once, over the width of the sheet metal because the upper blade is inclined. The angle 
of inclination of the upper blade usually falls between 4° and 8° and must not exceed 15° 
so that the sheet metal is not pushed out by the horizontal component of the reaction. 

When cutting takes place along an open curve (or on an open corrugated line), the 
operation is referred to as cutoff, provided that the blanks match each other or can be 
fully nested, as shown in Figure 6.2. The cutoff operation results in almost no waste of 
stock and is, therefore, considered to be very efficient with respect to material utiliza- 
tion. This operation is usually performed in a die that is mounted on a crank press. If the 
blanks do not match each other, it is necessary for cutting to take place along two open 
curves (or lines), as shown in Figure 6.3. In this case, the operation is called parting. It 


Shearing operation with 
inclined blades 


>ZL blade 

6.1 Press Working Operations 



Examples of cutoff 




Cutting takes 

place along these 

two lines, each 



blank shape 


An example of a parting 

Cutting takes place 
along these lines 


is clear from the figure that a parting operation results in some waste of stock and is, 
therefore, less efficient than shearing and cutoff operations. 

In blanking operations, cutting occurs along a closed contour and results in a 
relatively high percentage of waste in stock metal, a fact that makes blanking oper- 
ations less efficient than other cutting operations. Nevertheless, this process is used 
for mass production of blanks that cannot be manufactured by any of the preceding 
operations. An efficient layout of blanks on the strip of sheet metal can result in an 
appreciable saving of material. An example of a good layout is shown in Figure 
6.4a, where circular blanks are staggered. The in-line arrangement shown in Figure 
6.4b is less efficient in terms of material utilization. Because a blanking operation 
is performed in a die, there is a limit to the minimum distance between two adja- 
cent blanks. It is always advantageous to keep this minimum distance larger than 70 
percent of the thickness of the sheet metal. In blanking, the part separated from the 
sheet metal is the product, and it is usually further processed. But if the remaining 


Two methods for laying 
out circular blanks for 
blanking operations: 

(a) staggered layout; 

(b) in-line arrangement 











6 Sheet Metal Working 


Different patterns of 
holes produced by 
perforating operations 

o o 
o o 
o o 


o o o o o 

o o o o o 

o o o o o 

o o o o o 

o o o o o 

o o o o o 


Progressive working 

(2) \ (3) 

pilot holes 

imw m 




Cut off along this line 
to separate the product 

part of the sheet is required as a product, the operation is then termed punching. 
Sometimes, it is required to simultaneously punch a pattern of small holes as an or- 
nament, for light distribution, or for ventilation; the operation is then referred to as 
perforating. Figure 6.5 illustrates some patterns of perforated holes. 

A notching operation is actually a special case of punching, where the removed 
part is adjacent to the edge of the strip. It is clear that any required shape can be ob- 
tained by carrying out several notching operations. For this reason, notching is usually 
employed in progressive dies. A similar operation, called seminotching, in which the 
separated part is not attached to the side of the strip, is also used in progressive work- 
ing of sheet metals. In Figure 6.6, we can see both of these operations and how they 
can be employed progressively to produce a blank with an intricate shape. 

Mechanics of sheet metal cutting. Let us now look further at the process of cutting 
sheet metal. For simplicity, consider the simple case where a circular punch, together 
with a matching die, are employed to punch a hole. Figure 6.7 shows the punch, die, 
and sheet metal during a punching operation. When a load is applied through the 
punch, the upper surface of the metal is elastically bent over the edge of the punch, 
while the lower surface is bent over the edge of the die. With further increase in the 
punch load, the elastic curvature becomes permanent or plastic and is referred to as the 
rollover. Next, the punch sinks into the upper surface of the sheet, while the lower sur- 
face sinks into the die hole. This stage involves mainly plastic flow of metal by shear- 
ing as there are two forces equal in magnitude and opposite in direction, subjecting the 
cylindrical surface within the metal to intense shear stress. The result will be a cylin- 
drical smooth surface in contact with the cylindrical surface of the punch as it sinks 
into the sheet metal. Also, a similar surface forms the border of the part of the metal 
sinking into the die hole. Each of these smooth surfaces is called a burnish. The extent 
of a burnish depends upon the metal of the sheet as well as on the design features of 

6.1 Press Working Operations 



Stages of a blanking 



Final hole 

Final blank 


Fracture surface 




Blanking operations 

where the punch-die 

clearance is: 

(a) excessive; (b) too 


the die. The burnish ranges approximately between 40 and 60 percent of the stock 
thickness, the higher values being for soft ductile materials like lead and aluminum. At 
this stage, two cracks initiate simultaneously in the sheet metal, one at the edge of the 
punch and the other at the edge of the die. These two cracks propagate and finally meet 
each other to allow separation of the blank from the sheet metal. This zone has a rough 
surface and is called the fracture surface (break area). Finally, when the newly formed 
blank is about to be completely separated from the stock, a burr is formed all around 
its upper edge. Thus, the profile of the edge of a blank involves four zones: a rollover, 
a burnish, a fracture surface, and a burr. In fact, the profile of the edge of the gener- 
ated hole consists of the same four zones, but in reverse order. 

We are now in a position to discuss the effects of some process parameters, such 
as the punch-die clearance. Figure 6.8a illustrates the case where the punch-die clear- 
ance is excessive and is almost equal to the thickness of the sheet. Initially, the metal 
is bent onto the round edges of the punch and the die, and it then forms a short circu- 
lar wall connecting the flat bottom and the bulk of the sheet. With further increase in 
the applied load, the wall elongates under the tensile stress, and tearing eventually oc- 
curs. As can be seen in Figure 6.8a, the blank resulting in this case has a bent, torn 
edge all around and, therefore, has no value. On the other hand, if the punch-die clear- 
ance is too tight, as shown in Figure 6.8b, the two cracks that initiate toward the end 
of the operation do not meet, and another shearing must take place so that the blank 
can be separated. This operation is referred to as the secondary shear. As can be seen, 
the obtained blank has an extremely rough side. In addition, the elastically recovering 

Location where 

secondary shear 


ES f- F S3 

Edge of 




6 Sheet Metal Working 


Elastic recovery of the 
metal around the hole 
gripping the punch 

FIGURE 6.10 

Elastic recovery of the 
blank necessitating die 

sheet stock tends to grip the punch, as shown in Figure 6.9, thus increasing the force 
required to withdraw the punch from the hole, which is usually called the stripping 
force. This results in excessive punch wear and shorter tool life. On the other hand, the 
blank undergoes elastic recovery, and it is, therefore, necessary to provide relief by en- 
larging the lower part of the die hole, as shown in Figure 6.10. 

Between these two extremes for the punch-die clearance, there exists an optimum 
value that reduces or minimizes the stripping force and the tool wear and also gives a 
blank with a larger burnish and smaller fracture surface. This recommended value for 
the punch-die clearance is usually taken as about 10 to 15 percent of the thickness of 
the sheet metal, depending upon the kind of metal being punched. 

Forces required. Based on the preceding discussion, the force required for cutting 
sheet metal is equal to the area subjected to shear stress (the product of the perimeter 
of the blank multiplied by the thickness of the sheet metal) multiplied by the ultimate 
shear strength of the metal being cut. The blanking force can be expressed by the fol- 
lowing equation: 

F=KxQxtx x ultimate (6.1) 

where: Q is the perimeter 
/ is the thickness 
^ultimate is trie ultimate shear strength 

Note that K is an experimentally determined factor to account for the deviation of the 
stress state from that of pure shear and is taken as about 1.3. The ultimate shear stress 
can either be obtained from handbooks or be taken as approximately 0.8 of the ulti- 
mate tensile strength of the same metal. 

We can now see that one of the tasks of a manufacturing engineer is to calculate the 
required force for blanking (or punching) and to make sure that it is below the capacity 
of the available press. This is particularly important in industries that involve blanking 
relatively thick plates. There is, however, a solution to the problem when the required 
force is higher than the capacity of the available press. It is usually achieved by bevel- 
ing (or shearing) the punch face in punching operations and the upper surface of the die 
steel in blanking operations. Shearing the punch results in a perfect hole but a distorted 
blank, whereas shearing the die yields a perfect blank but a distorted hole. Nevertheless, 
in both cases, cutting takes place gradually, not all at once, along the contour of the hole 
(or the blank), with the final outcome being a reduction in the required blanking force. 
The shear angle is usually taken proportional to the thickness of the sheet metal and 
ranges between 2° and 8°. Double-sheared punches are quite common and are employed 

6.1 Press Working Operations 


to avoid the possibility of horizontal displacement of sheet metals during punching. Fig- 
ure 6. 1 1 illustrates the basic concept of punch and die shearing. It also provides a sketch 
of a double-sheared punch. 

Another important aspect of the punching (or blanking) operation is the stripping 
force (i.e., the force required to pull the punch out of the hole). It is usually taken as 
10 percent of the cutting force, although it depends upon some process parameters, 
such as the elasticity and plasticity of the sheet metals, the punch-die clearance, and 
the kind of lubricant used. 

Bar cropping. Bar cropping is similar to sheet metal cutting. Although bars, not 
sheets, are cut, the mechanics of the process are similar to those of sheet metal cutting, 
and separation of the cropped part is due to plastic flow caused by intense shear stress. 
The process is used for mass production of billets for hot forging and cold forming 
processes. Nevertheless, the distortion and work-hardening at the sheared cross section 
limit the application of bar cropping when the billets are to be cold formed. Therefore, 
a modified version of the cropping operation has to be used. It involves completely 
confining the cropped billet and applying an axial stress of approximately 20 percent 
of the tensile strength of the bar material. This bar-cropping technique, which is shown 
in Figure 6.12, yields a very smooth cropped surface and distortion-free billets. 

Fine blanking. As we saw previously, the profile of the edge of a blank is not smooth 
but consists of four zones: the rollover, the burnish, the fracture surface (break area), 
and the burr. Sometimes, however, the blank must have a straight, smooth side for 
some functional reasons. In this case, an operation called fine blanking is employed, as 

FIGURE 6.11 

Shearing of the punch 

and the die: 

(a) sheared punch 

resulting in distorted 

blanks; (b) sheared die 

resulting in distorted 


FIGURE 6.12 

Bar cropping with 
workpiece totally 




6 Sheet Metal Working 

FIGURE 6.13 

Fine-blanking operation 

Upper punch 

Pressure pad 

Sheet metal 

Die steel 

Figure 6.13 shows. This operation necessitates the use of a triple-action press and a 
special die with a very small punch-die clearance. As can be seen in the figure, the 
metal is squeezed and restrained from moving in the lateral directions in order to con- 
trol the shear flow along a straight vertical direction. A variety of shapes can be pro- 
duced by this method. They can have any irregular outer contour and a number of 
holes as well. The fine-blanking operation has found widespread application in preci- 
sion industries. 

Miscellaneous cutting operations. The primary operation that is used for preparing 
strips for blanking is needed because the available sheets vary in width between 32 and 
80 inches (800 to 2000 mm), a range that is usually not suitable because of the di- 
mensions of the die and the press. Therefore, coils having a suitable width have to be 
obtained first. The operation performed is called slitting, and it employs two circular 
cutters for each straight cut. Sometimes, slitting is carried out in a rolling plant, and 
coils are then shipped ready for blanking. 

A secondary operation that is sometimes carried out on blanks (or holes) to elim- 
inate rough sides and/or to adjust dimensions is the shaving operation. The excess 
metal in this case is removed in the form of chips. As can be seen in Figure 6.14, the 
punch-die clearance is very small. For this reason, the die must be rigid, and matching 
of its two halves must be carefully checked. 

Sometimes, punching operations are mistakenly called piercing. In fact, the me- 
chanics of sheet metal cutting in the two operations are completely different. We can 
see in Figure 6.15 that piercing involves a tearing action. We can also see the pointed 

FIGURE 6.14 

The shaving operation 


6.1 Press Working Operations 


FIGURE 6.15 

The piercing operation 


ie steel 

shape of the punch. Neither blanks nor metal waste result from the piercing operation. 
Instead, a short sleeve is generated around the hole, which sometimes has functional 
application in toy construction and the like. 

Cutting-die construction. The construction of cutting dies may take various forms. 
The simplest one is the drop-through die, which is shown in Figure 6.16. In addition 
to the punch and die steels, the die includes the upper and lower shoes, the guideposts, 
and some other auxiliary components for guiding and holding the metal strip. The 
stripper plate touches the strip first and holds it firmly during the blanking operation; 
it then continues to press it until the punch is totally withdrawn from the hole made in 
the strip. The generated blanks fall through the die hole, which has a relief for this rea- 
son, and are collected in a container located below the bed of the press. 

Consequently, this die construction is applicable only if the bed of the press has a 
hole. On the other hand, if the diameter of the required blanks is too large, the use of 
a drop-through die may result in a defect called dishing. As shown in Figure 6.17. this 
defect involves slackening of the middle of the blank in such a manner that it becomes 
curved and not flat. The answer to this problem lies in employing a return-type die. 

FIGURE 6.16 

Die construction for 
simple drop-through 
blanking die 



Die steel 

die shoe 


6 Sheet Metal Working 

FIGURE 6.17 

A vertical section 
through a blank with 
the dishing defect 

Figure 6.18 shows that in this type of die construction, the blank is supported through- 
out the operation by a spring-actuated block that finally pushes the blank upward 
above the surface of the strip, where it is automatically collected. A more complicated 
die construction, like that shown in Figure 6.19, can be used to perform two operations 
simultaneously. This is usually referred to as a compound die. As can be seen in Fig- 
ure 6.19, the hollow blanking punch is also a hole-punching die. This allows blanking 
and punching to be carried out simultaneously. The product, which is a washer, and the 
central scrap are removed by return blocks. 

Bending Operations 

Bending is the simplest operation of sheet metal working. It can, therefore, be carried 
out by employing simple hand tools. As opposed to cutting operations, there is always 
a clear displacement between the forces acting during a bending operation. The gener- 
ated bending moment forces a part of the sheet to be bent with respect to the rest of it 
through local plastic deformation. Therefore, all straight unbent surfaces are not sub- 
jected to bending stresses and do not undergo any deformation. Figure 6.20 illustrates 
the most commonly used types of bending dies: the V-type, the wiping, and the chan- 
nel (U-type) dies. We can see that the displacement between forces is maximum in the 

FIGURE 6.18 

A return-type die 



FIGURE 6.19 

A compound die for 
producing a washer 

Punch steel 



Die steel 

6.1 Press Working Operations 


FIGURE 6.20 

The three common 
types of bending dies: 
(a) V-type die; (b) wiping 
die; (c) channel (U-type) 




case of the V-type die, and, therefore, lower forces are required to bend sheet metal 
when using this kind of die. 

Mechanics of bending. The bending of sheet metal resembles the case of a beam with 
a very high width-to-height ratio. When the load is applied, the bend zone undergoes 
elastic deformation; then plastic deformation occurs with a further increase in the ap- 
plied load. During the elastic deformation phase, the external fibers in the bend zone 
are subjected to tension, whereas the internal fibers are subjected to compression. The 
distribution of stresses is shown in Figure 6.21a. Note that there is a neutral plane that 
is free of stresses at the middle of the thickness of the sheet. The length of the neutral 
axis remains constant and does not undergo either elongation or contraction. Next, 
when the plastic phase starts, the neutral plane approaches the inner surface of the 
bend, as can be seen in Figure 6.21b. The location of the neutral plane is dependent 
upon many factors, such as the thickness of the sheet metal, the radius, and the degree 
of bend. Nevertheless, the distance between the neutral plane and the inner surface of 
the bend is taken as equal to 40 percent of the thickness of the sheet metal as a first ap- 
proximation for blank-development calculations. 

Let us now consider a very important phenomenon — namely, springback, which 
is an elastic recovery of the sheet metal after the removal of the bending load. As Fig- 
ure 6.22 indicates, for bending by an angle of 90°, the springback amounts to a few de- 
grees. Consequently, the obtained angle of bend is larger than the required one. Even 

FIGURE 6.21 

Distribution of stress 
across the sheet 
thickness: (a) in the 
early stage of bending; 
(b) toward the end of a 
bending operation 










6 Sheet Metal Working 

FIGURE 6.22 

The springback 

Position of the sheet metal 
after partial elastic recovery 


toward the end of the bending operation, the zone around the neutral plane is subjected 
to elastic stresses and, therefore, undergoes elastic deformation (see Figure 6.21b). As 
a result, the elastic core tries to return to its initial flat position as soon as the load is 
removed. When doing so, it is impeded by the plastically deformed zones. The final 
outcome is, therefore, an elastic recovery of just a few degrees. Consequently, the way 
to eliminate springback involves forcing this elastic core to undergo plastic deforma- 
tion. This can be achieved through either of the techniques shown in Figure 6.23a and 
b. In the first case, the punch is made so that a projection squeezes the metal locally; 
in the second case, high tensile stress is superimposed upon bending. A third solution 
is overbending, as shown in Figure 6.23c. In this case, the amount of overbending 
should be equal to the springback so that the exact required angle is obtained after the 
elastic recovery. 

Blank development. We have previously referred to the fact that the neutral plane 
does not undergo any deformation during the bending operation and that its length, 
therefore, remains unchanged. Accordingly, the length of the blank before bending can 
be obtained by determining the length of the neutral plane within the final product. The 
lengths of the straight sections remain unchanged and are added together. The follow- 
ing equation can be applied to any general bending product, such as the one shown in 
Figure 6.24: 

L = total length of blank before bending 

3 4 180 ' 180 2 180 3 


FIGURE 6.23 

Methods used to 
eliminate springback: 

(a) bottoming; 

(b) overbending; 

(c) stretch-forming 

Bending moment 

The final 


6.1 Press Working Operations 


FIGURE 6.24 

A bending product 
divided into straight 
and circular sections 
for blank development 

where: R is equal to r + 0.4/ 

r is the inner radius of a bend 

t is the thickness of the sheet metal 

R is the radius of the neutral axis 

Classification of bending operations. Various operations can be classified as bending, 
although each one has its own industrial name. They include, for example, conven- 
tional bending, flanging, hemming, wiring, and corrugating. The flanging operation is 
quite similar to conventional bending, except that the ratio of the lengths of the bent 
part to that of the sheet metal is small. Flanging is usually employed to avoid a sharp 
edge, thus eliminating the possibility of injury. It is also used to add stiffness to the 
edges of sheet metal and for assembly purposes. 

Among the bending operations, hemming used to be a very important one, before 
the recent developments in welding and can-forming technologies. A hem is a flange 
that is bent by 180°; it is used now to get rid of a sharp edge and to add stiffness to 
sheet metal. A few decades ago, hems were widely employed for seaming sheet met- 
als. Figure 6.25 shows four different kinds of hems. A similar operation is wiring, 
which is shown in Figure 6.26. True wiring involves bending the edge of the sheet 
metal around a wire. Sometimes, the operation is performed without a wire, and it is 
then referred to as false wiring. 

Corrugating is another operation that involves bending sheet metal. Different 
shapes, like those shown in Figure 6.27, are obtained by this operation. These shapes 
possess better rigidity and can resist bending moments normal to the corrugated cross 

FIGURE 6.25 

Different kinds of hems 


Flat hem 

Open hem 

Teardrop hem 

Seaming using 
two hems 


6 Sheet Metal Working 

FIGURE 6.26 

Wiring operation 

True wiring 

False wiring 

FIGURE 6.27 

Different shapes of 
corrugated sheet metal 


section mainly because of the increase in the moment of inertia of the section due to 
corrugation and because of the work-hardened zones resulting from bending. 

Miscellaneous bending operations. Conventional bending operations are usually car- 
ried out on a press brake. However, with the developments in metal forming theories 
and machine tool design and construction, new techniques have evolved that are em- 
ployed in bending not only sheet metal but also iron angles, structural beams, and 
tubes. Figure 6.28 illustrates the working principles and the stages involved in roll 
bending. As can be seen in the figure, the rolls form a pyramid-type arrangement. Two 
rolls are used to feed the material, whereas the third (roll B) gradually bends it (see 
Figure 6.28a and b). The direction of feed is then reversed, and roll A now gradually 
bends the beam (see Figure 6.28c and d). 

Another bending operation that recently emerged and that is gaining industrial ap- 
plication is rotary bending. Figure 6.29 illustrates the working principles of this oper- 

FIGURE 6.28 

Stages involved in roll 
bending a structural 
beam: (a) feeding; 

(b) initial bending; 

(c) further bending; 

(d) reversing the 
direction of feed 

Roll A 

Roll B 



FIGURE 6.29 

Working principles of 
rotary bending 




6.1 Press Working Operations 


ation. As can be seen, the rotary bender includes three main components: the saddle, 
the rocker, and the die anvil. The rocker is actually a cylinder with a V-notch along its 
length. The rocker is completely secured inside the saddle (i.e., the saddle acts like a 
housing) and can rotate but cannot fall out. The rotary bender can be mounted on a 
press brake. The rocker acts as both a pressure pad and a bending punch. Among the 
advantages claimed for rotary bending are the elimination of the pressure pad and its 
springs (or nitrogen cylinders), lower required tonnage, and the possibility of over- 
bending without the need for any horizontal cams. This new method has been patented 
by the Accurate Manufacturing Association and is nicknamed by industrial personnel 
as the "Pac Man" bending operation. 

A bending process that is usually mistakenly mentioned among the rolling 
processes is the manufacture of thin-walled welded pipes. Although rolls are the form- 
ing tools, the operation is actually a gradual and continuous bending of a strip that is 
not accompanied by any variation in the thickness of that strip. Figure 6.30 indicates 
the basic principles of this process. Notice that the width of the strip is gradually bent 
to take the form of a circle. Strip edges must be descaled and mechanically processed 
before the process is performed to improve weldability. Either butt or high-frequency 
induction welding is employed to weld the edges together after the required circular 
cross section is obtained. This process is more economical and more productive than 
seamless tube rolling. Poor strength and corrosion resistance of seams are considered 
as its main disadvantages. 

Deep Drawing Operation 

Deep drawing involves the manufacture of deep, cuplike products from thin sheet 
metal. As can be seen in Figure 6.31, the tooling basically involves a punch with a 
round corner and a die with a large edge radius. It can also be seen that the punch-die 
clearance is slightly larger than the thickness of the sheet metal. When load is applied 
through the punch, the metal is forced to flow radially and sink into the die hole to 
form a cup. This is an oversimplification of a rather complex problem. For the proper 
design of deep-drawn products as well as the tooling required, we have to gain a 
deeper insight into the process and understand its mechanics. 

Mechanics of deep drawing. Consider what happens during the early stages of ap- 
plying the load. As Figure 6.32a shows, the blank is first bent onto the round edge of 
the die hole. With further increase in the applied load, the part of the blank that was 
bent is straightened in order to sink into the annular clearance between the punch and 

FIGURE 6.30 

Roll bending as 
employed in the 
manufacture of seamed 
















6 Sheet Metal Working 

FIGURE 6.31 

Basic concept of deep 

Blank holder 
The drawn cup 


the die, thus forming a short, straight, vertical wall. Next, the rest of the blank starts to 
flow radially and to sink into the die hole, but because the lower surface of the blank 
is in contact with the upper flat surface of the die steel, frictional forces try to impede 
that flow. These forces are a result of static friction; their magnitude drops as the blank 
metal starts to move. Now consider what happens to a sector of the blank, such as that 
shown in Figure 6.32b, when its metal flows radially. It is clear that the width of the 
sector shrinks so that the large peripheral perimeter of the blank can fit into the smaller 
perimeter of the die hole. This is caused by circumferential compressive stresses act- 
ing within the plane of the blank. With further increase in the applied load, most of the 
blank sinks into the die hole, forming a long vertical wall, while the remaining part of 
the blank takes the form of a small annular flange (see Figure 6.31). The vertical wall 
is subjected to uniaxial tension whose magnitude is increasing when going toward the 
bottom of the cup. 

We can see from the preceding discussion that the deep drawing process involves 
five stages: bending, straightening, friction, compression, and tension. Different parts 
of the blank being drawn are subjected to different states of stress. As a result, the de- 
formation is not even throughout the blank, as is clear in Figure 6.33, which shows an 
exaggerated longitudinal section of a drawn cup. While the flange gets thicker because 
of the circumferential compressive stress, the vertical wall gets thinner, and thinning is 

FIGURE 6.32 

Mechanics of deep 

drawing: (a) first stage 

of deep drawing (i.e., 


(b) compression stage 

in deep drawing 



6.1 Press Working Operations 


FIGURE 6.33 

An exaggerated 
longitudinal section of a 
drawn cup, with the 
states of stress at 
different locations 

Maximum thinning 
occurs here 

G compression, 

Uniaxial tension, 

maximum at the lowest part of the wall adjacent to the bottom of the cup. Accordingly, 
if the cup is broken during the drawing process, failure is expected to occur at the lo- 
cation of maximum thinning. An upper bound for the maximum drawing force can, 
therefore, be given by the following equation: 

F = K x (d + t)tC T (6.3) 

where: F is the maximum required drawing force 

d is the diameter of the punch 

t is the thickness of the blank 

<3 T is the ultimate tensile strength of the blank material 

The blank holder. As previously mentioned, the thin blank is subjected to compres- 
sive stresses within its plane. This is similar to the case of a slender column subjected 
to compression, where buckling is expected to occur if the slenderness ratio (i.e., 
length/thickness) is higher than a certain value. Therefore, by virtue of similarity, if the 
ratio of the diameter of the blank to its thickness exceeds a certain value, buckling oc- 
curs. Actually, if (D -d)lt> 18, where D is the blank diameter, d is the punch diam- 
eter, and t is the thickness, the annular flange will buckle and crimple. This is a product 
defect referred to as wrinkling. 

One way to eliminate wrinkling (buckling) of the thin blank is to support it over 
its entire area. This is done by sandwiching the blank between the upper surface of the 
die steel and the lower surface of an annular ring that exerts pressure upon the blank, 
as shown in Figure 6.31. This supporting ring is called the blank holder, and the force 
exerted on it can be generated by die springs or a compressed gas like nitrogen. On the 
other hand, higher frictional forces will initiate at both the upper and lower surfaces of 
the blank as a result of the blank-holding force. For this reason, lubricants like soap in 
water, waxes, mineral oil, and graphite are applied to both surfaces of the blank. More- 
over, the upper surface of the die steel as well as the lower surface of the blank holder 
must be very smooth (ground and lapped). As a rule of thumb, the blank-holding force 
is taken as 1/3 the force required for drawing. 

228 6 Sheet Metal Working 

Variables affecting deep drawing. Now that we understand the mechanics of the 
process, we can identify and predict the effect of each of the process variables. For ex- 
ample, we can see that poor lubrication results in higher friction forces, and, accord- 
ingly, a higher drawing force is required. In fact, in most cases of poor lubrication, the 
cup cross section does not withstand the high tensile force, and failure of the wall at 
the bottom takes place during the process. A small die corner radius would increase the 
bending and straightening forces, thus increasing the drawing force, and the final out- 
come would be a result similar to that caused by poor lubrication. 

In addition to these process variables, the geometry of the blank has a marked ef- 
fect not only on the process but also on the attributes of the final product. An appro- 
priate quantitative way of expressing the geometry is the number indicating the 
thickness as a percentage of the diameter, or (t/D) x 100. For smaller values of this 
percentage (e.g., 0.5), excessive wrinkling should be expected, unless a high blank- 
holding force is used. If the percentage is higher than 3, no wrinkling occurs, and a 
blank holder is not necessary. 

Another important variable is the drawing ratio, wJaich is given by the following 

R = 4 < 6 - 4 > 


where: R is the drawing ratio 

D is diameter of the blank 
d is the diameter of the punch 

It has been experimentally found that the deep drawing operation does not yield a 
sound cup when the drawing ratio is higher than 2 (i.e., for successful drawing, R must 
be less than 2). 

Another number that is commonly used to characterize drawing operations is the 
percentage reduction. It can be given by the following equation: 

r = ^^xl00 (6.5) 


where: r is the percentage reduction 
D is the diameter of the blank 
d is the diameter of the punch 

It is a common industrial practice to take the value of r as less than 50 percent in order 
to have a sound product without any tearing. When the final product is long and neces- 
sitates a value of r higher than 50 percent, an intermediate cup must be obtained first, as 
shown in Figure 6.34. The intermediate cup must have dimensions that keep the per- 
centage reduction below 50. It can then be redrawn, as illustrated in Figure 6.35, once or 
several times until the final required dimensions are achieved. The maximum permissi- 
ble percentage reduction in the redrawing operations is always far less than 50 percent. 
It is usually taken as 30 percent, 20 percent, and 13 percent, in the first, second, and third 
redraws, respectively. If several redrawing operations are required, the product should 

6.1 Press Working Operations 


FIGURE 6.34 

The use of an 
intermediate cup when 
the total required 
reduction ratio is high 

t I ' 'I 


r = D ~ d X 100 > 50 

? 9 



FIGURE 6.35 

Redrawing an 
intermediate cup 


then be annealed after every two operations in order to eliminate work-hardening and 
thus avoid cracking and failure of the product. 

Blank-development calculations. For the sake of simplicity, it is always assumed that 
the thickness of the blank remains unchanged after the drawing operation. Because the 
total volume of the metal is constant, it can then be concluded that the surface area of 
the final product is equal to the surface area of the original blank. This rule forms the 
basis for the blank-development calculations. Consider the simple example shown in 
Figure 6.36. The surface area of the cup is the area of its bottom plus the area of the 

surface area of cup = —d + ndh 

FIGURE 6.36 

A simple example of 
blank development 



Surface area of blank -D 2 = -d 2 + ndh, i.e., surface area of the cup 
4 4 

230 6 Sheet Metal Working 

This is equal to the surface area of the original blank; therefore, we can state that 

—Dr = —d+ ndh 

4 4 


D 2 = d 2 + 4dh 

Therefore, the original diameter of the blank, which is unknown, can be given by the 
following equation: 

D = Vd 2 + 4dh 


Equation 6.6 gives an approximate result because it assumes the cup has sharp corners, 
which is not the case in industrial practice. However, this equation can be modified to 
take round corners into account by adding the area of the surface of revolution result- 
ing from the rotation of the round corner around the centerline of the cup, when equat- 
ing the area of the product to that of the original blank. Note that the area of any 
surface of revolution can be determined by employing Pappus's first theorem, which 
gives that area as the product of the path of the center of gravity of the curve around 
the axis of rotation multiplied by the length of that curve. 

Planning for deep drawing. The process engineer usually receives a blueprint of the 
required cup from the product designer. His or her job is to determine the dimensions 
of the blank and the number of drawing operations needed, together with the dimen- 
sions of intermediate cups, so that the tool designer can start designing the blanking 
and the deep drawing dies. That job requires experience as well as close contact be- 
tween the product designer and the process engineer. The following steps can be of 
great help to beginners: 

1. Allow for a small flange around the top of the cup after the operation is completed. 
This flange is trimmed at a later stage and is referred to as the trimming allowance. It 
is appropriate to take an allowance equal to 10 to 15 percent of the diameter of the cup. 

2. Calculate the total surface area of the product and the trimming allowance. Then, 
equate it to the area of the original blank with an unknown diameter. Next, solve 
for the diameter of the original blank. 

3. Calculate the thickness as a percentage of the diameter or (t/D) x 100, in order to 
get a rough idea of the degree of wrinkling to be expected (see the preceding dis- 
cussion on process variables). 

4. Calculate the required percentage reduction. If it is less than or equal to 50, then the 
required cup can be obtained in a single drawing. But if the required r is greater 
than 50, then a few redrawing operations are required; the procedure to be followed 
is given in the next steps. 

5. For the first draw, assume r to be equal to 50 and calculate the dimensions of the 
intermediate cup. Then, calculate r required for the first redraw. If r < 30, only a 
single redraw is required. 

6.1 Press Working Operations 


6. If r > 30 for the first redraw, take it as equal to 30 and calculate the dimensions of 
a second intermediate cup. The percentage reduction for the second redraw should 
be less than 20; otherwise, a third redraw is required, and so on. 

Ironing. We can see from the mechanics of the deep drawing operation that there is 
reasonable variation in the thickness of the drawn cup. In most cases, such thickness 
variation does not have any negative effect on the proper functioning of the product, 
and, therefore, the drawn cups are used as is. However, close control of the dimensions 
of the cups is sometimes necessary. In this case, cups are subjected to an ironing op- 
eration, in which the wall of the cup is squeezed in the annular space between a punch 
and its corresponding die. As can be seen in Figure 6.37, the punch-die clearance is 
smaller than the thickness of the cup and is equal to the final required thickness. Large 
reductions in thickness should be avoided in order to obtain a sound product. It is good 
industrial practice to take the value of the punch-die clearance in the range between 30 
and 80 percent of the thickness of the cup. Also, the percentage reduction in thickness, 
which is given next, should fall between 40 and 60 in a single ironing operation. This 
is a safeguard against fracture of the product during the operation. Following is the 
equation to be applied: 

percentage reduction in thickness = 


x 100 


where: t Q is the original thickness of the cup 

tf is the final thickness of the cup after ironing 

Drawing of stepped, conical, and domed cups. Stepped cups are those with two (or 
more) shell diameters (see Figure 6.38a). They are produced in two (or more) stages. 
First, a cup is drawn to have the large diameter, and, second, a redrawing operation is 
performed on only the lower portion of the cup. Tapered or conical cups (see Figure 

FIGURE 6.37 

The ironing operation 

FIGURE 6.38 

Deep-drawn cups: 

(a) stepped; (b) conical; 

(c) domed 









6 Sheet Metal Working 

6.38b) cannot be drawn directly. They first have to be made into stepped cups, which 
are then smoothed and stretched out to give the required tapered cups. A complex deep 
drawing operation is used for producing domed cups (see Figure 6.38c). So that the 
sheet metal stretches properly over the punch nose, higher blank-holding forces are re- 
quired. Therefore, the process actually involves stretch-forming, and its variables 
should be adjusted to eliminate either wrinkling or tearing. 

Drawing of box-shaped cups. When all press working operations of sheet metal are 
reviewed, there would be almost no doubt that the box drawing process is the most 
complex and difficult to control. Nevertheless, in an attempt to simplify the problem, 
we can divide a box into four round corners and four straight sides. Each of these 
round corners represents 1/4 of a circular cup, and, therefore, the previous analysis 
holds true for it. On the other hand, no lateral compression is needed to allow the blank 
metal to flow toward the die edge at each of the straight sides. Accordingly, the process 
in these zones is not drawing at all; it is just bending and straightening. For this rea- 
son, the metal in these zones flows faster than in the round corners, and a square blank 
takes the form shown in Figure 6.39 after drawing. Note that there is excess metal at 
each of the four round corners, which impedes the drawing operations at those loca- 
tions. It also results in localized higher stresses and tears almost always beginning at 
one (or more) of the corners during box drawing, as can be seen in Figure 6.40. 

Several variables affect this complex operation as well as the quality of the products 
obtained. They include the die bending radius, the die corner radius, and the shape of the 
original blank. These process variables have been investigated by research workers, and 
it has been found that in order to obtain sound box-shaped cups, it is very important to 
ensure easy, unobstructed flow of metal during the drawing operation. The absence of 
this condition results in the initiation of high tensile stresses in the vertical walls of the 
box, especially at the round corners, and results in considerable thinning, which is fol- 
lowed by fracture. Among the factors that can cause obstruction to the metal flow are 
smaller die radii, higher reduction ratios (at the corners), and poor lubrication. These are 
added to the presence of excess metal at the corners, which causes an appreciable in- 
crease in the transverse compressive stresses. Therefore, an optimum blank shape with- 
out excess metal at the corners is necessary for achieving successful drawing operations 
of box-shaped cups. A simple method for optimizing the shape of the blank is shown in 
Figure 6.41. It involves printing a square grid on the surface of the blank and determin- 
ing the borders of the undeformed zone on the flanges at each corner (by observing the 

FIGURE 6.39 

Final shape of a box- 
shaped cup, obtained 
by deep drawing a 
square blank 

6.1 Press Working Operations 


FIGURE 6.40 

Tears occurring in box 

FIGURE 6.41 

Optimized blank shape 
for drawing box-shaped 

undistorted grid) so that it can be taken off the original blank. It has been found that the 
optimum shape is a circle with four cuts corresponding to the four corners. Also, the 
blank-holding force has been found to play a very important role. Better products are ob- 
tained by using a rubber-actuated blank holder that exerts low forces during the first 
third of the drawing stroke, followed by a marked increase in those forces during the rest 
of the drawing stroke to eliminate wrinkling and stretch out the product. 


Sheet Metal Working 

FIGURE 6.42 

Optimized blank shape 
for drawing cups with 
an irregular cross 



Cross section of the 
deep-drawn part 

The preceding discussion can be generalized to include the drawing of a cup with 
an irregular cross section. This can be achieved by dividing the perimeter into straight 
sides and circular arcs. Professor Kurt Lange and his coworkers (Institute Fur Um- 
formstechnik, Stuttgart Universitate) have developed a technique for obtaining the op- 
timum blank shape in this case by employing the slip-line field theory. The technique 
was included in an interactive computer expert system that is capable of giving direct 
answers to any drawing problem. An optimized blank shape obtained by that system is 
shown in Figure 6.42. 

Recent developments In deep drawing. A recent development in deep drawing in- 
volves cup drawing without a blank holder. Cupping of a thick blank has been ac- 
complished by pushing the blank through a die having a special profile, as shown in 
Figure 6.43, without any need for a blank holder. This process has the advantages of 
reducing the number of processing stages, eliminating the blank holder, and using 
considerably simpler tool construction. A further advantage is that the operation can 
be performed on a single-acting press, resulting in an appreciable reduction in the ini- 
tial capital cost required. 

Another new development is the employment of ultrasonics to aid the deep draw- 
ing operation. The function of the ultrasonic waves is to enlarge the die bore and then 
leave it to return elastically to its original dimension in a pulsating manner. This re- 
duces the friction forces appreciably, resulting in a marked reduction in the required 
drawing force and in a clear improvement of the quality of the drawn cup. In many 
cases, the cup can be drawn by the force exerted by the human hand without the need 

FIGURE 6.43 

Drawing cups without a 
blank holder 



6.1 Press Working Operations 


for any mechanical force-generating device. It is, therefore, obvious that low-tonnage, 
high-production-rate presses can be used, which makes the process economically at- 

Defects in deep-drawn parts. These defects differ in shape and cause, depending 
upon the prevailing conditions and also on the initial dimensions of the blank. Fol- 
lowing is a brief description of the most common defects, some of which are shown in 
Figure 6.44: 

1. Wrinkling. Wrinkling is the buckling of the undrawn part of the blank under com- 
pressive stresses; it may also occur in the vertical walls (see Figure 6.44a and b). If 
it takes place on the punch nose when drawing a domed cup, it is referred to as 

2. Tearing. Tearing, which always occurs in the vicinity of the radius connecting the 
cup bottom and the wall, is caused by high tensile stresses due to the obstruction of 
the flow of the metal in the flange. 

3. Earing. Earing is the formation of ears at the free edges of a deep-drawn cylindri- 
cal cup (see Figure 6.44c). It is caused by the anisotropy of the sheet metal. Ears 
are trimmed after a drawing operation, resulting in a waste of material. 

4. Surface irregularities. Surface irregularities are caused by nonuniform yielding, 
like the orange-peel effect of Luder's lines. 

5. Surface marks. Surface marks are caused by improper punch-die clearance or poor 
lubrication. These include draw marks, step rings, and burnishing. 

Forming Operations 

In this section, we will discuss the various forming operations performed on sheet met- 
als — not just flat sheets, but tubular sheets (i.e., thin-walled tubes) as well. Therefore, 
not only will operations like embossing and offsetting be discussed, but also tube 
bulging, expanding, and necking will be considered. 

Forming of sheets. True forming involves shaping the blank into a three-dimensional 
(or sculptured) surface by sandwiching it between a punch and a die. The strain is not 
uniform, and the operation is complex. The nonhomogeneity (or complexity) depends 
upon the nature and the unevenness of the required shape. Experience and trial and 
error were employed in the past to obtain an optimum blank shape and to avoid thin- 
ning the blank or tearing. 

FIGURE 6.44 

Some defects occurring 
in deep drawing 
operations: (a) wrinkling 
in the flange; 

(b) wrinkling in the wall; 

(c) earing 




6 Sheet Metal Working 

A printed grid on the original blank helps to detect the locations of overstraining 
where tearing is expected. It also helps in optimizing the shape of the original blank. 
With recent advances in computer graphics and simulation of metal deformation, ra- 
tional design of the blank can be performed by the computer, without any need for trial 
and error. In fact, a successful software package has been prepared by the Mechanical 
Engineering Department of Michigan Technological University. 

Embossing operations. Embossing operations involve localized deflection of a flat 
sheet to create depressions in the form of beads and offsets. This is sometimes called 
oil canning. Beads and offsets are usually employed to add stiffness to thin sheets, 
whether flat or tubular (e.g., barrels), as well as for other functional reasons. A typi- 
cal example of a part that is subjected to embossing is the license plate of an auto- 
mobile. The cross section of a bead can take different forms, such as those shown in 
Figure 6.45. Because this operation involves stretching the sheet, the achieved local- 
ized percentage elongation within the bead cross section must be lower than that al- 
lowable for the metal of the sheet. On the other hand, Figure 6.46 shows two kinds 
of offsets, where it is common practice to take the maximum permissible depth as 
three times the thickness of the sheet metal. 

Rubber forming of flat sheets. Rubber forming is not new and actually dates back to 
the nineteenth century, when a technique for shearing and cutting paper and foil was 
patented by Adolph Delkescamp in 1872. Another rubber forming technique, called the 
Guerin process, was widely used during World War II for forming aircraft panels. It in- 
volved employing a confined rubber pad on the upper platen of the press and a steel 
form block on the lower platen, as shown in Figure 6.47a. This method is still some- 
times used. As can be seen in the figure, when a block of elastomer (usually incom- 
pressible artificial rubber) is confined in a rigid box, the only way it can flow when the 
punch sinks into it is up, thus forming the blank around the punch under uniform pres- 
sure over the whole surface. It is also common industrial practice to place spacers on 
the base of the metal box in order to provide a relief for the elastomer block, which, in 
turn, helps to avoid the initiation of high localized strains in the blank area directly be- 
neath the punch. Rubber forming has real potential when the number of parts required 
is relatively small and does not justify designing and constructing a forming die. 

A modified version of this process, called the hydroform process, involves em- 
ploying a pressurized fluid above the rubber membrane, as shown in Figure 6.47b. 

FIGURE 6.45 

Different kinds of 


1 c 


Flat V bead 

Round bead 

FIGURE 6.46 

Offsetting operations 


Interior offset 

Edge offset 

6.1 Press Working Operations 


FIGURE 6.47 

Rubber forming of flat 
sheets: (a) conventional 
rubber forming; 
(b) hydroform process 








This is similar in effect to drawing the cup into a high-pressure container, as previously 
mentioned. Therefore, percentage reductions higher than those obtained in conven- 
tional deep drawing can be achieved. 

Forming of tubular sheets. Figure 6.48a through d indicates tubular parts after they 
were subjected to beading, flattening, expanding, and necking operations, respectively. 
Tube bulging is another forming operation, in which the diameter of the tube, in 
its middle part, is expanded and then restrained by a split die and forced to conform 
to the details of the internal surface of the die. This can be achieved by internal hy- 
draulic pressure or by employing an elastomer (polyurethane) rod as the pressure- 
transmitting medium, causing expansion of the tube. A schematic of this operation is 

FIGURE 6.48 

Different tubular parts 
after forming 
operations: (a) beading; 

(b) flattening; 

(c) expanding; 

(d) necking 



-^7777 . 




6 Sheet Metal Working 

FIGURE 6.49 

The tube-bulging 
operation with an 
elastomer rod 


Die holder 


given in Figure 6.49. At the beginning of the operation, the elastomer rod fits freely 
inside the tube and has the same length. Compressive forces are then applied to both 
the rod and the tube simultaneously so that the tube bulges outward in the middle 
and the frictional forces at the tube-rod interface draw more metal into the die space, 
thus decreasing the length of the tube. The method of using a polyurethane rod is 
simpler and cleaner, and there is no need for using oil seals or complicated tooling 
construction. A further advantage of rubber bulging is that it can be used for simul- 
taneous forming, piercing, and shearing of thin tubular sheets. 


In HERF, the energy of deformation is delivered within a very short period of time — 
on the order of milliseconds or even microseconds. HERF methods include explosive, 
electrohydraulic, and electromagnetic forming techniques. These techniques are usu- 
ally employed when short-run products or large parts are required. HERF is also rec- 
ommended for manufacturing prototype components and new shapes in order to avoid 
the unjustifiable cost of dies. Rocket domes and other aerospace structural panels are 
typical examples. During a HERF process, the sheet metal is given an extremely high 
acceleration in a very short period of time and is thus formed as a result of consuming 
its own kinetic energy to cause deformation. 

Explosive Forming 

Explosive forming of sheet metal received some attention during the past decade. The 
various explosive forming techniques fall under one or the other of two basic systems: 
confined and unconfined. In a confined system, which is shown in Figure 6.50a, a 
charge of low explosives is detonated and yields a large amount of high-pressure gas, 
thus forcing the sheet metal to take the desired shape. This system is mainly used for 

6.2 High-Energy-Rate Forming (HERF) 


FIGURE 6.50 

Explosive forming of 
sheet metal: 

(a) confined system; 

(b) standoff system 



Die steel 



bulging and flaring of small tubular parts. Its main disadvantage is the hazard of die 
failure because of the high pressure generated. 

In an unconfined, or standoff, system, which is shown in Figure 6.50b, the charge is 
maintained at a distance from the sheet blank (the standoff distance), and both the blank 
and the charge are kept immersed in water. When the charge is detonated, shock waves 
are generated, thus forming a large blank into the desired shape. It is obvious that the ef- 
ficiency of the standoff system is less than that of the confined system because only a 
portion of the surface over which the shock waves act is utilized (actually, shock waves 
act in all directions, forming a spherical front). However, the standoff system has the ad- 
vantages of a lower noise level and of largely reducing the hazard of damaging the 
workpiece by particles resulting from the explosion. In a simple standoff system, the dis- 
tance from the explosive charge to the water surface is usually taken as twice the stand- 
off distance. The latter depends upon the size of the blank and is taken as equal to D (the 
blank diameter) for D less than 2 feet (60 cm) and is taken as equal to 0.5D for D greater 
than that. Best results are obtained when the blank is clamped lightly around its periph- 
ery and when a material with a low modulus of elasticity, like plastic, is used as a die 
material. This eliminates springback, thus obtaining closer tolerances. A modified ver- 
sion of this method is illustrated in Figure 6.5 1, where a reflector is used to collect and 

FIGURE 6.51 

Increasing the 
efficiency of explosive 
forming by using a 


To vacuum 




Sheet Metal Working 

reflect the explosion energy that does not fall directly onto the blank surface. This leads 
to improved efficiency over the standoff system because a smaller amount of charge is 
needed for the same job. 

Electrohydraulic Forming 

The basic idea for the process of electrohydraulic forming, which has been known for 
some time, is based on discharging a large amount of electrical energy across a small 
gap between two electrodes immersed in water, as shown in Figure 6.52. The high- 
amperage current resulting from suddenly discharging the electrical energy from the 
condensers melts the thin wire between the electrodes and generates a shock wave. 
The shock wave lasts for a few microseconds; it travels through water to hit the 
blank and forces it to take the shape of the die cavity. The use of a thin wire between 
the electrodes has the advantages of initiating and guiding the path of the spark, en- 
abling the use of nonconductive liquids; also, the wire can be shaped to suit the 
geometry of the required product. The method is also safer than explosive forming 
and can be used for simultaneous operations like piercing and bulging. Nevertheless, 
it is not suitable for continuous production runs because the wire has to be replaced 
after each operation. Moreover, the level of energy generated is lower than that of ex- 
plosive forming. Therefore, the products are generally smaller than those produced 
by explosive forming. 

Electromagnetic Forming 

Electromagnetic forming is another technique based on the sudden discharge of elec- 
trical energy. As we know from electricity and magnetism in physics, when an electric 
current passes through a coil, it initiates a magnetic field whose magnitude is a func- 
tion of the current. We also know that when a magnetic field is interrupted by a con- 
ductive material (workpiece), a current is induced in that material that is proportional 
to the rate of change of the flux. This is called eddy current and produces its own mag- 
netic field that opposes the initial one. As a result, repulsive forces between the coil 
and the workpiece force the workpiece to conform to the die cavity. This technique can 
be used to form flat as well as tubular sheets. As can be seen in Figure 6.53, it is em- 
ployed in expanding as well as compressing tubes. It has proven to be very effective 
when forming relatively thin materials. 

FIGURE 6.52 

Electrohydraulic forming 



IN-h. , r 




6.3 Spinning of Sheet Metal 


FIGURE 6.53 

Examples of 
forming of tubes 



Spinning is the forming of axisymmetric hollow shells over a rotating-form mandrel by 
using special rollers. Generally, the shapes produced by spinning can also be manu- 
factured by drawing, compressing, or flanging. However, spinning is usually used for 
forming large parts that require very large drawing presses or when there is a diversity 
in the products (i.e., when various shapes are needed but only a small number of each 
shape is required). 

A schematic of the spinning operation is shown in Figure 6.54. At the beginning, 
the semifinished product (circular blank) is pushed by the tail stock against the front of 
the form mandrel (usually a wooden one) that is fixed on the rotating faceplate of the 
spinning machine (like a lathe). A pressing tool is pushed by the operator onto the ex- 
ternal surface of the blank. The blank slips under the pressing devices, which causes 
localized deformation. Finally, the blank takes the exact shape of the form mandrel. 
This technique can also be used to obtain hollow products with a diameter at the end 
(neck) smaller than that at the middle. In this case, it is necessary to use a collapsible- 
form mandrel, which is composed of individual smaller parts that can be extracted 
from the neck of the final product after the process is completed. Figure 6.55 shows a 
group of parts produced by spinning. 

FIGURE 6.54 

A schematic of the 
spinning operation 


Sheet Metal Working 

FIGURE 6.55 

A group of parts 
produced by spinning 

A modified version of this method involves replacing the operator by a numeri- 
cally controlled (NC) tool. Auxiliary operations, like removing the excess metal, are 
also carried out on the same machine. Better surface quality and more uniform thick- 
ness are the advantages of NC spinning over the conventional techniques. 

Review Questions 


1. What main design feature characterizes sheet 
metal products? 

2. List some of the advantages of press working 
sheet metals. 

3. When are sheet metals formed in their hot 
state? Give examples. 

4. What are the two main groups of press working 

5. What main condition must be fulfilled so that 
cutting of sheet metal (and not any other opera- 
tion) takes place? 

6. Use sketches to explain why the angle of incli- 
nation of the upper blade of guillotine shears 
must not exceed 15°. 

7. Use sketches to differentiate between the fol- 
lowing operations: shearing, cutoff, parting, 
blanking, and punching. 

8. Why must attention be given to careful layout 
of blanks on a sheet metal strip? 

9. Describe a perforating operation. 

10. What does an edge of a blank usually look like? 
Draw a sketch. 

11. What is meant by the percentage penetration? 

12. What does an edge of a blank look like when 
the punch-die clearance is too large? 

13. What does an edge of a blank look like when 
the punch-die clearance is too tight? 

Chapter 6 Review Questions 


14. When are punches sheared and why? 

15. When are dies sheared and why? 

16. In what aspect is fine blanking different from 
conventional blanking? 

17. Use sketches to explain each of the following 
operations: shaving, piercing, and cropping. 

18. Can a drop-through die be used on any press? 
Why not? 

19. What is the function of a stripper plate? 

20. List two types of die constructions for blanking 

21. How can a washer be produced in a single 

22. What condition must be fulfilled so that bend- 
ing of sheet metal takes place? 

23. Sketch the common types of bending dies. 

24. Which die requires the minimum force for the 
same thickness of sheet metal? 

25. Where is tearing expected to occur and where 
is wrinkling expected to occur when a sheet 
metal is subjected to bending? 

26. What is springback? Why does it occur? 

27. List three methods for eliminating the effects of 

28. On what assumption is blank development 

29. List some operations that can be classified as 
bending. Use sketches and explain design func- 
tions of the products. 

30. How can structural angles be bent? 

31. Explain rotary bending, using sketches, and list 
some of the advantages of this operation. 

32. Explain how a seamed tube can be produced by 
continuous bending. 

33. What are the disadvantages of seamed tubes? 

34. Explain deep drawing, using sketches. 

35. What are the stages involved in deep drawing a 
circular cup? Explain, using sketches. 

36. Indicate the states of stress at different locations 
in a cup toward the end of a drawing operation. 

37. Where is thickening expected to occur? 

38. At what location is thinning maximum? To 
what would this lead? 

39. Why is a blank holder sometimes needed? 

40. List some of the variables affecting the deep 
drawing operation. 

41. What is wrinkling? Why does it occur? 

42. Describe an ironing operation. 

43. Is there any limitation on ironing? 

44. Why are conical cups not drawn directly? 

45. What is actually taking place when drawing 
domed cups? 

46. Is it feasible to take any blank shape for box 
drawing operations and then trim the excess 
metal? Why? 

47. What are the mechanics of deformation in the 
straight-sides areas? 

48. What is the advantage of ultrasonic deep draw- 

49. How can plates be drawn without a blank 

50. List some of the defects experienced in deep- 
drawn products. 

51. As a product designer, how can you make use 
of the embossing operation when designing 
sheet metal parts? 

52. When would you recommend using rubber 
forming techniques? 

53. What is meant by high-energy-rate forming? 

54. When would you recommend using explosive 

55. Should the dies used in explosive forming be 
made of a hard material, like alloy steel, or a 
softer one, like plastic? Why? 

56. What happens if you make the hydraulic head 
very small in explosive forming? 


6 Sheet Metal Working 

57. What are the advantages of electrohydraulic 
forming? What are the disadvantages? 

58. Use a sketch to explain the electromagnetic 
forming operation. 

59. Describe spinning. When is it recommended? 

60. Can products with a diameter at the neck 
smaller than at the middle be produced by spin- 
ning? How? 



1. The blank shown in Figure 6.56 is to be pro- 
duced from a sheet metal strip 0.0625 inch (1.6 
mm) in thickness. Material is low-carbon steel 
AISI 1020. Estimate the required blanking force. 

2. The products shown in Figure 6.57a, b, and c are 
produced by bending. Obtain the length of the 
original blank to the nearest 0.01 inch (0.25 
mm). Take / as 0.0625 inch (1.6 mm). 

3. A cup is drawn from a sheet of 1020 steel. The 
thickness is 0.03 inch (0.8 mm), and the inner di- 

ameter is 1 inch (25 mm). Estimate the maxi- 
mum force required for drawing. If the material 
is aluminum, what would the force be? 

A cup with a height of 0.75 inch (18.75 mm) and 
an inner diameter of 1 inch (25 mm) is to be 
drawn from a steel strip 0.0625 inch ( 1 .6 mm) in 
thickness. Plan for the drawing process by carry- 
ing out blank development, determining the 
number of drawings, and looking at the draw 
severity analysis. 

FIGURE 6.56 

The blank shape 
required in Problem 1 

1.5 in. 
(37.5 mm) 

R = 0.25 in. 
(6 mm) 

Chapter 6 Design Example 


FIGURE 6.57 

Products produced by 
bending in Problem 2 

R = 0.5 in. 
(12.5 mm) 

R =0.75 in. 

18.75 mm) 


Design Example 



Design a simple wrench to loosen (or tighten) a 1/2-inch (12.5-mm) nut (or bolt head). 
The 1/2 inch (12.5 mm) measures across the nut flats. The torque is 1 lb ft (1.356 
Nm), and 50,000 pieces are required annually. The wrench is to be produced by press 


A suitable method for production is fine blanking as there will be no need for any fur- 
ther machining operations. We cannot select a steel that has a high carbon content be- 
cause it will create problems during the fine-blanking operation. An appropriate choice 
is AISI 1035 CD steel. The dimensions of the wrench are the same as those given in 
the examples on forging and casting, although the tolerances can be kept much tighter. 
A detailed design is given in Figure 6.58. 


Sheet Metal Working 

FIGURE 6.58 

Detailed design of a 
wrench produced by 

0.75 inch 

Now it is time to check the maximum tensile stress due to bending: 

/ = -±-bh 3 = ^(0.25)(0.75) 3 = 0.10546 x 10" 2 in. 4 

q = Afr = l x 12 x 0.375 42851b/in2 
/ 0.10546 x 10" 2 
It is less than the allowable stress for 1035 steel, which is about 20,000 lb/in. 


Design Project s 

1. A pulley (for a V-belt) that has 4-inch (100-mm) outer diameter and is mounted on 
a shaft that is 3/4 inch (19 mm) in diameter was manufactured by casting. The 
process was slow, and the rejects formed a noticeable percentage of the production. 
As a product designer, you are required to redesign this pulley so that it can be pro- 
duced by sheet metal working and welding. 

2. Design a connecting rod for a sewing machine so that it can be produced by sheet 
metal working, given that the diameter of each of the two holes is 0.5 inch (12.5 
mm) and the distance between the centers of the holes is 4 inches (100 mm). 

3. If a connecting rod four times smaller than the one of Design Project 2 is to be used 
in a little toy, how would the design change? 

4. Design a table for the machine shop. The table should be 4 feet in height, with a 
surface area of 3 by 3 feet (900 by 900 mm), and should be able to carry a load 

Chapter 6 Design Projects 


of half a ton. Assume that 4000 pieces are required annually and that different 
parts will be produced by sheet metal working and then joined together by nuts 
and bolts. 

5. A trash container having a capacity of 1 cubic foot (0.02833 m 3 ) is to be designed 
for manufacturing by sheet metal working. Assume that it is required to withstand 
an axial compression load of 200 pounds (890 N) and that the production rate is 
50,000 pieces per year. Provide a detailed design for this trash container. 

6. A connecting lever is produced by forging. The lever has two short bosses, each at 
one of its ends and each with a vertical hole that is 3/4 inch (19 mm) in diameter. 
The horizontal distance between the centers of the holes is 12 inches (300 mm), and 
the vertical distance is 3 inches (75 mm). The lever during functioning is subjected 
to a bending moment of 200 lb ft (271 Nm). Because of the high percentage of re- 
jects and low production rate, this connecting lever is to be produced by sheet metal 
working. Provide a detailed design so that it can be produced by this manufactur- 
ing method. 

Chapter 7 

wder — 


Powder metallurgy is the technology of producing useful components shaped 
from metal powders by pressing and simultaneous or subsequent heating to 
produce a coherent mass. The heating operation is usually performed in a con- 
trolled-atmosphere furnace and is referred to as sintering. The sintering tem- 
perature must be kept below the melting point of the powder material or the 
melting point of the major constituent if a mixture of metal powders is used. 
Therefore, sintering involves a solid-state diffusion process that allows the 
compacted powder particles to bond together without going through the molten 
state. This, in fact, is the fundamental principle of powder metallurgy. 

Historical background. Although powder metallurgy is becoming increasingly im- 
portant in modern industry, the basic techniques of this process are very old 
indeed. The ancient Egyptians used a crude form of powder metallurgy as early 
as 3000 b.c. to manufacture iron implements. The technique involved reducing 
the ore with charcoal to obtain a spongy mass of metal that was formed by fre- 
quent heating and hammering to eject the slag and consolidate the iron parti- 
cles together into a mass of wrought iron. This process was used because the 
primitive ovens then available were not capable of melting iron. The same tech- 
nique was used later by smiths in India about a.d. 300 to manufacture the well- 
known Delhi pillar weighing 6.5 tons. This method was superseded when more 
advanced ovens capable of melting ferrous metals came into being. 

At the beginning of the nineteenth century, powder metallurgy had its first 
truly scientific enunciation, in England, when Wallaston published details of the 


7.1 Metal Powders 249 

preparation of malleable platinum. As had happened in the past, Wallaston's 
technique was superseded by melting. However, the need for the powder met- 
allurgy process arose again to satisfy the industrial demand for high-melting- 
point metals. An important application was the production of ductile tungsten 
in 1909 for manufacturing electric lamp filaments. 

Why powder metallurgy? As a result of the development of furnaces and melt- 
ing techniques, the powder-consolidation process is now usually used when 
melting metal is undesirable or uneconomical. Fusion is not suitable when it is 
required to produce parts with controlled, unique structures, such as porous 
bearings, filters, metallic frictional materials, and cemented carbides. Also, it 
has been found that powder metallurgy can produce certain complicated 
shapes more economically and conveniently than other known manufacturing 
processes. For this reason, the process currently enjoys widespread industrial 
application. As the price of labor and the cost of materials continue to rise, the 
powder-consolidation technique is becoming more and more economical be- 
cause it eliminates the need for further machining operations, offers more ef- 
ficient utilization of materials, and allows components to be produced in 
massive numbers with good surface finish and close tolerances. 


The Manufacture of Metal Powders 

Different methods are used for producing metal powders. They include reduction of 
metal oxides, atomization of molten metals, electrolytic deposition, thermal decompo- 
sition of carbonyls, condensation of metal vapor, and mechanical processing of solid 

Reduction. In reduction, the raw material is usually an oxide that is subjected to a se- 
quence of concentration and purification operations before it is reduced. Carbon, car- 
bon monoxide, and hydrogen are used as reducing agents. Following is the chemical 
formula indicating the reaction between carbon and iron oxide: 

2Fe,0, + 3C — > 4Fe + 3C0 2 T 


Because the reaction takes place at a high temperature, the resulting metal particles 
sinter together and form sponges that are subsequently crushed and milled to a powder 
suitable for consolidation. Such powders have low apparent densities and often contain 
impurities and inclusions, but they are cheap. Metal powders produced by this method 
include iron, cobalt, nickel, tungsten, and molybdenum. 


7 Powder Metallurgy 

Atomization. Atomization is frequently used for producing powders from low-melt- 
ing-point metals such as tin, lead, zinc, aluminum, and cadmium. Iron powder can also 
be produced by atomization. The process involves forcing a molten metal through a 
small orifice to yield a stream that is disintegrated by a jet of high-pressure fluid. When 
compressed gas is used as the atomizing medium, the resulting powder particles will 
be spherical. The reason is that complete solidification takes a relatively long period, 
during which surface tension forces have the chance to spheroidize the molten metal 
droplets. However, when water is used, the droplets solidify very quickly and have a 
ragged or irregular form. Figure 7.1 illustrates the atomization technique. 

Electrolytic deposition. Electrolytic deposition involves obtaining metal powders 
from solutions by electrolysis. Process parameters such as current density and solution 
concentration are controlled to give a loose deposit instead of the coherent layer ac- 
quired in electroplating. The electrolytically deposited powders are then carefully 


Production of metal 
powders by atomization 

Stream of 
molten metal 


7.1 Metal Powders 251 

washed, dried, and annealed. Such powders are relatively expensive, but their impor- 
tant advantage is their high purity and freedom from nonmetallic inclusions. 

Thermal decomposition of carbonyls. Nickel and iron carbonyls are volatile liquids 
having low boiling points of 110°F and 227°F (43°C and 107°C), respectively. They 
decompose at temperatures below 572°F (300°C), and the metal is precipitated in the 
form of a very fine powder. 

Condensation of metal vapor. Condensation is employed only with some low-melt- 
ing-point metals. For example, zinc powder can be obtained directly by condensation 
of the zinc vapor. 

Mechanical processing of solid metals. Production of metal powders by comminua- 
tion of solid metals is accomplished by either machining, crushing, milling, or any 
combination of these. This method is limited to the production of beryllium and mag- 
nesium powders because of the expenses involved. 

Properties of Metal Powders 

The particular method used for producing a metal powder controls its particle and bulk 
properties, which, in turn, affect the processing characteristics of that powder. There- 
fore, comprehensive testing of all the physical and chemical properties of powders is 
essential prior to use in order to avoid variations in the final properties of the com- 
pacts. Following are the important characteristics of metal powders. 

Chemical composition. In order to determine the chemical composition, conventional 
chemical analysis is used in addition to some special tests that are applicable only to 
metal powders, such as weight loss after reduction in a stream of hydrogen, which is 
an indirect indication of the amount of oxide present. For example, in the case of iron 
powder, the following equation is used: 

159 7 
% iron oxide = % weight loss x — — — (7.2) 


= % weight loss x 3.33 (7.3) 

In Equation 7.2, the fraction on the right-hand side is the ratio of the total weight of 
iron oxide to the weight of the combined oxygen in it, or (Fe 2 3 )/(0 3 ), which can be 
calculated by summing up the atomic weights of each element in the numerator and 

It is also important to mention that the percentages of nonmetallic inclusions will 
affect the maximum achievable density of the compacted powder (i.e., the full theo- 
retical density). For example, if an iron powder (density of iron is 7.87 g/cnv ) consists 
of a percent Fe 2 3 , b percent carbon, and c percent sulfur, the following equation can 
be applied: 


max. achievable density = ioo - (a + b + c) a b c ^'^ 

+ + + 

'•o/ Poxide Pcarbon Psulfur 

252 7 Powder Metallurgy 

where p oxide , p ca rbon> and Psuifur are the densities of oxide, carbon, and sulfur, respec- 
tively. Equation 7.4 can also be used in calculating the maximum achievable density 
for a mixture of powders. 

Particle shape. The particle shape is influenced by the method of powder production 
and significantly affects the apparent density of the powder, its pressing properties, and 
its sintering ability. 

Particle size. The flow properties and the apparent density of a metal powder are 
markedly influenced by the particle size, which can be directly determined by mea- 
surement on a microscope, by sieving, or by sedimentation tests. 

Particle-size distribution. The particle-size distribution has a considerable effect on 
the physical properties of the powder. Sieve testing is the standard method used for the 
determination of the particle-size distribution in a quantitative manner. The apparatus 
used involves a shaking machine on which a series of standard sieves are stacked with 
the coarsest at the top and the finest at the bottom. The particle-size distribution is ob- 
tained from the percentage (by weight) of the sample that passes through one sieve but 
is retained on the next finer sieve. These sieves are defined by the mesh size, which in- 
dicates the number of apertures per linear inch. After the test is performed, the results 
are stated in a suitable form, such as a table of weight percentages, graphs of frequency 
distribution, or cumulative oversize and undersize curves where the cumulative size is 
the total weight percentage above or below a particular mesh size. 

Specific surface. Specific surface is the total surface area of the particles per unit 
weight of powder, usually expressed in square centimeters per gram (cm'/g). The spe- 
cific surface has a considerable influence on the sintering process. The higher the spe- 
cific surface, the higher the activity during sintering because the driving force for 
bonding during the sintering operation is the excess energy due to the large area (high 
specific surface). 

Flowability. Flowability is the ease with which a powder will flow under gravity 
through an orifice. A quantitative expression of the flowability of a powder is its flow 
rate, which is determined using a Hall flowmeter. As illustrated in Figure 7.2, this ap- 
paratus involves a polished conical funnel made of brass having a half-cone angle of 
30° and an orifice of 0.125 inch (3.175 mm). The funnel is filled with 50 grams of the 
powder, and the time taken for the powder to flow from the funnel is determined, the 
flow rate being expressed in seconds. The flow properties are dependent mainly upon 
the particle shape, particle size, and particle-size distribution. They are also affected by 
the presence of lubricants and moisture. Good flow properties are required if high pro- 
duction rates are to be achieved in pressing operations because the die is filled with 
powder flowing under gravity and because a shorter die-filling time necessitates a high 
powder-flow rate. 

Bulk (or apparent) density. The bulk (or apparent) density is the density of the bulk 
of a powder mass. It can be easily determined by filling a container of known volume 
with the powder and then determining the weight of the powder. The bulk density is 
the quotient of the powder mass divided by its volume and is usually expressed in 
grams per cubic centimeter (g/cm 3 ). The apparent density is influenced by the same 

7.1 Metal Powders 



A sketch of the Hall 

1/8 in. or 1/10 

factors as the flowability — namely, the particle configuration and the particle-size dis- 

Compressibility and compactibility. Compressibility and compactibility are very im- 
portant terms that indicate and describe the behavior of a metal powder when com- 
pacted in a die. Compressibility indicates the densification ability of a powder, whereas 
compactibility is the structural stability of the produced as-pressed compact at a given 
pressure. A generalized interpretation of these terms involves graphs indicating the as- 
pressed density versus pressure (for compressibility) and the as-pressed strength ver- 
sus pressure (for compactibility). It must be noted that these two terms are not 
interchangeable: A brittle powder may have good compressibility but usually has a 
weak as-pressed compactibility. 

Sintering ability. Sintering ability is the ability of the adjacent surfaces of particles in 
an as-pressed compact to bond together when heated during the sintering operation. 
Sintering ability is influenced mainly by the specific surface of the powder used and is 
the factor responsible for imparting strength to the compact. 

Factors Affecting the Selection 
of Metal Powders 

Probably all metallic elements can be made in powderous form by the previously dis- 
cussed manufacturing methods. However, the powder characteristics will differ in each 
case and will depend mainly upon the method of manufacture. The task of the manu- 
facturing engineer is to select the type of powder appropriate for the required job. The 
decision generally depends upon the following factors: 

1. Economic considerations 

2. Purity demands 

3. Desired physical, electrical, or magnetic characteristics of the compact 

These considerations will be discussed in a later section. 

254 7 Powder Metallurgy 


The conventional powder metallurgy process normally consists of three operations: 
powder blending and mixing, powder pressing, and compact sintering. 

Blending and Mixing 

Blending and mixing the powders properly is essential for uniformity of the finished 
product. Desired particle-size distribution is obtained by blending in advance the types 
of powders used. These can be either elemental powders, including alloying powders 
to produce a homogeneous mixture of ingredients, or prealloyed powders. In both 
cases, dry lubricants are added to the blending powders before mixing. The commonly 
used lubricants include zinc stearate, lithium stearate, calcium stearate, stearic acid, 
paraffin, acra wax, and molybdenum disulfide. The amount of lubricant added usually 
ranges between 0.5 and 1.0 percent of the metal powder by weight. The function of the 
lubricant is to minimize the die wear, to reduce the friction that is initiated between the 
die surface and powder particles during the compaction operation, and, hence, to ob- 
tain more even density distribution throughout the compact. Nevertheless, it is not rec- 
ommended that the just-mentioned limits of the percentage of lubricant be exceeded, 
as this will result in extruding the lubricant from the surfaces of the particles during 
compaction to fill the voids, preventing proper densification of the powder particles 
and impeding the compaction operation. 

The time for mixing may vary from a few minutes to days, depending upon oper- 
ator experience and the results desired. However, it is usually recommended that the 
powders be mixed for 45 minutes to an hour. Overmixing should always be avoided 
because it may decrease particle size and work-harden the particles. 


Pressing consists of filling a die cavity with a controlled amount of blended powder, 
applying the required pressure, and then ejecting the as-pressed compact, usually 
called the green compact, by the lower punch. The pressing operation is usually per- 
formed at room temperature, with pressures ranging from 10 tons/in. 2 (138 MPa) to 
60 tons/in.^ (828 MPa), depending upon the material, the characteristics of the pow- 
der used, and the density of the compact to be achieved. 

Tooling is usually made of hardened, ground, and lapped tool steels. The final 
hardness of the die walls that will come in contact with the powder particles during 
compaction should be around 60 R c in order to keep the die wear minimal. The die 
cavity is designed to allow a powder fill about three times the volume (or height) of 
the green compact. The ratio between the initial height of the loose powder fill and the 
final height of the green compact is called the compression ratio and can be deter- 
mined from the following equation: 

7.2 Powder Metallurgy: The Basic Process 


compression ratio 

height of loose powder fill 
height of green compact 

density of green compact 
apparent density of loose powder 


When pressure is first applied to metal powders, they will undergo repacking or 
restacking to reduce their bulk volume and to attain better packing density. The extent 
to which this occurs depends largely on the physical characteristics of the powder par- 
ticles. The movement of the powder particles relative to one another will cause the 
oxide films covering their surfaces to be rubbed off. These oxide films will also col- 
lapse at the initial areas of contact between particles because these areas are small and 
the magnitude of the localized pressures are, therefore, extremely high. This leads to 
metal-to-metal contact and, consequently, to cold-pressure welding between the pow- 
der particles at the points of contact. When the pressure is further increased, interlock- 
ing and plastic deformation of the particles take place, extending the areas of contact 
between the individual particles and increasing the strength and density of the coher- 
ent compacted powder. Plasticity of the metal-powder particles plays a major role dur- 
ing the second stage of the pressing operation. As the compaction pressure increases, 
further densification is increasingly retarded by work-hardening of the particle mater- 
ial and by friction. Figure 7.3 shows a typical plot of the relationship between the 
achieved density and the compaction pressure. As can be seen, the density first goes up 
at a high rate, and then the rate of increase in density decreases with increasing pres- 
sure. Consequently, it is very difficult to achieve the full density because prohibitive 
pressure is required. 

Frictional forces between the powder and the die wall always oppose the trans- 
mission of the applied pressure in its vicinity. Therefore, the applied pressure diminishes 
with depth in the case of single-ended pressing (i.e., when the compaction pressure is 
applied on only one side). This is accompanied by an uneven density distribution 
throughout the compact. The density always decreases with increasing distance from 
the pressing punch face. Figure 7.4 indicates the variation of pressure with depth along 


A typical plot of the 
relationship between 
achieved density and 
compaction pressure 

Compaction pressure 


7 Powder Metallurgy 


The variation of 
pressure with depth 
along the compact 

the compact as well as the resulting variation in density. It is always recommended that 
the value of the length-to-diameter ratio of the compact be kept lower than 2.0 in order 
to avoid considerable density variations. 

In order to improve pressure transmission and to obtain more even density distri- 
bution, lubricants are either admixed with the powder or applied to the die walls. Other 
techniques are also used to achieve uniform density distribution, such as compacting 
from both ends and suspending the die on springs or withdrawing it to reduce the ef- 
fects of die-wall friction. 

During the pressing of a metal powder in a die, elastic deformation of the die oc- 
curs in radial directions, leading to bulging of the die wall. Meanwhile, the compact 
deforms both elastically and plastically. When the compaction pressure is released, the 
elastic deformation tries to recover. But because some of the compact expansion is due 
to plastic deformation, the die tightly grips the compact, which hinders the die from re- 
turning to its original shape. Accordingly, a definite load, called the ejection load, has 
to be exerted on the compact in order to push it out of the die. Figure 7.5 illustrates the 
sequence of steps in a pressing operation. 


Sintering involves heating the green compact in a controlled-atmosphere furnace to a 
temperature that is slightly below the melting point of the powder metal. When the 
compact is composed of mixed elemental powders (e.g., iron and copper), the sinter- 


Sequence of steps in a 
pressing operation 




v//// / , 

7.2 Powder Metallurgy: The Basic Process 257 

ing temperature will then have to be below the melting point of at least one major con- 
stituent. The sintering operation will result in the following: 

1. Strong bonding between powder particles 

2. Chemical, dimensional, or phase changes 

3. Alloying, in the case of mixed elemental powders 

Such effects of the sintering operation are influenced by process variables such as sin- 
tering temperature, time, and atmosphere. 

The amount, size, shape, and even nature of the pores are changed during sin- 
tering. There are two kinds of porosity: open, or interconnected, porosity (connected 
to the compact surface) and closed, or isolated, porosity. In a green compact, most 
of the porosity is interconnected and is characterized by extremely irregular pores. 
After sintering, interconnected porosity becomes isolated, and pore spheroidization 
takes place because of the surface tension forces. Also, the oxide films covering the 
particle surfaces of a green compact can be reduced by using the appropriate sin- 
tering atmosphere. 

The most important atmospheres used in industrial sintering are carbon monoxide, 
hydrogen, and cracked ammonia. The latter is commonly used and is obtained by cat- 
alytic dissociation of ammonia, which gives a gas consisting of 25 percent nitrogen 
and 75 percent hydrogen by volume. Inert gases like argon and helium are occasion- 
ally used as sintering atmospheres, but cost is a decisive factor in limiting their use. 
Vacuum sintering is also finding some industrial application in recent years; neverthe- 
less, the production rate is the main limitation of this method. 

There are two main types of sintering furnaces: continuous and batch-operated. 
In continuous furnaces, the charge is usually conveyed through the furnace on mesh 
belts. These furnaces are made in the form of tunnels or long tubes having a diame- 
ter of not more than 18 inches (45 cm). Heating elements are arranged to provide 
two heating zones: a relatively low-temperature zone, called a dewaxing zone, in 
which lubricants are removed so that they will not cause harmful reactions in the 
next zone, and a uniform heating zone, which has the required high temperature 
where sintering actually takes place. A third zone of the furnace tube is surrounded 
by cooling coils in order to cool the compacts to ambient temperature in the con- 
trolled atmosphere of the furnace, thus avoiding oxidation of the compacts. Flame 
curtains (burning gases like hydrogen) are provided at both ends of the furnace tube 
to prevent air from entering into the furnace. Figure 7.6 is a sketch of a continuous 
sintering furnace. This type of furnace is suitable for mass production because of its 
low sintering cost per piece and its ability to give more consistent products. When 
small quantities of compacts must be sintered, however, batch-operated furnaces are 
used. These furnaces (e.g., vacuum furnaces) are also more suitable when high-purity 
products are required. 

The sintering time varies with the metal powder and ranges between 30 minutes 
and several hours. However, 40 minutes to an hour is the most commonly used sinter- 
ing time in industry. 


Powder Metallurgy 


A sketch of a continuous sintering furnace 

Temperature , 




Uniform heating zone 






Because of the wide variety of powder metallurgy operations, it may be difficult for 
a person who is not familiar with this process to pursue the proper sequence of op- 
erations. The flowchart in Figure 7.7 is intended to clearly show the relationship be- 
tween the various powder metallurgy operations (which will be discussed later) and 
to give a bird's-eye view of the flow of material to yield the final required product. 
Nevertheless, it must be remembered that there are exceptions and that some opera- 
tions cannot be shown on the flowchart because they would make it overly detailed 
and complicated. 


There are many techniques of consolidating metal powders. They are classified, as 
shown in Figure 7.8, into two main groups: pressureless and pressure forming. The 
pressureless methods are those in which no external pressure is required. This group 
includes loose sintering, slip casting, and slurry casting. The, pressure forming methods 
include conventional compaction, vibratory compaction, powder extrusion, powder 

7.4 Alternative Consolidation Techniques 



A flowchart showing the relationship between the various powder metallurgy operations 

Metal powders 

Mixing and blending 

die pressing 















Finished P/M components 



Classification of the techniques for consolidating metal powders 




Loose sintering 


Conventional Vibratory 
compaction compaction 





CIP Explosive Forming 

compaction with binders 

260 7 Powder Metallurgy 

rolling, hot and cold isostatic pressing, explosive forming, and forming with binders. 
A detailed account of conventional powder metallurgy has been given; following is a 
brief description of these other consolidation techniques. 

Loose Sintering 

Loose sintering is employed in manufacturing filters. It involves sintering of loose 
metal powder in molds made of graphite or ceramic material. The temperature used is 
similar to that of conventional sintering, but the time involved is usually longer (two 
days when manufacturing stainless steel filters). 

Slip Casting 

The application of slip casting is usually limited to the production of large, intricate 
components made from refractory metals and cermets (mixtures of metals and ceram- 
ics). The slip, which is a suspension of fine powder particles in a viscous liquid, is 
poured into an absorbent plaster-of-paris mold. Both solid and hollow articles can be 
produced by this method. When making hollow objects, excess slip is poured out after 
a layer of metal has been formed on the mold surface. 

Slurry Casting 

Slurry casting is very similar to slip casting, except that the mixture takes the form of 
a slurry and binders are usually added. Also, because the slurry contains less water, 
nonabsorbent molds can be used. 

Vibratory Compaction 

Vibratory compaction involves superimposing mechanical vibration on the pressing 
load during the compaction operation. The advantages of this process include the con- 
siderable reduction in the pressure required and the ability to compact brittle particles 
that cannot be pressed by conventional techniques because the high compaction load 
required would result in fragmentation rather than consolidation of the powder parti- 
cles. The main application involves the consolidation of stainless steel and uranium 
oxide powders for nuclear fuel elements. 

Isostatic Pressing 

In isostatic pressing (IP), equal all-around pressure is applied directly to the powder 
mass via a pressurized fluid. Accordingly, die-wall friction is completely eliminated, 
which explains the potential of the process to produce large, dense parts having uni- 
form density distribution. The process can be performed at room temperature (cold iso- 
static pressing) or can be carried out at elevated temperatures (hot isostatic pressing). 
In cold isostatic pressing (CIP), a flexible envelope (usually made of rubber or 
polymers) that has the required shape is filled with the packed powder. The envelope 
is then sealed and placed into a chamber that is, in turn, closed and pressurized to con- 

7.4 Alternative Consolidation Techniques 



The isostatic pressing 




o o o 
o o o 

o o o 
o o o 

o o o 
o o o 

o o o 
o o o 

o o o 
o o o 

o o o 
o o o 

o o o 
o o o 

o o o 
o o o 

solidate the powder. The lack of rigidity of the flexible envelope is countered by using 
a mesh or perforated container as a support (see Figure 7.9). The main disadvantage of 
this process is the low dimensional accuracy due to the flexibility of the mold. 

In hot isostatic pressing (HIP), both isostatic pressing and sintering are combined. 
Powder is canned in order to separate it from the pressurized fluid, which is usually 
argon. The can is then heated in an autoclave, with pressure applied isostatically. Com- 
plete densification and particle bonding occur. The elevated temperature at which the 
powder is consolidated results in a softening of the particles. For this reason, the 
process is used to compact hard-to-work materials such as tool steels, beryllium, 
nickel-base superalloys, and refractory metals. A good example is the manufacture of 
jet-engine turbine blades, where a near-net shape is made from nickel-base superal- 
loys. A main disadvantage of this method is the long processing time. 

Powder Extrusion 

Powder extrusion is a continuous compaction process and can be performed hot or 
cold. It is employed in producing semifinished products having a high length-to- 
diameter ratio, a geometry that makes producing them by conventional powder met- 
allurgy impossible. The conventional technique involves packing metal powder into 
a thin container that is, in turn, evacuated, sealed, and then extruded. An emerging 
technique involves the extrusion of suitable mixtures of metal (or ceramic) powders 
and binders such as dextrin and sugars. It has been successfully employed in the 
production of highly porous materials used as filters or fuel cells in batteries. 

262 7 Powder Metallurgy 

Powder Rolling 

Direct powder rolling, or roll compacting, is another type of continuous compaction 
process. It is employed mainly for producing porous sheets of nonferrous powders like 
copper and nickel. This process involves feeding the metal powder into the gap be- 
tween the two rolls of a simple mill, where it is squeezed and pushed forward to form 
a sheet that is sintered and further rolled to control its density and thickness. 

High-Energy-Rate Compaction 

The various HERF compaction techniques are based on the same principle, which is 
the application of the compaction energy within an extremely short period of time. 
Several methods were developed for compacting metal powders at high speeds. Ex- 
amples are explosives, high-speed presses, and spark sintering. It is believed that ex- 
plosive compaction is suitable only when the size of the compact and the density 
required cannot be achieved by the isostatic compaction process. Nevertheless, the 
danger of handling explosives and the low cycling times impose serious limitations on 
this technique in production. 

The use of high-speed presses like the Dynapak (built by General Dynamics) 
and the Petro-Forge (built by Mechanical Engineering Department, Birmingham 
University, England) for powder compaction is, in practicality, an extension of the 
die-pressing technique. These high-speed presses are particularly advantageous for 
pressing hard-to-compact powders and large components. 

There are also some other powder-consolidation methods that can be classified as 
high-speed techniques. These include electrodynamic pressing, electromagnetic press- 
ing, and spark sintering. Electrodynamic pressing involves utilizing the high pressure 
produced by the sudden discharge of electrical energy to compact powders at high 
speeds. Electromagnetic pressing is based upon the phenomenon that a strong mag- 
netic field is generated when electric current is suddenly discharged through an in- 
ductance. This strong magnetic field is used for pressing a thin-walled metallic tube 
that contains the powder. Spark sintering involves the sudden discharge of electrical 
energy into the powder mass to puncture the oxide films that cover each individual 
powder particle and to build up pure metallic contacts between the particles. After 
about 10 seconds of impulsive discharging, the current is shut off, and a pressure of 
about 14,500 lb/in 2 (100 MPa) is applied to compact the powder to the final required 

Injection Molding 

Although injection molding is an emerging process, it can be considered as a version 
of forming with binders, which is a rather old method. The process involves injection 
molding metal powders that are precoated with a thermoplastic polymer into a part 
similar in shape to the final required component but having larger dimensions. After re- 
moving the polymer by an organic solvent, the porous compact is then sintered for a 
long time in order to allow for volume shrinkage and, consequently, an increase in 
density. The main advantage of this process is that it offers promise in the forming of 
intricate shapes. 

7.5 Secondary Consolidation Operations 263 

Hot Pressing 

Hot pressing is a combination of both the compaction and the sintering operations. It 
is basically similar to the conventional powder metallurgy process, except that pow- 
ders are induction heated during pressing, and, consequently, a protective atmosphere 
is necessary. For most metal powders, the temperatures used are moderate (above re- 
crystallization temperature), and dies made of superalloys are used. The hot pressing 
of refractory metals (e.g., tungsten and beryllium), however, necessitates the use of 
graphite dies. The difficulties encountered in this technique limit its application to lab- 
oratory research. 


In most engineering applications, the physical and mechanical properties of the as- 
sintered compact are adequate enough to make it ready for use. However, secondary 
processing is sometimes required to increase the density and enhance the mechani- 
cal properties of the sintered component, thus making it suitable for heavy-duty en- 
gineering applications. The operations involved are similar to those used in forming 
fully dense metals, though certain precautions are required to account for the porous 
nature of the sintered compacts. Following is a survey of the common secondary 

Coining (Repressing) 

Coining involves the pressing of a previously consolidated and sintered compact in 
order to increase its density. This operation is performed at room temperature, and con- 
siderable pressures are thus required. It is often possible to obtain significant improve- 
ment in strength not only because of the increased densification but also because of the 
work-hardening that occurs during the operation. A further advantage of this process is 
that it can be employed to alter shape and dimensions slightly. Repressing is a special 
case of coining where no shape alteration is required. 

Extrusion, Swaging, or Rolling 

Sintered powder compacts, whether in their cold or hot state, can be subjected to any 
forming operation (extrusion, swaging, or rolling). When processing at elevated tem- 
peratures, either a protective atmosphere or canning of the compacts has to be em- 
ployed. Such techniques are applied to canned sintered compacts of refractory metals, 
beryllium, and composite materials. 

Forging of Powder Preforms 

Repressing and coining of sintered compacts cannot reduce porosity below 5 percent 
of the volume of the compact. Therefore, if porosity is to be completely eliminated, 
hot forging of powder preforms must be employed. Sintered powder compacts hav- 
ing medium densities (80 to 85 percent of the full theoretical density) are heated. 


Powder Metallurgy 

lubricated, and fed into a die cavity. The preform is then forged with a single 
stroke, as opposed to conventional forging of fully dense materials, where several 
blows and manual transfer of a billet through a series of dies are required. This ad- 
vantage is a consequence of using a preform that has a shape quite close to that of 
the final forged product. The tooling used involves a precision flashless closed die; 
therefore, the trimming operation performed after conventional forging is eliminated. 
The forging of powder preforms combines the advantages of both the basic pow- 
der metallurgy and the conventional hot forging processes while eliminating their 
shortcomings. For this reason, the process is extensively used in the automotive in- 
dustry in producing transmission and differential-gear components. Examples of some 
forged powder metallurgy parts are shown in Figure 7.10. 


Many powder metallurgy products are ready for use in their as-sintered state; however, 
finishing processes are frequently used to impart some physical properties or geomet- 
rical characteristics to them. Following are some examples of the finishing operations 
employed in the powder metallurgy industry. 


Sizing is the pressing of a sintered compact at room temperature to secure the desired 
shape and dimensions by correcting distortion and change in dimensions that may have 
occurred during the sintering operation. Consequently, this operation involves only 

FIGURE 7.10 

Some forged powder 
metallurgy parts 
(Courtesy of the Metal 
Powder Industries 
Federation, Princeton, 
New Jersey) 


7.6 Finishing Operations 


FIGURE 7.10 

Some forged powder 
metallurgy parts 
(Courtesy of the Metal 
Powder Industries 
Federation, Princeton, 
New Jersey) 


limited deformation and slight density changes and has almost no effect on the me- 
chanical properties of the sintered compact. 


Features like side holes, slots, or grooves cannot be formed during pressing, and, there- 
fore, either one or two machining operations are required. Because cooling liquids 
can be retained in the pores, sintered components should be machined dry whenever 
possible. An air blast is usually used instead of coolants to remove chips and cool the 

266 7 Powder Metallurgy 

Oil Impregnation 

Oil impregnation serves to provide either protection against corrosion or a degree of 
self-lubrication or both. It is usually carried out by immersing the sintered porous com- 
pact in hot oil and then allowing the oil to cool. Oil impregnation is mainly used in the 
manufacturing of self-lubricating bearings made of bronze or iron. 


Infiltration is permeation of a porous metal skeleton with a molten metal of a lower 
melting point by capillary action. Infiltration is performed in order to fill the pores and 
give two-phase structures with better mechanical properties. The widely used applica- 
tion of this process is the infiltration of porous iron compacts with copper. The process 
is then referred to as copper infiltration and involves placing a green compact of cop- 
per under (or above) the sintered iron compact and heating them to a temperature 
above the melting point of copper. 

Heat Treatment 

Conventional heat treatment operations can be applied to sintered porous materials, 
provided that the inherent porosity is taken into consideration. Pores reduce the ther- 
mal conductivity of the porous parts and thus reduce their rate of cooling. For sintered 
porous steels, this means poorer hardenability. Also, cyanide salts, which are very poi- 
sonous and are used in heat treatment salt baths, are retained in the pores, resulting in 
extreme hazards when using such heat-treated compacts. Therefore, it is not advisable 
to use salt baths for surface treatment of porous materials. 

Steam Oxidizing 

A protective layer of magnetite (Fe 3 4 ) can be achieved by heating the sintered ferrous 
parts and exposing them to superheated steam. This will increase the corrosion resis- 
tance of the powder metallurgy parts, especially if it is followed by oil impregnation. 


Metallic coatings can be satisfactorily electroplated directly onto high-density and 
copper-infiltrated sintered compacts. For relatively low-density compacts, electro- 
plating must be preceded by an operation to seal the pores and render the compacts 
suitable for electroplating. 


The structure of a powder metallurgy part consists of a matrix material with a mi- 
crostructure identical to that of a conventional fully dense metal and pores that are a 
unique and controllable feature of sintered porous materials. For this reason, powder 

7.7 Porosity in Powder Metallurgy Parts 


metallurgy materials are grouped according to their porosity, which is quantitatively 
expressed as the percentage of voids in a part. Those materials having less than 10 per- 
cent porosity are considered to be high density; those with porosity more than 25 per- 
cent, low density. There is a relationship between porosity and density (both being 
expressed as fractions of the full theoretical density), and it can be expressed by the 
following equation: 

porosity = 1 - density (7.6) 

As previously explained, the theoretical density is not that of the fully dense pure 
metal but is the mean value of the densities of all constituents. These include not only 
alloying additives but also impurities. When considering green densities, the effect of 
lubricants must be taken into consideration. 

Pores are classified with respect to their percentage, type, size, shape, and distrib- 
ution. The type can be either interconnected or isolated. The volume of interconnected 
porosity can be determined by measuring the amount of a known liquid needed to sat- 
urate the porous powder metallurgy sample. The interconnected porosity is essential 
for successful oil impregnation and thus is very important for the proper functioning 
of self-lubricating bearings. 

At this stage, it is appropriate to differentiate between the following three techni- 
cal terms used to describe density: 

„ , . mass of compact n -, 

bulk density = — ; j^ V J > 

bulk volume of compact 

mass of compact 

apparent density = ; 

rr apparent volume 


mass of compact 

bulk volume of compact - volume of open pores 
mass of compact 

true density = 

true volume 

mass of compact "™ 

~ bulk volume of compact - (volume of open pores 
+ volume of closed pores) 

For a green compact produced by admixed lubrication, these densities are mis- 
leading and do not indicate the true state of densification due to the presence of lubri- 
cant within the space between metal particles. Therefore, the bulk density must be 
readjusted to give the true metal density (TMD) as follows: 

, , „ , % of metal n im 

TMD = actual bulk density x — I 7 -™) 

268 7 Powder Metallurgy 


The design of a powder metallurgy part and the design of the tooling required to pro- 
duce it cannot be separated. A part design that needs either long, thin tubular punches, 
tooling with sharp corners, or lateral movement of punches cannot be executed. For 
this reason, the design of powder metallurgy parts is often different from that of parts 
produced by machining, casting, or forging, and a component that is being produced 
by these methods has to be redesigned before being considered for manufacture by 
powder metallurgy. Following are various tooling and pressing considerations, some of 
which are illustrated in Figure 7.11. 


Holes in the pressing direction can be produced by using core rods. In this case, there 
is almost no limitation on the general shape of the hole. But side holes and side slots 
are very difficult to achieve during pressing and must be made by secondary machin- 
ing operations (see Figure 7.11a). 

Wall Thickness 

It is not desirable to have a wall thickness less than 1/16 inch (1.6 mm) because the 
punch required to produce the thickness will not be rigid enough to withstand the high 
stresses encountered during the pressing operation. 


It is recommended that sharp corners be avoided whenever possible. Fillets with gen- 
erous radii are desirable, provided that they do not necessitate the use of punches with 
featherlike edges (see Figure 7.11b). 


Tapers are not always required. However, it is desirable to have them on flange-type 
sections and bosses to facilitate the ejection of the green compact. 


As mentioned earlier, it is sometimes not desirable to use radii on part edges. Cham- 
fers are the proper alternative in preventing burrs. 


A small flange, or overhang, can be easily produced. However, for a large overhang, 
ejection without breaking the flange is very difficult (see Figure 7.11c). 

7.8 Design Considerations for Powder Metallurgy Parts 


FIGURE 7.11 

Design considerations 
for powder metallurgy 
parts: (a) holes; 
(b) fillets; (c) flanges; 

(d) bosses; 

(e) undercuts 

Required ^ 



\ / 




Bosses can be made, provided that they are round in shape (or almost round) and that 
the height does not exceed 15 percent of the overall height of the component (see 
Figure 7. lid). 


Undercuts that are perpendicular to the pressing direction cannot be made because they 
prevent ejection of the part. If required, they can be produced by a secondary machin- 
ing operation (see Figure 7.1 le). 

270 7 Powder Metallurgy 


Like any other manufacturing process, powder metallurgy has advantages as well as 
disadvantages. The decision about whether to use this process or not must be based on 
these factors. The advantages of powder metallurgy are as follows: 

1. Components can be produced with good surface finish and close tolerances. 

2. There is usually no need for subsequent machining or finishing operations. 

3. The process offers a high efficiency of material utilization because it virtually 
eliminates scrap loss. 

4. Because all steps of the process are simple and can be automated, only a mini- 
mum of skilled labor is required. 

5. Massive numbers of components with intricate shapes can be produced at high 

6. The possibility exists for producing controlled, unique structures that cannot be 
obtained by any other process. 

The main disadvantages of the process are as follows: 

1. Powders are relatively high in cost compared with solid metals. 

2. Sintering furnaces and special presses, which are more complicated in principle 
and construction than conventional presses, are necessary. 

3. Tooling is very expensive as several punches or die movements are often used. 

4. High initial capital cost is involved, and the process is generally uneconomical 
unless very large numbers of components are to be manufactured. 

5. Powder metallurgy parts have inferior mechanical properties due to porosity (this 
does not apply to forged powder metallurgy parts), and the process is thus primar- 
ily suitable for the production of a large number of small, lightly stressed parts. 


The applications of powder metallurgy parts fall into two main groups. The first group 
consists of those applications in which the part is used as a structural component that 
can also be produced by alternative competing manufacturing methods, powder metal- 
lurgy being used because of the low manufacture cost and high production rate. The 
second group includes those applications in which the part usually has a controlled, 
unique structure and cannot be made by any other manufacturing method. Examples 
are porous bearings, filters, and composite materials. Following is a quick review of 
the various applications. 

7.10 Applications of Powder Metallurgy Parts 


FIGURE 7.12 

Some powder 
metallurgy products 
(Courtesy of the Metal 
Powder Industries 
Federation, Princeton, 
New Jersey) 

Structural Components 

Powder metallurgy used to be limited to the production of small, lightly stressed parts. 
However, with the recent development in forging powder preforms, the process is 
commonly used in producing high-density components with superior mechanical prop- 
erties. Cams, gears, and structural parts of the transmission system are some applica- 
tions of the powder metallurgy process in the automotive, agricultural machinery, and 
domestic appliance industries. Figures 7.12 and 7.13 show some examples of powder 
metallurgy products. 

The structural powder metallurgy components are usually made of iron-base pow- 
ders, with or without additions of carbon, copper, and other alloying elements like 

FIGURE 7.13 

More powder metallurgy 
products (Courtesy of 
the Metal Powder 
Industries Federation, 
Princeton, New Jersey) 

272 7 Powder Metallurgy 

nickel. Prealloyed powders are also employed, although they are less common than the 
mixed elemental powders. 

Self-Lubricating Bearings 

Self-lubricating bearings are usually made by the conventional die-pressing technique, 
in which a porosity level between 20 and 40 percent is achieved. A sizing operation is 
performed for dimensional accuracy and in order to obtain smooth surfaces. The bear- 
ings are oil impregnated either before or after sizing. Bronze powders are used in the 
manufacturing of porous bearings, but iron-base powders are also employed to give 
higher strength and hardness. 


In manufacturing filters, the appropriate metal powder (e.g., bronze) is screened in 
order to obtain uniform particle size. The powder is then poured into a ceramic or 
graphite mold. The mold is put into a sintering furnace at the appropriate sintering 
temperature so that loose sintering can take place. The products must have generous 
tolerances, especially on their outer diameters, where 3 percent is typical. 

Friction Materials 

Clutch liners and brake bands are examples of friction materials. They are best manu- 
factured by powder metallurgy. The composition includes copper as a matrix, with ad- 
ditions of tin, zinc, lead, and iron. Nonmetallic constituents like graphite, silica, emery, 
or asbestos are also added. The mixture is then formed to shape by cold pressing. After 
sintering, some finishing operations like bending, drilling, and cutting are usually re- 
quired. It must be noted that friction materials are always joined to a solid plate, which 
gives adequate support to these weak parts. 

Electrical Contact Materials 

Electrical contact materials include two main kinds: sliding contacts and switching 
contacts. It is not possible to produce any of these contact materials except by powder 
metallurgy as both involve duplex structures. 

Sliding contacts are components of electrical machinery employed when current is 
transferred between sliding parts (e.g., brushes in electric motors). The two main char- 
acteristics needed are a low coefficient of friction and good electrical conductivity. 
Compacts of mixtures of graphite and metal powder can fulfill such conditions. Pow- 
ders of metals having high electrical conductivity, such as brass, copper, or silver, are 
used. These graphite-metal contacts are produced by conventional pressing and sinter- 
ing processes. 

Switching contacts are used in high-power circuit breakers. The three characteris- 
tics needed are good electrical conductivity, resistance to mechanical wear, and less 
tendency of the contact surfaces to weld together. A combination of copper, silver, and 
a refractory metal like tungsten provides the required characteristics. These contacts 

7.10 Applications of Powder Metallurgy Parts 273 

are produced either by conventional pressing and sintering or by infiltrating a porous 
refractory material with molten copper or silver. 


Magnets include soft magnets and permanent magnets. Soft magnets are used in dc 
motors or generators as armatures, as well as in measuring instruments. They are made 
of iron, iron-silicon, and iron-nickel alloys. Electrolytic iron powder is usually used 
because of its high purity and its good compressibility, which allows the high compact 
densities required for maximum permeability to be attained. 

Permanent magnets produced by powder metallurgy have the commonly known 
name Alnico. This alloy consists mainly of nickel (30 percent), aluminum (12 percent), 
and iron (58 percent) and possesses outstanding permanent magnetic properties. Some 
other additives are often used, including cobalt, copper, titanium, and niobium. 


The cores produced by powder metallurgy are used with ac high-frequency inductors 
in wireless communication systems. Such cores must possess high constant perme- 
ability for various frequencies as well as high electrical resistivity. Carbonyl iron pow- 
der is mixed with a binder containing insulators (to insulate the powder particles from 
one another and thus increase electrical resistivity) and then compacted using ex- 
tremely high pressures, followed by sintering. 

Powder Metallurgy Tool Steels 

The production of tool steels by powder metallurgy eliminates the defects encountered 
in conventionally produced tool steels — namely, segregation and uneven distribution 
of carbides. Such defects create problems during tool fabrication and result in shorter 
tool life. The technique used involves compacting prealloyed tool-steel powders by hot 
isostatic pressing to obtain preforms that are further processed by hot working. 


Superalloys are nickel- and cobalt-base alloys, which exhibit high strength at elevated 
temperatures. They are advantageous in manufacturing jet-engine parts like turbine 
blades. The techniques used in consolidating these powders include HIP, hot extrusion, 
and powder metallurgical forging. 

Refractory Metals 

The word refractory means "difficult to fuse." Therefore, metals with high melting 
points are considered to be refractory metals. These basically include four metals: 
tungsten, molybdenum, tantalum, and niobium. Some other metals can also be con- 
sidered to belong to this group. Examples are platinum, zirconium, thorium, and tita- 
nium. Refractory metals, as well as their alloys, are best fabricated by powder 
metallurgy. The technique used usually involves pressing and sintering, followed by 


Powder Metallurgy 

working at high temperatures. The applications are not limited to incandescent lamp 
filaments and heating elements; they also include space technology materials, the 
heavy metal used in radioactive shielding, and cores for armor-piercing projectiles. 
Titanium is gaining an expanding role in the aerospace industry because of its excel- 
lent strength-to-specific-weight ratio and its good fatigue and corrosion resistance. 

Cemented Carbides 

Cemented carbides are typical composite materials that possess the superior properties 
of both constituents. Cemented carbides consist of hard wear-resistant particles of 
tungsten or titanium carbides embedded in a tough strong matrix of cobalt or steel. 
They are mainly used as cutting and forming tools; however, there are other applica- 
tions, including gages, guides, rock drills, and armor-piercing projectiles. They possess 
excellent red hardness and have an extremely long service life as tools. Cemented car- 
bides are manufactured by ball-milling carbides with fine cobalt (or iron) powder, fol- 
lowed by mixing with a lubricant and pressing. The green compact is then presintered 
at a low temperature, machined to the required shape, and sintered at an elevated tem- 
perature. A new dimension in cemented carbides is Ferro-Tic, involving titanium car- 
bide particles embedded in a steel matrix. This material can be heat treated and thus 
can be easily machined or shaped. 

Review Questions 


1. Define each of the following technical terms: 

a. compressibility 

b. compactibility 

c. green density 

d. impregnation 

e. infiltration 

f. flowability 

g. particle-size distribution 

2. List five advantages of the powder metallurgy 

3. List four disadvantages of the powder metal- 
lurgy process. 

4. What are the important characteristics of a 
metal powder? 

5. Describe three methods for producing metal 


Explain briefly the mechanics of pressing. 

Why are lubricants added to metal powders be- 
fore pressing? 

Is it possible to eliminate all voids by conven- 
tional die pressing? Why? 

9. Explain briefly the mechanics of sintering. 

10. Why is it necessary to have controlled atmos- 
pheres for sintering furnaces? 

Explain why it is not possible to use the con- 
ventional pressing techniques as a substitute for 
each of the following operations: isostatic 
pressing, slip casting, HERF compaction. 

Differentiate between the following: coining, 
repressing, sizing. 

How is copper infiltration accomplished and 
what are its advantages? 




Chapter 7 Problems 


14. Can powder metallurgical forging be replaced 
by conventional forging? Why? 

15. How can machining of some powder metal- 
lurgy components be inevitable? 

16. How is plating of powder metallurgy compo- 
nents carried out? 

17. Name five products that can only be produced 
by powder metallurgy. 

18. Why are cemented carbides presintered? 

19. Why is electrolytic iron powder used in manu- 
facturing soft magnets? 

20. Discuss four design limitations in connection 
with powder metallurgy components. 



1. Following are the experimentally determined 
characteristics of three kinds of iron powder: 


Sponge Iron Powder (1) 



Screen Analysis 




Row Rate 

+70 0% 

H 2 loss 


-70 to +100 1% 



2.4 g/cm 3 

35 s/50 g 

-100 to +325 74% 

Si0 2 


-325 25% 



Sponge Iron Powder (II) 


Screen Analysis 




Flow Rate 

+40 2% 

H 2 loss 


-40 to +60 40% 



2.4 g/cm 3 

35 s/50 g 

-70 to +100 30% 

Si0 2 


-100 to +200 20% 



-200 8% 




7 Powder Metallurgy 

Atomized Iron Powder 

Screen Analysis 




Flow Rate 



H 2 loss 






2.9 g/cm 3 

24 s/50 g 









Plot the following for each powder: 

a. Cumulative oversize graph 

b. Cumulative undersize graph 

c. Frequency distribution curve (obtain median 
particle size for the powder) 

d. Histogram of particle-size distribution 

The axis usually indicates the particle size in mi- 
crons and not mesh size. Use the following 

4. Plot a graph indicating the maximum achievable 
green density versus the percentage of admixed 
zinc stearate for atomized iron powder (density 
of zinc stearate is 1 . 1 g/cm ). What can you de- 
duce from the curve? 

5. Which powder in Problem 1 would fill the die 
cavity faster? 













Calculate the full theoretical density for a com- 
pact made of atomized iron powder, knowing 
that the density of carbon equals 2.2 g/cm and 
the density of iron oxide equals 2.9 g/cm . 

Determine the approximate height of the pow- 
der fill for each kind of iron powder given in 
Problem 1 if the green density of the compact is 
6.8 g/cm 3 and its height is 2.1 cm. 

Calculate the maximum achievable green density 
of a mixture of atomized iron powder plus 1 per- 
cent zinc stearate and 10 percent pure copper 
(density of copper is 8.9 g/cm 3 ). 

Chapter 7 Design Project 


7. Following is the relationship between density 
and pressure for atomized iron powder contain- 
ing 1 percent zinc stearate: 

8. A cylindrical compact of atomized iron powder 
plus 1 percent zinc stearate had a green bulk den- 
sity of 7.0 g/cm\ a diameter equal to 2 cm, and 

Green density, g/cm 3 
Pressure, MN/m 2 






If it is required to manufacture a gear wheel hav- 
ing a green density of 6.8 g/cm 3 using a press 
with a capacity of 1 MN, calculate the diameter 
of the largest gear wheel that can be manufac- 
tured. How can you produce a larger gear by 
modifying the design? 

a height equal to 3 cm. After sintering, its bulk 
density increased to 7.05 g/cm 3 . Calculate its 
new dimensions. 

9. The sintered density of atomized iron compact 
containing 10 percent copper was 7.2 g/cm 3 . 
What is the porosity? 

Design Project 


Figure 7.14 shows a part that is currently produced by forging and subsequent ma- 
chining. Because the part is not subjected to high stresses during its actual service con- 
ditions, the producing company is considering the idea of manufacturing it by powder 
metallurgy in order to increase the production rate. Redesign this component so that it 
can be manufactured by the conventional die-pressing technique. 

FIGURE 7.14 

A part to be redesigned 
for production by 
powder metallurgy 

Chapter 8 


Plastics, which are more correctly called polymers, are products of macromo- 
lecular chemistry. In fact, the term polymer is composed of the two Greek 
words poly and meres, which mean "many parts." This is, indeed, an accurate 
description of the molecule of a polymer, which is made up of a number of iden- 
tical smaller molecules that are repeatedly linked together to form a long chain. 
As an example, consider the commonly used polymer polyethylene, which is 
composed of many ethylene molecules (C 2 H 4 ) that are joined together, as 
shown in Figure 8.1. These repeated molecules are always organic compounds, 
and, therefore, carbon usually forms the backbone of the chain. The organic 
compound whose molecules are linked together (like ethylene) is referred to as 
the monomer. 

Now, let us examine why the molecules of a monomer tend to link together. 
We know from chemistry that carbon has a valence of 4. Therefore, each car- 
bon atom in an ethylene molecule has an unsaturated valence bond. Conse- 
quently, if two ethylene molecules attach, each to one side of a third molecule, 
the valence bonds on the two carbon atoms of the center molecule will be sat- 
isfied (see Figure 8.1). In other words, the molecules of the monomer tend to 
attach to one another to satisfy the valence requirement of the carbon atoms. 

The molecules of a monomer in a chain are strongly bonded together. Nev- 
ertheless, the long chains forming the polymer molecules tend to be more or 
less amorphous and are held together by weaker secondary forces that are 
known as the van der Waals forces (named after the Dutch physicist). There- 


8.1 Classification of Polymers 



The molecular chain of 

H H 

H , H 


r __ 

H I H 

H i H H 

1 I 

H H 

H H 

fore, polymers are generally not as strong as metals or ceramics. It is also ob- 
vious that properties of a polymer such as the strength, elasticity, and relax- 
ation are dependent mainly upon the shape and size of the long chainlike 
molecules, as well as upon the mutual interaction between them. 

A common, but not accurate, meaning of the term polymer involves syn- 
thetic organic materials that are capable of being molded. Actually, polymers 
form the building blocks of animal life; proteins, resins, shellac, and natural 
rubber are some examples of natural polymers that have been in use for a long 
time. On the other hand, the synthetic, or manufactured, polymers have come 
into existence fairly recently. The first synthetic polymer, cellulose nitrate (cel- 
luloid), was prepared in 1869. It was followed in 1909 by the phenolics, which 
were used as insulating materials in light switches. The evolution of new poly- 
mers was accelerated during World War II due to the scarcity of natural materi- 
als. Today, there are thousands of polymers that find application in all aspects 
of our lives. 


There are, generally, two methods for classifying polymers. The first method involves 
grouping all polymers based on their elevated-temperature characteristics, which actu- 
ally dictate the manufacturing method to be used. The second method of classification 
groups polymers into chemical families, each of which has the same monomer. As an 
example, the ethenic family is based on ethylene as the monomer, and different poly- 
mers (members of this family such as polyvinyl alcohol or polystyrene) can be made 
by changing substituent groups on the basic monomer, as shown in Figure 8.2. As we 
will see later, this enables us to study most polymeric materials by covering just a lim- 
ited number of families instead of considering thousands of polymers individually. But 
before reviewing the commonly used chemical families of polymers, let us discuss in 
depth their elevated-temperature behavior. Based on this behavior, polymers can be 
split into two groups: thermoplastics and thermosets. 


8 Plastics 


Structural formula of some polymers of the ethenic group 


H H 

Polyvinyl chloride 



— C 

-c — 






H CH, 

H H 



Thermoplastics generally have linear structures, meaning that their molecules look 
like linear chains having little breadth but significant length. This structure, as shown 
in Figure 8.3, is analogous to a bowl of spaghetti. Bonds between the various molecu- 
lar chains are mainly of the van der Waals type (i.e., secondary forces). Therefore, this 
type of polymer softens by heating and can then flow viscously to take a desired shape 
because elevated temperatures tend to decrease the intermolecular coherence of the lin- 
ear chains. When the solidified polymers are reheated and melted again, they can be 
given a different shape. This characteristic enables plastics fabricators to recycle ther- 
moplastic scrap, thus increasing the efficiency of raw-material utilization. 

Usually, a thermoplastic polymer consists of a mixture of molecular chains having 
different lengths. Therefore, each structure has a different melting point, and, conse- 
quently, the whole polymer melts, not at a definite temperature, but within a range 
whose limits are referred to as the softening point and the flow point. It has been ob- 
served that when a thermoplastic is given a shape at a temperature between the soft- 


The molecular chains of 
a thermoplastic 


The molecular chains of 
a thermosetting 

8.1 Classification of Polymers 281 

ening and the flow points, the intermolecular tension is retained after the thermoplas- 
tic cools down. Therefore, if the part is reheated to a temperature above the softening 
point, it will return to its original shape because of this intermolecular tension. This 
phenomenon, which characterizes most thermoplastic polymers, is known as shaping 

Many thermoplastic polymers are soluble in various solvents. Consequently, any 
one of these polymers can be given any desired shape by dissolving it into an appro- 
priate solvent and then casting the viscous solution in molds. When the solvent com- 
pletely evaporates, it leaves the rigid resin with the desired shape. 

Several chemical families of polymeric materials can be categorized as thermo- 
plastic. These include the ethenics, the polyamides, the cellulosics, the acetals, and the 
polycarbonates. Their characteristics, methods of manufacture, and applications are 
discussed in detail later in this chapter. 


The molecules of thermosets usually take the form of a three-dimensional network 
structure that is mostly cross-linked, as shown in Figure 8.4. When raw (uncured) 
thermosetting polymers are heated to elevated temperatures, they are set, cross- 
linked, or polymerized. If reheated after this curing operation, thermosets will not 
melt again but will char or burn. Therefore, for producing complex shapes of ther- 
mosetting polymers, powders (or grains) of the polymers are subjected to heat and 
pressure until they are cured as finished products. Such polymers are referred to as 
heat-convertible resins. 

Some raw thermosets can take the form of liquids at room temperature. When re- 
quired, they are converted into solids by curing as a result of heating and/or additives 
(hardeners). This characteristic enables fabricators to produce parts by casting mix- 
tures of liquid polymers and hardeners into molds. Therefore, these polymers are re- 
ferred to as casting resins. 

The cured thermosets are insoluble in solvents and do not soften at high tempera- 
tures. Thus, products made of thermosets can retain their shape under combined load 
and high temperatures, conditions that thermoplastics cannot withstand. 

282 8 Plastics 


Properties of plastics differ significantly from those of metals, and they play a very im- 
portant role in determining the form of the product. In other words, the form is dictated 
not only by the function but also by the properties of the material used and the method 
of manufacture, as we will see later. Following is a discussion of the effect of the prop- 
erties characterizing plastics on the design of plastic products. 

Mechanical Properties 

The mechanical properties of polymers are significantly inferior to those of metals. 
Strength and rigidity values for plastics are very low compared with the lowest values 
of these properties for metals. Therefore, larger sections must be provided for plastic 
products if they are to have a similar strength and/or rigidity as metal products. Un- 
fortunately, these properties get even worse when plastic parts are heated above mod- 
erate temperatures. In addition, some plastics are extremely brittle and notch-sensitive. 
Accordingly, any stress raisers like sharp edges or threads must be avoided in such 

A further undesirable characteristic of plastics is that they tend to deform contin- 
ually under mechanical load even at room temperature. This phenomenon is acceler- 
ated at higher temperatures. Consequently, structural components made of plastics 
should be designed based on their creep strength rather than on their yield strength. 
This dictates a temperature range in which only a plastic product can be used. It is ob- 
vious that such a range is dependent principally upon the kind of polymer employed. 

In spite of these limitations, the strength-to-weight ratio as well as the stiffness-to- 
weight ratio of plastics can generally meet the requirements for many engineering ap- 
plications. In fact, the stiffness-to-weight ratio of reinforced polymers is comparable to 
that of metals like steel or aluminum. 

Physical Properties 

Three main physical properties detrimentally affect the widespread industrial applica- 
tion of polymers and are not shared by metals. First, plastics usually have a very high 
coefficient of thermal expansion, which is about ten times that of steel. This has to be 
taken into consideration when designing products involving a combination of plastics 
and metals. If plastics are tightly fastened to metals, severe distortion will occur when- 
ever a significant temperature rise takes place. Second, some plastics are inflammable 
(i.e., not self-extinguishing) and keep burning even after the removal of the heat 
source. Third, some plastics have the ability to absorb large amounts of moisture from 
the surrounding atmosphere. This moisture absorption is unfortunately accompanied 
by a change in the size of the plastic part. Nylons are a typical example of this kind of 

8.3 Polymeric Systems 



This section surveys the commonly used polymeric materials and discusses their man- 
ufacturing properties and applications. Also discussed are the different additives that 
are used to impart certain properties to the various polymers. 

Commonly Used Polymers 

Following are some polymeric materials that are grouped into chemical families ac- 
cording to their common monomer. 

Ethenic group. The monomer is ethylene. This group includes the following 

1. Polyethylene. 



— c — 

— c — 



The properties of polyethylene depend upon factors like degree of crystallinity, 
density, molecular weight, and molecular weight distribution. This thermoplastic 
polymer is characterized by its chemical resistance to solvents, acids, and alkalies, as 
well as by its toughness and good wear properties. Polyethylenes also have the ad- 
vantage of being adaptable to many processing techniques, such as injection mold- 
ing, blow molding, pipe extrusion, wire and cable extrusion, and rotational molding. 

The applications of polyethylene are dependent upon the properties, which, in 
turn, depend upon the density and molecular weight. Low-density polyethylene is 
used in manufacturing films, coatings, trash bags, and throwaway products. High- 
density polyethylene is used for making injection-molded parts, tubes, sheets, and 
tanks that are used for keeping chemicals. The applications of the ultrahigh molec- 
ular weight (UHMW) polyethylene include wear plates and guide rails for filling 
and packaging equipment. 

2. Polypropylene. 

H H 

H Chi, 


8 Plastics 

Polypropylene is a thermoplastic material. A molecule of this polymer has all 
substituent groups (i.e., CH 3 ) on only one of its sides. This promotes crystallinity 
and, therefore, leads to strength higher than that of polyethylene. The resistance of 
polypropylene to chemicals is also good. 

Polypropylene is mainly used for making consumer goods that are subjected to 
loads during their service life, such as ropes, bottles, and parts of appliances. This 
polymer is also used in tanks and conduits because of its superior resistance to 

3. Polybutylene. 

H CH 3 

C C- 

H CHo 

Polybutylene is a thermoplastic polymer that has high tear, impact, and creep 
resistances. It also possesses good wear properties and is not affected by chemicals. 
Polybutylene resins are available in many grades, giving a wide range of properties 
and, therefore, applications. 

The properties of polybutylene have made it an appropriate material for piping 
applications. These pipes can be joined together by heat fusion welding or by me- 
chanical compression. Some grades of polybutylene are used as high-performance 
films for food packaging and industrial sheeting. 

4. Polyvinyl chloride. 

H H 


Polyvinyl chloride (PVC) is a thermoplastic polymer that can be processed by 
a variety of techniques like injection molding, extrusion, blow molding, and com- 
pression molding. It is fairly weak and extremely notch-sensitive but has excellent 
resistance to chemicals. When plasticized (i.e., additives are used to lubricate the 
molecules), it is capable of withstanding large strains. 

The applications of rigid PVC include low-cost piping, siding, and related pro- 
files, toys, dinnerware, and credit cards. Plasticized PVC is used in upholstery, im- 
itation leather for seat covers and rainwear, and as insulating coatings on wires. 

8.3 Polymeric Systems 


5. Polyvinyleidene chloride. 

CH 2 C - 


Polyvinyleidene chloride (PVDC) is nonpermeable to moisture and oxygen. It 
also possesses good creep properties. It is a preferred food-packaging material 
(e.g., saran wrap). Rigid grades are used for hot piping. 

6. Polystyrene. 


This thermoplastic polymer is known as "the cheap plastic." It has poor me- 
chanical properties, can tolerate very little deflection, and breaks easily. Because of 
its cost, polystyrene is used for cheap toys and throwaway articles. It is also made 
in the form of foam (Styrofoam) for sound attenuation and thermal insulation. 

7. Polymethyl methacrylate (Plexiglas acrylics). 

CH 2 C 


O OCH 3 


8 Plastics 

This polymer has reasonably good toughness, good stiffness, and exceptional 
resistance to weather. In addition, it is very clear and has a white-light transmission 
equal to that of clear glass. Consequently, this polymer finds application in safety 
glazing and in the manufacture of guard and safety glasses. It is also used in mak- 
ing automotive and industrial lighting lenses. Some grades are used as coatings and 
lacquers on decorative parts. 

8. Fluorocarbons like polytetrafluoroethylene (Teflon). 

F F 

F F 

Teflon is characterized by its very low coefficient of friction and by the fact 
that even sticky substances cannot adhere to it easily. It is also the most chemi- 
cally inert polymer. Nevertheless, it has some disadvantages, including low 
strength and poor processability. Because of its low coefficient of friction, Teflon 
is commonly used as a dry film lubricant. It is also used as lining for chemical 
and food-processing containers and conduits. 

Polycarbonate group. These are actually polyesters. They are thermoplastic and have 
linear molecular chains. Polycarbonate exhibits good toughness, good creep resistance, 
and low moisture absorption. It also has good chemical resistance. It is widely used in 
automotive and medical and food packaging because of its cost effectiveness. It is also 
considered to be a high-performance polymer and has found application in the form of 
solar collectors, helmets, and face shields. 

Polyacetal group. Included in this group is formaldehyde, with ending groups. 



Formaldehyde is a thermoplastic polymer that can be easily processed by injection 
molding and extrusion. It has a tendency to be highly crystalline, and, as a result, this 
polymer possesses good mechanical properties. It also has good wear properties and a 
good resistance to moisture absorption. 

Its applications involve parts that were made of nonferrous metals (like zinc, 
brass, or aluminum) by casting or stamping. These applications are exemplified by 
shower heads, shower-mixing valves, handles, good-quality toys, and lawn sprinklers. 

8.3 Polymeric Systems 


Cellulosic group. The monomer is cellulose. 



Cellulose itself is not a thermoplastic polymer. It can be produced by the viscous 
regeneration process to take the form of a fiber as in rayon, or a thin film, as in cello- 
phane. Cellophane applications involve mainly decoration. Nevertheless, cellulose can 
be chemically modified to produce the following thermoplastics: 

1. Cellulose nitrate. Good dimensional stability and low water absorption are the posi- 
tive characteristics of this polymer. The major disadvantage that limits its widespread 
use is its inflammability. Cellulose nitrate is used in making table-tennis balls, fash- 
ion accessories, and decorative articles. It is also used as a base for lacquer paints. 

2. Cellulose acetate. This polymer has good optical clarity, good dimensional stabil- 
ity, and resistance to moisture absorption. The uses of cellulose acetate include 
transparent sheets and films for graphic art, visual aids, and a base for photographic 
films. It is also used in making domestic articles. 

3. Cellulose acetate butyrate. This thermoplastic polymer is tough, has good surface 
quality and color stability, and can readily be vacuum formed. It finds popular use 
in laminating with thin aluminum foil. 

4. Cellulose acetate propionate. This thermoplastic polymer has reasonably good 
mechanical properties and can be injection molded or vacuum formed. It is used for 
blister packages, lighting fixtures, brush handles, and other domestic articles. 

Polyamide group. This family includes high-performance melt-processable thermo- 


R is a chemical group that differs for different members of this family. 

One group of common polyamides is the nylons. These are characterized by their 
endurance and retention of their good mechanical properties even at relatively high 
temperatures. They also possess good lubricity and resistance to wear. The chief limi- 
tation is their tendency to absorb moisture and change size. 


8 Plastics 

These polymers find use in virtually every market (e.g., automotive, electrical, 
wire, packaging, and appliances). Typical applications include structural components 
up to 10 pounds (4 kg), bushings, gears, cams, and the like. 

ABS. The three monomers are acrylonitrile, butadiene, and styrene. Based on this 
three-monomer system (similar to an alloy in the case of metals), the properties of this 
group vary depending upon the components. Fifteen different types are commercially 
used. They possess both good mechanical properties and processability. Applications 
of the ABS group include pipes and fittings, appliances and automotive uses, tele- 
phones, and components for the electronics industry. 

Polyesters. These polymers result from a condensation reaction of an acid and an al- 
cohol. The type and nature of the polymer obtained depend upon the acid and alcohol 
used. This multitude of polymers are mostly thermoplastic and can be injection molded 
and formed into films and fibers. Their uses include bases for coatings and paints, 
ropes, fabrics, outdoor applications, construction, appliances, and electrical and elec- 
tronic components. Polyester is also used as a matrix resin for fiberglass to yield the 
composite fiber-reinforced polymer. 

Phenolic group. The monomer is phenol formaldehyde. 

As previously mentioned, phenolics are actually the oldest manufactured ther- 
mosetting polymers. They are processed by compression molding, where a product 
with a highly cross-linked chain structure is finally obtained. Phenolics are character- 
ized by their high strength and their ability to tolerate temperatures far higher than 
their molding temperature. 

Phenolics are recommended for use in hostile environments that cannot be toler- 
ated by other polymers. They are used in electrical switchplates, electrical boxes, and 
similar applications. Nevertheless, the chief field of application is as bonding agents 
for laminates, plywood-grinding wheels, and friction materials for brake lining. 

Polyimides. Polyimides are mostly thermosetting and have very complex structures. 
They are considered to be one of the most heat-resisting polymers. They do not melt 

8.3 Polymeric Systems 289 

and flow at elevated temperatures and are, therefore, manufactured by powder metal- 
lurgy techniques. 

The polyimides are good substitutes for ceramics. Applications include jet-engine 
and turbine parts, gears, coil bobbins, cages for ball bearings, bushings and bearings, 
and parts that require good electrical and thermal insulation. 

Epoxies. Epoxies and epoxy resins are a group of polymers that become highly cross- 
linked by reaction with curing agents or hardeners. These polymers have low molecu- 
lar weight and got their name from the epoxide group at the ends of the molecular 
chains. Epoxy resins are thermosetting and have good temperature resistance. They ad- 
here very well to a variety of substrates. Another beneficial characteristic is their sta- 
bility of dimensions upon curing. 

The common application of epoxy resins is as adhesives. With the addition of 
fibers and reinforcements, laminates and fiber-reinforced epoxy resins can be obtained 
and are used for structural applications. 

Polyurethanes. Polyurethanes involve a wide spectrum of polymers ranging from 
soft thermoplastic elastomers to rigid thermosetting foams. While all polyurethanes are 
products of a chemical reaction of an isocyanate and an alcohol, different polymers are 
apparently obtained by different reacting materials. 

Elastomers are used as die springs, forming-die pads, and elastomer-covered rolls. 
Some of these elastomers are castable at room temperature and find popular applica- 
tion in rubber dies for the forming of sheet metals. Flexible foam has actually replaced 
latex rubber in home and auto seating and interior padding. The rigid thermosetting 
foam is used as a good insulating material and for structural parts. Other applications 
of polyurethanes include coating, varnishes, and the like. 

Silicones. In this group of polymers, silicon forms the backbone of the molecular 
chain and plays the same role as that of carbon in other polymers. 

Silicones can be oils, elastomers, thermoplastics, or thermosets, depending upon 
the molecular weight and the functional group. Nevertheless, they are all characterized 
by their ability to withstand elevated temperatures and their water-repellent properties. 

Silicones in all forms are mainly used for high-temperature applications. These in- 
clude binders for high-temperature paints and oven and good-handling tubing gaskets. 
Silicone oils are used as high-temperature lubricants, mold release agents, and damp- 
ing or dielectric fluids. 

Elastomers. These polymeric materials possess a percentage elongation of greater 
than 100 percent together with significantly high resilience. This rubberlike behavior 
is attributed to the branching of the molecular chains. Elastomers mainly include five 

290 8 Plastics 

polymers: natural rubber, neoprene, silicone rubber, polyurethane, and fiuoroelas- 
tomers. Natural rubber is extracted as thick, milky liquid from a tropical tree. Next, 
moisture is removed, additives (coloring, curing agents, and fillers) are blended with it, 
and the mixture is then vulcanized. The latter operation involves heating up to a tem- 
perature of 300°F (150°C) to start cross-linking and branching reactions. 

The application of elastomers includes seals, gaskets, oil rings, and parts that pos- 
sess rubberlike behavior such as tires, automotive and aircraft parts, and parts in form- 
ing dies. 


Additives are materials that are compounded with polymers in order to impart and/or 
enhance certain physical, chemical, manufacturing, or mechanical properties. They are 
also sometimes added just for the sake of reducing the cost of products. Commonly 
used additives include fillers, plasticizers, lubricants, colorants, antioxidants, and sta- 

Fillers. Fillers involve wood flour, talc, calcium carbonate, silica, mica flour, cloth, 
and short fibers of glass or asbestos. Fillers have recently gained widespread industrial 
use as a result of the continued price increase and short supply of resin stocks. An ex- 
pensive or unavailable polymer can sometimes be substituted by another filled poly- 
mer, provided that an appropriate filler material is chosen. 

The addition of inorganic fillers usually tends to increase the strength because this 
kind of additive inhibits the mobility of the polymers' molecular chains. Nevertheless, 
if too much filler is added, it may create enclaves or weak spots and cause problems 
during processing, especially if injection molding is employed. 

Plasticizers. Plasticizers are organic chemicals (high-boiling-temperature solvents) 
that are admixed with polymers in order to enhance resilience and flexibility. This is a re- 
sult of facilitating the mobility of the molecular chains, thus enabling them to move eas- 
ily relative to one another. On the other hand, plasticizers reduce the strength. Therefore, 
a polymer that meets requirements without the addition of plasticizers is the one to use. 

Lubricants. Lubricants are chemical substances that are added in small quantities to 
the polymer to improve processing and flowability. They include fatty acids, fatty al- 
cohols, fatty esters, metallic stearates, paraffin wax, and silicones. Lubricants are clas- 
sified as external (applied externally to the polymer), internal, or internal-external. The 
last group includes most commercially used lubricants. 

Colorants. Colorants may be either dyes or pigments. Dyes have smaller molecules 
and are transparent when dissolved. Pigment particles are relatively large (over 1 |im) 
and are, therefore, either translucent or opaque. Pigments are more widely used than 
dyes because dyes tend to extrude from the polymers. 

Antioxidants. The use of antioxidants is aimed at enhancing the resistance to oxida- 
tion and degradation of polymers, thus extending their useful temperature range and 
service life. These substances retard the chemical reactions that are caused by the pres- 
ence of oxygen. 

8.4 Processing of Plastics 291 

Stabilizers. Stabilizers are substances that are added to polymers to prevent degrada- 
tion as a result of heat or ultraviolet rays. The mechanism of inhibiting degradation of 
polymers differs for different stabilizers. However, ultraviolet stabilizers usually func- 
tion by absorbing ultraviolet radiation. 


A variety of processing methods can be employed in manufacturing plastic products. 
However, it must be kept in mind that no single processing method can successfully be 
employed in shaping all kinds of plastics. Each process has its own set of advantages 
and disadvantages that influence product design. Following is a survey of the common 
methods for plastic processing. 


Casting is a fairly simple process that requires no external force or pressure. It is usually 
performed at room temperature and involves filling the mold cavity with monomers or 
partially polymerized syrups and then heating to cure. After amorphous solidification, 
the material becomes isotropic, with uniform properties in all directions. Nevertheless, 
a high degree of shrinkage is experienced during solidification and must be taken into 
consideration when designing the mold. Sheets, rods, and tubes can be manufactured by 
casting, although the typical application is in trial jigs and fixtures as well as in insulat- 
ing electrical components. Acrylics, epoxies, polyesters, polypropylene, nylon, and 
PVC can be processed by casting. The casting method employed is sometimes modified 
to suit the kind of polymer to be processed. Whereas nylon is cast in its hot state after 
adding a suitable catalyst, PVC film is produced by solution casting. This process in- 
volves dissolving the PVC into an appropriate solvent, pouring the solution on a sub- 
strate, and allowing the solvent to evaporate in order to finally obtain the required film. 

Blow Molding 

Blow molding is a fast, efficient method for producing hollow containers of thermo- 
plastic polymers. The hollow products manufactured by this method usually have thin 
walls and range in shape and size from small, fancy bottles to automobile fuel tanks. 
Although there are different versions of the blow molding process, they basically 
involve blowing a tubular shape (parison) of heated polymer in a cavity of a split mold. 
As can be seen in Figure 8.5, air is injected through a needle into the parison, which 
expands in a fairly uniform thickness and finally conforms to the shape of the cavity. 

Injection Molding 

Injection molding is the most commonly used method for mass production of plastic 
articles because of its high production rates and the good control over the dimensions 
of the products. The process is used for producing thermoplastic articles, but it can also 
be applied to thermosets. The main limitation of injection molding is the required high 


8 Plastics 


The blow molding 




View normal to 
the separation line 

initial capital cost, which is due to the expensive machines and molds employed in the 

The process basically involves heating the polymer, which is fed from a hopper in 
granular pellet or powdered forms, to a viscous melted state and then forcing it into a 
split-mold cavity, where it hardens under pressure. Next, the mold is opened, and the 
product is ejected by a special mechanism. Molds are usually made of tool steel and 
may have more than a single cavity. 

Figure 8.6 shows a modern screw-preplasticator injection unit employed in injection 
molding of thermoplastics. As can be seen, the diverter valve allows the viscous polymer 
to flow either from the plasticating screw to the pressure cylinder or from the cylinder to 
the cooled mold. When thermosets are to be injection molded, a machine with a differ- 
ent design has to be used. Also, the molds must be hot so that the polymer can cure. 

Once the decision has been made to manufacture a plastic product by injection 
molding, the product designer should make a design that facilitates and favors this 
process. Following are some design considerations and guidelines. 

Make the thickness of a product uniform and as small as possible. Injection mold- 
ing of thermoplastics produces net-shaped parts by going from a liquid state to a solid 
state. (These net-shaped parts are used as manufactured; they do not require further 
processing or machining.) This requires time to allow the heat to dissipate so that the 


The injection molding 




8.4 Processing of Plastics 


polymer melt can solidify. The thicker the walls of a product, the longer the product 
cycle, and the higher its cost. Consequently, a designer has to keep the thickness of a 
product to a minimum without jeopardizing the strength and stiffness considerations. 
Also, thickness must always be kept uniform; if change in thickness is unavoidable, it 
should be made gradually. It is better to use ribs rather than increase the wall thickness 
of a product. Figure 8.7 shows examples of poor design and how they can be modified 
(by slight changes in constructional details) so that sound parts are produced. 

Provide generous fillet radii. Plastics are generally notch-sensitive. The designer 
should, therefore, avoid sharp corners for fillets and provide generous radii instead. 
The ratio of the fillet radius to the thickness should be at least 1 .4. 

Ensure that holes will not require complex tooling. Holes are produced by using core 
pins. It is, therefore, clear that through holes are easier to make than blind holes. Also, 
when blind holes are normal to the flow, they require retractable core pins or split 
tools, thus increasing the production cost. 


Examples of poor and 
good designs of walls 
of plastic products 

V &77A WZZ& 




Improved design 

V ////////////A 






8 Plastics 


Examples of poor and 
good designs of bosses 
in injection-molded 




Poor design 

Good design 



Through holes are better than blind holes 

Provide appropriate draft. As is the case with forging, it is important to provide a 
draft of 1 ° so that the product can be injected from the mold. 

Avoid heavy sections when designing bosses. Heavy sections around bosses lead to 
wrappage and dimensional control problems. Figure 8.8 shows poor and good designs 
of bosses. 

Compression Molding 

Compression molding is used mainly for processing thermosetting polymers. The 
process involves enclosing a premeasured charge of polymer within a closed mold and 
then subjecting that charge to combined heat and pressure until it takes the shape of the 
mold cavity and cures. Figure 8.9 shows a part being produced by this process. 

Although the cycle time for compression molding is very long when compared 
with that for injection molding, the process has several advantages. These include low 
capital cost (because the tooling and the equipment used are simpler and cheaper) and 
the elimination of the need for sprues or runners, thus reducing the material waste. There 

8.4 Processing of Plastics 



The compression moldin 



are, however, limitations upon the shape and size of the products manufactured by this 
method. It is generally difficult to produce complex shapes or large parts as a result of 
the poor flowability and long curing times of the thermosetting polymers. 

Transfer Molding 

Transfer molding is a modified version of the compression molding process, and it is 
aimed at increasing the productivity by accelerating the production rate. As can be 
seen in Figure 8.10, the process involves placing the charge in an open, separate "pot," 
where the thermosetting polymer is heated and forced through sprues and runners to 
fill several closed cavities. The surfaces of the sprues, runners, and cavities are kept at 
a temperature of 280 to 300°F (140 to 200°C) to promote curing of the polymer. Next, 
the entire shot (i.e., sprues, runners, product, and the excess polymer in the pot) is 

Rotational Molding 

Rotational molding is a process by which hollow objects can be manufactured from 
thermoplastics and sometimes thermosets. It is based upon placing a charge of solid or 
liquid polymer in a mold. The mold is heated while being rotated simultaneously 
around two perpendicular axes. As a result, the centrifugal force pushes the polymer 
against the walls of the mold, thus forming a homogeneous layer of uniform thickness 

FIGURE 8.10 

The transfer molding 





8 Plastics 

FIGURE 8.11 

The extrusion process 

Changeable die 

Extruded section 



that conforms to the shape of the mold, which is then cooled before the product is 
ejected. The process, which has a relatively long cycle time, has the advantage of of- 
fering almost unlimited product design freedom. Complex parts can be molded by em- 
ploying low-cost machinery and tooling. 


In extrusion, a thermoplastic polymer in powdered or granular form is fed from a hop- 
per into a heated barrel, where the polymer melts and is then extruded out of a die. Fig- 
ure 8.11 shows that plastics extrusion is a continuous process capable of forming an 
endless product that has to be cooled by spraying water and then cut to the desired 
lengths. The process is employed to produce a wide variety of structural shapes, such 
as profiles, channels, sheets, pipes, bars, angles, films, and fibers. Extrusions like bars, 
sheets, and pipes can also be further processed by other plastic manufacturing methods 
until the desired final product is obtained. 

A modification of conventional extrusion is a process known as coextrusion. It in- 
volves extruding two or more different polymers simultaneously in such a manner that 
one polymer flows over and adheres to the other polymer. This process is used in in- 
dustry to obtain combinations of polymers, each contributing some desired property. 
Examples of coextrusion include refrigerator liners, foamed-core solid-sheath tele- 
phone wires, and profiles involving both dense material and foam, which are usually 
used as gasketing in automotive and appliance applications. 


Thermoforming involves a variety of processes that are employed to manufacture cup- 
like products from thermoplastic sheets by a sequence of heating, forming, cooling, 
and trimming. First, the sheet is clamped all around and heated to the appropriate tem- 
perature by electric heaters located above it. Next, the sheet is stretched under the ac- 
tion of pressure, vacuum, or male tooling and is forced to take the shape of a mold. 
The polymer is then cooled to retain the shape. This is followed by removing the part 
from the mold and trimming the web surrounding it. Figure 8.12a through d illustrates 
the different thermoforming processes. 

Although thermoforming was originally developed for the low-volume production 
of containers, the process can be automated and made suitable for high-volume appli- 
cations. In this case, molds are usually made of aluminum because of its high thermal 

8.4 Processing of Plastics 


FIGURE 8.12 

processes: (a) straight 
vacuum forming; 

(b) drape forming; 

(c) matched-mold 
forming; (d) vacuum 




-■• » f i "»-» in »» -' 

v ^---CLOngina 


(Upper half 
of mold) 

Lower half 
of mold) 

atmosphere ~| 

Vent for 


entrapped air 





[1) First stage 






(2) Second stage 

conductivity. For low-volume or trial production, molds are made of wood or even 
plaster of paris. 

Examples of the parts produced by thermoforming include containers, panels, 
housings, machine guards, and the like. The only limitation on the shape of the prod- 
uct is that it should not contain holes. If holes are absolutely required, they should be 
made by machining at a later stage. 


Calendering is the process employed in manufacturing thermoplastic sheets and films. 
This process is similar to rolling with a four-high rolling mill, except that the rolls that 
squeeze the polymer are heated. The thermoplastic sheet is fed and metered in the first 
and second roll gaps, whereas the third roll gap is devoted to gaging and finishing. 

298 8 Plastics 

FIGURE 8.13 

The calendering 

Most of the calendering products are flexible or rubberlike sheets and films, although 
the process is sometimes applied to ABS and polyethylene. Figure 8.13 illustrates the 
calendering process. 

Machining of Plastics 

In some cases, thermoplastic and thermosetting polymers are subjected to machining 
operations like sawing, drilling, or turning. Some configurations and small lot sizes can 
be more economically achieved by machining than by any other plastic-molding 
method. Nevertheless, there are several problems associated with the machining of plas- 
tics. For instance, each type of plastic has its own unique machining characteristics, and 
they are very different from those of the conventional metallic materials. A further prob- 
lem is the excessive tool wear experienced when machining plastics, which results in the 
interruption of production as well as additional tooling cost. Although much research is 
needed to provide solutions for these problems, there are some general guidelines: 

1. Reduce friction at the tool-workpiece interface by using tools with honed or pol- 
ished surfaces. 

2. Select tool geometry so as to generate continuous-type chips. Recent research has 
revealed that there exists a critical rake angle (see Chapter 9) that depends upon 
the polymer, depth of cut, and cutting speed. 

3. Use twist drills that have wide, polished flutes, low helix angles, and tool-point 
angles of about 70° and 120°. 

Recently, lasers have been employed in cutting plastics. Because a laser acts as a ma- 
terials eliminator, its logical application is cutting and hole drilling. High-pressure water 
jets also currently find some application in the cutting of polymers and composites. 

Welding of Plastics 

There are several ways for assembling plastic components. The commonly used meth- 
ods include mechanical fastening, adhesive bonding, thermal bonding, and ultrasonic 
welding. Only thermal bonding and ultrasonic welding are discussed next because the 
first two operations are similar to those used with metals. 

8.4 Processing of Plastics 


FIGURE 8.14 

Steps involved in hot- 
plate joining 




Thermal bonding of plastics. Thermal bonding, which is also known as fusion bond- 
ing, involves the melting of the weld spots in the two plastic parts to be joined and then 
pressing them together to form a strong joint. Figure 8.14 illustrates the steps involved 
in the widely used thermal bonding method known as hot-plate joining. As can be seen 
in the figure, a hot plate is inserted between the edges to be mated in order to melt the 
plastic parts; melting stops when the plate comes in contact with the holding fixture. 
Next, the plate is withdrawn, and the parts are pressed together and left to cool to yield 
a strong joint. The cycle time usually ranges from 15 to 20 seconds, depending upon 
the relationship between the melt time and the temperature (of the hot plate) for the 
type of plastic to be bonded. Also, this process is applied only to thermoplastics. 

Figure 8.15 illustrates different types of joint design. The one to select is depen- 
dent upon both the desired strength and the appearance of the joint. The product de- 
signer must keep in mind that a small amount of material is displaced from each side 
to form the weld bead. This must be taken into account when dimensional tolerance is 
critical, such as when fusion-bonded parts are to be assembled together. 

Another thermal bonding process, which is equivalent to riveting in the case of 
metals, is referred to as the rmo staking. As can be seen in Figure 8.16, the process 

FIGURE 8.15 

Different joint designs 
for fusion bonding 


butt joint 

butt joint 

Bead enclosed 

Bead covered 

Recessed weld 


8 Plastics 

FIGURE 8.16 

The thermostaking 

Hot air 

involves the softening of a plastic stud by a stream of hot air and then forming the 
softened stud and holding it while it cools down. Thermal bonding processes find 
widespread application in the automotive, appliance, battery, and medical industries. 

Ultrasonic welding of plastics. Ultrasonic welding is gaining popularity in industry 
because of its low cycle time of about 0.5 second and the strong, tight joints that are 
easily obtainable. The process is used for thermoplastics and involves conversion of 
high-frequency electrical energy to high-frequency mechanical vibrations that are, in 
turn, employed to generate highly localized frictional heating at the interface of the 
mating parts. This frictional heat melts the thermoplastic polymer, allowing the two 
surfaces to be joined together. 

The product designer must bear in mind that not all thermoplastics render them- 
selves suitable for ultrasonic welding. Whereas amorphous thermoplastics are good 
candidates, crystalline polymers are not suitable for this process because they tend to 
attenuate the vibrations. Hydroscopic plastics (humidity-absorbing polymers, such as 
nylons) can also create problems and must, therefore, be dried before they are ultra- 
sonically welded. In addition, the presence of external release agents or lubricants re- 
duces the coefficient of friction, thus making ultrasonic welding more difficult. 

The equipment used involves a power supply, a transducer, and a horn. The power 
supply converts the conventional 115-V, 60-Hz (or 220- V, 50-Hz) current into a high- 
frequency current (20,000 Hz). The transducer is usually a piezoelectric device that 
converts the electrical energy into high-frequency, axial-mechanical vibrations. The 
horn is the part of the system that is responsible for amplifying and transmitting the 
mechanical vibrations to the plastic workpiece. Horns may be made of aluminum, 
alloy steel, or titanium. The latter material possesses superior mechanical properties 
and is, therefore, used with heavy-duty systems. The horns amplify the mechanical vi- 
bration via a continuous decrease in the cross-sectional area and may take different 
forms to achieve that goal, as shown in Figure 8.17. 

The task of joint design for ultrasonic welding is critical because it affects the de- 
sign of the molded parts to be welded. Fortunately, there are a variety of joint designs, 
and each has its specific features, advantages, and limitations. The type of joint to be 
used should obviously depend upon the kind of plastic, the part geometry, the strength 
required, and the desired cosmetic appearance. Following is a discussion of the com- 
monly used joint designs, which are illustrated in Figure 8.18. 

8.4 Processing of Plastics 


FIGURE 8.17 

Different horn shapes 
employed in ultrasonic 
welding of plastics 

\ r 

Catenoidal horn 

Step horn 

Exponential horn 

FIGURE 8.18 

Different joint designs 
for ultrasonic welding: 
(a) butt joint; (b) step 
joint; (c) tongue-and- 
groove joint; (d) 
interference joint; 
(e) scarf joint 





Parts to be 






1. Butt joint with energy director. The butt joint (see Figure 8.18a) is the most 
commonly used joint design in ultrasonic welding. As can be seen in the figure, 
one of the mating parts has a triangular-shaped projection. This projection is 
known as an energy director because it helps to limit the initial contact to a very 
small area, thus increasing the intensity of energy at that spot. This causes the 
projection to melt and flow and cover the whole area of the joint. This type of 
joint is considered to be the easiest to produce because it is not difficult to mold 
into a part. 

2. Step joint with energy director. The step joint (see Figure 8.18b) is stronger than 
the butt joint and is recommended when cosmetic appearance is desired. 


8 Plastics 

FIGURE 8.19 

Ultrasonic installation 
of metal insert into 
plastic part 

Metal insert 

(diameter bigger 

than the hole) 

Plastic part 


3. Tongue-and-groove joint with energy director. The tongue-and-groove joint (see 
Figure 8.18c) promotes the self-locating of parts and prevents flash. It is stronger 
than both of the previously mentioned methods. 

4. Interference joint. The interference joint (see Figure 8.18d) is a high-strength joint 
and is usually recommended for square corners or rectangular-shaped parts. 

FIGURE 8.20 

Ultrasonic staking 


Flared stake 
diameter less than ^ in. (1.6 mm) 

Spherical stake 
diameter less than -^ in. (1.6 mm) 




Hollow stake 
diameter more than ^- in. (4 mm) 


Knurled stake 

(used for high-volume production 

and/or where appearance and 

strength are not critical) 

Flush stake 

(recommended when the thickness of 

the sheet allows a chamber or a counterbase) 

8.5 Fiber-Reinforced Polymeric Composites 303 

5. Scarf joint. The scarf joint (see Figure 8.18e) is another high-strength joint and is 
recommended for components with circular or oval shapes. 

In addition to welding, ultrasonics are employed in inserting metallic parts into 
thermoplastic components. Figure 8.19 illustrates an arrangement for the ultrasonic in- 
stallation of a metal insert into a plastic part. 

Another useful application of these systems is ultrasonic staking, which is equiv- 
alent to riveting or heading. Figure 8.20 indicates the different types of stakes, as well 
as their recommended applications. Notice that these stakes can be flared, spherical, 
hollow, knurled, or flush. 


In this present age of new materials, at the forefront of advancing developments are 
materials based on the combination of organic polymer resins and high-strength, high- 
stiffness synthetic fibers. This section addresses the materials, processing, and design 
methodology of fiber-reinforced polymeric composites. 

Historical Background 

Although the merits of fiber-reinforced materials have been known for centuries, 
(straw-reinforced clay was reportedly used as a building material by the Egyptians in 
600 B.C.), it is only in the past 40 years that fiber-reinforced polymers have become im- 
portant engineering materials. New synthetic high-strength, high-modulus fibers and 
new resins and matrix materials have elevated fiber-reinforced composites into the ma- 
terial of choice for innovative lightweight, high-strength engineered products. These de- 
velopments along with established engineering design criteria and special processing 
technology have advanced fiber-reinforced composites close to the realm of a commod- 
ity material of construction. In the areas of automobile bodies, recreational boat hulls, 
and bathroom fixtures (bathtubs and shower stalls), fiberglass-reinforced organic poly- 
mer resins have indeed become the material of choice. In more advanced applications, 
the first completely fiber-reinforced polymeric resin composite aircraft came into exis- 
tence in the 1980s. For the 1990s, some important nonaerospace applications are emerg- 
ing, such as sports equipment (sailboat spars) and, more recently, wind turbine blades. 
The utilization of composite materials in functional engineering applications con- 
tinues to grow. It is, therefore, important for engineering students to know about and 
understand these materials so that new uses may be developed and propagated. Con- 
sequently, a brief review of organic polymer engineering composites is presented next. 
A general description of these materials, their unique properties, processing tech- 
niques, and engineering design features will put into perspective present and future 
uses of fiber-reinforced polymer (FRP) engineering materials. 

* Section 8.5 was written by Dr. Armand F. Lewis, Lecturer at the University of Massachusetts Dartmouth. 


8 Plastics 

Nature of Composites 

A composite may be defined as a material made up of several identifiable phases, com- 
bined in an ordered fashion to provide specific properties different from or superior to 
those of the individual materials. Many types of composites exist, including laminated 
materials, filamentary-wound or -layered and particulate-filled compositions, and mul- 
tiphase alloys and ceramics. Most naturally occurring structural materials are compos- 
ites (wood, stone, bone, and tendon). 

Overall, composite materials can be classified according to Table 8.1. We will 
focus on fiber/resin composite materials composed of higher-strength, higher-modulus 
fibers embedded in an organic polymer/resin matrix. Table 8.2 lists some of the com- 
mon resin and fiber materials employed. These composite materials are generally re- 
ferred to as fiber-reinforced polymers (FRP). Currently, polyester and epoxy resins 
are the most common commercially used matrix resin polymers, while glass fibers are 
the most widely used reinforcing fiber. Resin matrix composites containing high- 
strength, high-elasticity-modulus carbon (graphite), polyaramid (Kevlar, a DuPont 
trade name), and boron fibers are also in use for specialty (advanced) composite mate- 
rial applications. 

The integral combination of high-strength, high-elasticity-modulus fibers and rel- 
atively low-strength, low-rigidity polymer matrices forms some unique engineering 
materials. FRP composites possess the material processing and fabrication properties 
of polymeric materials yet, due to their fiber reinforcement, can be designed to possess 
directional stiffness and strength properties comparable to those of metals. These me- 
chanical properties can be achieved at a very light weight. This feature can be illus- 
trated by comparing the specific strength (tensile strength/density) to the specific 
elastic modulus (tensile elasticity-modulus/density) of various fiber-reinforced com- 
posite materials with plastics and metals. Figure 8.21 compares the specific strengths 
and specific elastic moduli of these materials. Notice that commodity elastomers, plas- 

TABLE 8.1 

Classifications of 
composite materials 


Typical Example(s) 

Fiber/resin composites 

Glass fabric/mat reinforced polyester 


resin molded into sport boat hulls 


Heterophase polymer mixtures 

Aluminum and/or graphite powder 

Random particulate filled 

blended into nylon plastic to form a 

Flake or shaped particles 

machine gear 

Interstitial polymeric materials 

"Marbleized" decorative plastic for 

Interpenetrating polymer networks 

wall panels 

Skeletal composites 

Laminar and linear composites 

High-pressure laminates used in 

Material hybrids 

kitchen countertops and 


polyurethane rubber-impregnated 

polyaramid rope/cable 

8.5 Fiber-Reinforced Polymeric Composites 


TABLE 8.2 

Some materials used in 
organic polymer 
engineering composites 

/ Matrix Resin 

Maximum Service Temperature \ 


Up to 121°C (250°F) 


Up to 62°C (non HT) 

Vinyl ester 

Up to 145°C (HT type) 


Up to 149°C (300°F) 


Up to 260°C (500°F) 



Up to 80°C (175°F) 

Polyphenylene sulfide 

Up to 149°C (300°F) 

Polyetheretherktone (PEEK) 

Up to 200°C (392°F) 

Fiber (Continuous Yarn/Filament, 

Woven Fabric, 

Nonwoven Mat, Chopped Fiber) 

Glass (especially E and S glass) 

Polyaramid organic fiber (Kevlar)® 

(Dupont trademark) 



Form for Processing 

Liquid casting resin 

B-stage resin mixture 

Preimpregnated (prepreg) B-stage resin/fiber/fabr 

ic combination 

tics, and metals occupy only a very small portion of this structural materials map. 
Fibers and fiber-reinforced resin composites occupy the outer regions. Fiber-reinforced 
composites can have specific strengths and moduli up to six times those of common 
structural materials. For a given weight, fiber-reinforced composites far outperform 
other engineering materials in their strength and stiffness. These specific strengths and 
moduli approach the mechanical properties of theoretically perfect, ordered polymer 
crystals. This property makes composite materials unique among engineering struc- 
tural materials and opens new horizons for novel engineering designs. For example, 
composite materials are widely used in aircraft and aerospace applications: The FRP 
property of high specific strength with high elasticity modulus made possible the de- 
sign, construction, and functional deployment of the U.S. Air Force all-carbon fiber- 
reinforced epoxy resin composite Stealth reconnaissance aircraft. 

The observed high strength and stiffness-to-weight ratio of fiber-reinforced com- 
posites can be easily explained. Various material properties of composites can be esti- 
mated by a rule of mixtures approach. Micromechanic properties such as modulus 
(stiffness), strength, Poisson's ratio, and thermal expansion of fiber-reinforced polymer 
composites can be estimated by the following equation: 

M c = V f M t + V m M n 



8 Plastics 

FIGURE 8.21 

Specific strengths and 
specific elastic moduli 
of materials 

1 ' 



' ' 


Composite (unidirectional) 
















High Modulus 




High Strength — 









B _ 

~ x mn 


>x (HP 

■*— B/Ep 

. Gl/Ep 





Chopped \^ 

— ^-^y^^ffiPtx 



(py r^l^vr 

© W | \^ 

i_ _L 



200 400 600 

Specific modulus, inches x 10 -6 


where: M is the particular material property 

V is the volume fraction of the fiber (f ) or matrix (m) 
M c is the material property of the composite "mixture" 

The individual material component properties, therefore, contribute by a volume frac- 
tion ratio to the properties of the combined composite materials. For this rule of mix- 
tures equation to apply, several basic assumptions and limitations are involved: 

1. The fiber/polymer matrix composites as well as the polymer matrix are assumed 
to be linearly elastic and homogeneous. 

2. There are no voids in the composite, and there is good adhesion between the rein- 
forcing fibers and the polymer matrix. 

3. The proximity of the fiber and polymer does not alter the properties of the indi- 
vidual components. 

4. The rule of mixtures has some directional limitations as many FRPs are not 
isotropic materials. 

8.5 Fiber-Reinforced Polymeric Composites 307 

For example, if we are dealing with a continuous fiber-reinforced polymer resin 
composite, the modulus and strength properties of the composite will be very different 
in the direction longitudinal to the fiber length compared to the properties across or 
perpendicular to the fibers. For strength and modulus, Equation 8.1 is most appropri- 
ate for composites being tested in the longitudinal (fiber) direction. The mechanical 
contribution of the fibers are directly in line with the direction of pull. The fibers are 
strong and stiff in this longitudinal direction, and the polymer matrix is relatively weak 
and much less rigid. Note that the strength and stiffness of materials in fiber form 
are always much higher than bulk materials (e.g., bar, rod, plate) because the fiber 
form of a material has a more atomically ordered internal structure. Fibers have an 
internal crystalline structure that favorably alters the stiffness and fracture behavior 
of this form of material. The presence of fibers makes composites stiffer and stronger 
in the longitudinal (fiber) direction than the polymer matrix by itself. The term fiber- 
reinforced polymer is thus appropriate. Property directionality effects are very impor- 
tant to consider in the use of fiber-reinforced composites in engineering designs. 

Fiber Reinforcement 

Generally, reinforcement in FRPs can be either fibers, whiskers, or particles. In composite 
materials of the most commercial interest, fibers are the most important and have the most 
influence on composite properties. Table 8.3 presents a comparison of the most common re- 
inforcement fibers used in preparing organic polymer engineering composites. Nylon fiber 
is included here as areference fiber. All the materials listed in Table 8.3 are textile fibers and 
can, for the most part, be processed into manufactured products in the same manner as tex- 
tile fibers (e.g., continuous yarn, wound filaments, woven and knitted fabrics, nonwoven 
mats). The high-strength and high-stiffness properties of the glass (S-2), carbon, and pol- 
yaramid fibers are evident. These reinforcing fibers, when used in composite material fab- 
rication, can take several forms, such as <0. 1 inch (3^4- mm) fiber "whiskers," 0. 1 -0.3 inch 
(3-10 mm) chopped fibers, 0.1-2.0 inch (3-50 mm) (nonwoven) matted fiber sheets, 
woven fabric (continuous) fiber with plain weave, and unidirectional/longitudinal (contin- 
uous) fiber ribbons. These fiber reinforcement forms are illustrated in Figure 8.22. When 
using fiber reinforcement in polymer composites, the surface of the fibers or yarns is pre- 

TABLE 8.3 

Comparison properties of various fibers 

S-2 Glass 

Carbon T-300 


Nylon 6/6 

Tensile Strength, 





lb/in. 2 (MPa) 





Modulus of Elasticity, 





lb/in. 2 (MPa) 





Elongation, % 










lb/in. 3 (g/cm 3 ) 






8 Plastics 

FIGURE 8.22 

Comparison of fiber 
reinforcement forms 

\ s N ' ' " S-- \ ^ X ' 

' ' 

~ ' .", - x ' \ ~~ ' ', - 

N ~" 


-. 1 / N ' ^ - __ - 1 / 

\ / 


, V / - v - ^ X 

~ • 

<3-4 mm 

Chopped fibers 
3-1 mm 

Fiber (non-woven) mat 
>3-50 mm 




Woven fabric 

Parallel aligned yarns 

8.5 Fiber-Reinforced Polymeric Composites 309 

treated with a chemical coupling agent to enhance wetting and adhesion of the matrix resin 
to the fibers. Here, the chemical coupling agents are made specific to the chemical nature of 
the matrix resin being used. It is important that the fiber supplier be consulted for the proper 
type of fiber/resin coupling agent when fiber reinforcement materials are purchased. 

As Table 8.3 shows, the most commonly used reinforcing fiber material is glass. 
In particular, S-2 glass is used in most high-performance applications. There exists an 
extensive applications, manufacturing, and processing history involving the use of glass 
fiber in polymer composite applications. Various forms of carbon fiber are also used 
for high-performance applications. The processing of carbon-fiber-reinforced polymer 
composites follows similar procedures as glass fibers. However, in the continuous-yarn 
processing of carbon fibers, precautions must be taken to protect electrical processing 
equipment from damage. Airborne, electrically conducting graphite dust is generated 
when the carbon fibers or yarns are processed through guide rings and rollers. This can 
occur before the fibers are wetted by the matrix resin during material fabrication. The 
dust can ruin electrical equipment if it is allowed to penetrate an instrument's enclo- 
sure. Sometimes, explosion proof electrical equipment is used when processing carbon 
fibers. Another approach is to fit the electrical instrument housing with a positive pres- 
sure differential of clean air (or nitrogen gas). 

Matrix Resins 

Classification of polymer matrices. Many types of polymers and resins can be rein- 
forced by fibers to create FRP composite materials. Polymer matrices can be classified 
into two basic categories: thermoplastic and thermosetting. 

1. Thermoplastic. Many of the polymers previously discussed can be reinforced with 
fibers to form composites. The most common types are chopped-fiber-reinforced 
thermoplastics. These materials can be processed in the same way as nonfiber- 
reinforced plastics. Generally, chopped fibers are blended and mixed into a molten 
mass of the engineering thermoplastic (e.g., nylon, polycarbonate, acetal) in a melt- 
extruder type of plastics-compounding machine. The fiber containing plastic is ex- 
truded into a thin rod and cut into molding powder or pellets. This thermoplastic 
molding powder is then used for injection molding or extrusion of engineered parts 
similar to the unreinforced plastics discussed in the preceding sections. 

Continuous fibers such as glass, carbon, or polyaramid have also been prepared 
with thermoplastic resin matrices. The concept here is to first coat thermoplastic 
resins onto continuous-fiber yarn by a hot melt or a polymer solution-solvent-dip 
process. These thermoplastic polymer-coated yarns can then be fabricated into 
shaped structures by a (hot press) matched-die compression molding technique or 
other techniques for affecting molten-polymer controlled consolidation. At this 
time, discontinuous chopped-fiber thermoplastic composites are much more widely 
used than continuous fiber-reinforced composites. The main advantages of thermo- 
plastic matrix fiber composites is that they can be processed, for the most part, in 
conventional thermoplastic polymer fabrication equipment. Furthermore, any scrap 
or off-quality material can be recycled back into the injection molding or extruding 
machine. However, care must be taken during this thermoplastic processing not to 

310 8 Plastics 

overly "work" these materials in the molten state. Excessive processing in the 
molten state severely shortens the overall reinforcing fiber length, which can di- 
minish the reinforcement effect of the fiber in the polymer matrix. 

2. Thermosetting. Reinforced composites are traditionally associated with thermoset- 
ting polymers such as the unsaturated polyester and epoxy resins. In their cured 
state, thermosetting resins are composed of long polymer chains that are joined to- 
gether through cross-bridges that link together all the molecules in the resin mass. 
The final, hardened, tough, and glassy state of the cured resin is the terminal con- 
dition of the polymer resin matrix. In this state, the resin serves the all-important 
role of structurally consolidating, supporting, and cohesively tying together the 
fiber reinforcement in the composite. However, during initial processing, it is im- 
portant that thermosetting resins undergo a gradual liquid-to-solid conversion. It is 
this feature that renders thermosetting resins of the unsaturated polyester and epoxy 
type most readily adaptable to fiber-reinforced composite component fabrication. 

Sequence of FRP fabrication with respect to the resin system involved. At first, the 
resin is in a liquid state as it is received from the supplier. It may be more or less fluid 
depending on its viscosity (from a flowable waterlike consistency to a high-viscosity 
syrup). At this stage, rheological thickeners to increase resin viscosity or reactive dilu- 
ents to decrease resin viscosity may be added to the resin formulation. Frequently, the 
curative part of the resin system is much more fluid than the resin part. Here, the vis- 
cosity of the final mixed resin and curative system is low enough to accommodate 
proper flow in processing. Sometimes, the fluidity of the resin may be lowered by in- 
creasing the temperature of the resin upon its application to the fiber. In all, it is im- 
portant that the viscosity of the liquid resin be adjusted so that it has the proper fluidity 
to wet, impregnate, and saturate the reinforcing fiber yarns, fabric, or mat. 

The next consideration is the need to chemically catalyze the resin so that it prop- 
erly cross-links and cures the resin under the prescribed conditions. It is also necessary 
to have the catalyzed resin react very slowly at ambient temperature so that it remains 
fluid while it is in the process of wetting the reinforcing fibers. This resin-system fluid 
time is referred to as the pot life or open time of the resin. This fluid-time feature is 
controlled by the nature of the catalyst, the ambient temperature, and the bulk volume 
of resin in the resin container. Note that a bulk of catalyzed resin is a resin undergoing 
a heat-generating exothermic reaction. If the bulk volume of the resin is too large, heat 
cannot be easily dissipated. The reaction in the fiber/resin dip tank will automatically 
accelerate, the resin will cure, or, worse, the heat of the reaction may cause a serious 
fire as well as noxious fumes. Most often, however, the processing equipment will con- 
tain dual-component pumps and a mixing head that will continuously meter and mix 
the proper components of the resin system (resin: part A; curative: part B) at the ap- 
propriate moment and position for wetting the fibers. 

Once the resin part and the curative part of the resin system have been mixed, the 
liquid-to-solid cure reaction of the resin begins. The curing resin system will undergo 
several stages: liquid/fluid, gel stage, rubbery stage, and tough/glassy solid. Depending 
upon the processing temperature, the liquid-to-gel-to-rubber transition may occur from 
hours (for room temperature cures), to minutes, to seconds. The gel point in a ther- 

8.5 Fiber-Reinforced Polymeric Composites 


mosetting resin system is the point in the cure-time sequence when the resin undergoes 
a sharp rise in viscosity and ceases to be a fluid. Theoretically, the gel point is defined 
as the time in the cure when each polymer molecule in the system is tied together by 
at least one cross-link. Therefore, at the gel point, the polymer molecules in the resin 
system have combined and have reached an infinite molecular weight. After the gel 
point, the number of cross-links in the polymer system continues to increase, the cross- 
link network gets tighter and tighter, and the resin becomes a solid. It is at the gel stage 
that the influence of cross-linking takes hold. The rubbery stage is intermediate in 
cross-linking. In the solid glassy state of the resin, the ultimate number of cross-links 
in the resin system exists. Figure 8.23 illustrates the nature of the polymer resin and 
curative molecules during the curing sequence. Note that it is only in the solid-state 
stage that the fabricated composite part retains its shape and may be moved for addi- 
tional processing or given a postcure if necessary. Let us now examine the specific 
resin chemistry of the unsaturated polyester and epoxy resin systems. 

Chemistry of the unsaturated polyester resin system. Unsaturated organic polymers 
are polymer systems containing double bonds, or C = C. Double bonds can react with 
each other by an addition reaction that can be initiated by a free-radical catalyst. With 
the help of free-radical catalysts, unsaturated organic compounds can react with each 
other to form high-molecular-weight polymers: 



C = 

= C 





bond opening 

r i 

3D O I 





In unsaturated polyester resins, the resin part of the mixture is represented by high- 
molecular-weight polymer molecules having unsaturated groups in their chain. These 
unsaturated polyesters are readily soluble in the unsaturated organic liquid compound 
called styrene. Styrene (known here as a monomer) can easily react with itself (using a 
free-radical catalyst initiator) to form a styrene polymer, or polystyrene. Because the 
monomeric styrene can readily react with unsaturated groups, when liquid styrene is 
mixed with unsaturated polyester resin, it serves the dual role of a reactive diluent and 
cross-linking agent. If a free-radical catalyst is added to a solution mixture of unsatu- 
rated polyester resin and styrene, the styrene simultaneously reacts with both the unsat- 
uration in the backbone of the polyester chain and with itself. With free-radical catalysis, 
the polymerization reaction involving the growing polystyrene chains that react with the 
one polyester chain can also react with itself. When this reaction, in turn, connects with 
another polyester chain, a cross-link is formed between the two chains. In the molecu- 
lar mixture mass of styrene, growing polystyrene chains, and unsaturated polyester mol- 
ecules, an array of cross-links are formed between the multitude of polyester molecules 
(see Figure 8.24). As the polymer system reacts, from its initial mixing of the catalyst, 
the resin system will change from a liquid, to a gel-rubber when cross-linking starts to 


8 Plastics 

FIGURE 8.23 

Nature of molecules at 
various stages of 
thermosetting resin 

BACKBONE Polymer (Pre-Polymer) 


Liquid - All molecules 
are independent, 
can flow past each 

Gel - At least one 
crosslink attachment 
to each backbone 
ploymer molecule. 

Rubber - More 
crosslinks, backbone 
still flexible 

• Glass - High 
crosslink density, 
tight network 

8.5 Fiber-Reinforced Polymeric Composites 


FIGURE 8.24 

Chemical structure of 
polyester/styrene resin 

H H 

• H 

<{> H 


polyester resin 



H H H H H 
/™ — c — C — C — C — C'™ 

(j) H (j> H (J) 

H H \ 

1 1 \ 


c — c — 


1 1 


<t> H In 



-</WN C C /WWWW\ Q Q 

c Crosslinks 




occur, and, finally, to a solid glassy vitreous state when numerous cross-links form and 
tie together the polyester molecules in the resin system. This, in essence, is the chemi- 
cal mechanism that characterizes the cure of a typical polyester resin. 

In the commercial formulation of unsaturated polyester/styrene thermosetting FRP 
resins, to make the resin more sag resistant when applied to vertical and more con- 
toured surfaces, fumed silica is added to alter the rheology of the liquid resin. Another 
additive involves using a wax material that serves as a surface active agent that allows 
the resin to cure more evenly at its surface. Unsaturated polyester resin systems are by 
far the most widely used FRP matrix resins because of their low cost and availability. 
However, their use is being questioned because of environmental concerns. Styrene 
monomer is quite odiferous, and questions are being raised regarding its human toxic- 
ity after long-term process-operator exposure. 

Chemistry of the epoxy resin system. Because of their inherent good adhesion to all 
types of surfaces, epoxy resins are generally more difficult to work with than poly- 
esters. However, epoxies have much better thermal properties and exhibit very low 
shrinkage during cure. Their adhesive properties, while adding process difficulties, 
serve to enhance the structural integrity of the fiber/resin composite material system. 
Epoxies provide good adhesion of the resin matrix to the reinforcing fibers. The major 
hardeners for epoxy resins are amines and anhydrides. The chemistry of these hard- 
ener/curative systems is discussed next. 

314 8 Plastics 

Epoxy resins are characterized by the reaction of the epoxy group c c 

known as the oxirane ring. Polymerization reactions proceed by the opening of this 
oxirane ring to form a difunctional chemical-reacting specie similar to the unsaturated 
C = C group in polyesters. Epoxy resins are low-molecular- weight polymers contain- 
ing oxirane rings at each end of the chain. They are cured by adding a multifunctional 
chemical to the mixture that serves to cross-link the system by an addition reaction 
with the oxirane ring. The most common cross-linking agents for epoxies are the 
amines. Many of the amines used to cure epoxies are liquids, which makes the amines 
serve as reactive diluents. Such liquid material systems are also easily adaptable to 
dual-component pumps and the mixing of resin during dispensing for processing. The 
basic reaction between (primary) amine groups and the epoxy group is as follows: 


RNH, + CH 

1? — 

— CH *~ RNH CH 2 CH 

? H o 

/ \ 

CH ? 

CH + CH 2 CH >- RN (CH 2 

RNH CH 2 CH + CH 2 CH ► RN (CH 2 CHOH) 

As shown, each of the two hydrogen atoms of the primary amine, RNH 2 , where R is a 
generic unspecified organic grouping, is capable of reacting with one epoxide group. In 
this chemical process, as with polyester resins, the epoxy polymer passes from a liquid to 
a gel-rubber to the solid glassy state as the cross-linking reaction proceeds. During the lat- 
ter stages of the reaction, the resultant OH groups that are formed in the amine reaction can 
also react with epoxy groups and further increase the cross-link density of the polymer. 

One advantage of amine-cured epoxy resins is that they can harden or cure at 
room temperature. However, room temperature curing leads to polymers with low 
temperature stability. Also, the moisture resistance of these epoxy resins is generally 
low. Both temperature and moisture resistance can be improved by postcuring the 
resins above 212°F (100°C). Here, the chemical cross-links of the resin are maximized 
as complete reaction of all the epoxy groups is approached. 

The reaction of anhydride curing agents with epoxy resins is more complex than 
that of amine cures. With anhydrides, amine catalysts are required along with cures at 
high temperature. During reaction, several competing reactions can take place. The 
most important reactions are as follows: 

1. Opening of the anhydride ring with the OH groups from the catalytically reacted 
epoxy groups to form a carboxyl group: 

c c^ ! c c o CH 

O + HO CH *- : 

— c — c i — c — c — OH 

I V 



anhydride epoxy resin reaction product 


8.5 Fiber-Reinforced Polymeric Composites 315 

2. Subsequent reaction of the carboxyl group with the epoxy group: 

o o 

■ o 

C C O CH / \ C C O CH 

i + CH 2 — CH • — *~ i 



3. Epoxy groups, in turn, reacting with the formed OH groups: 

! /°\ 

HC OH + CH 2 CH ►■ = 

= HC O CH ? CH 

Although all three reactions can occur, which of the three reactions predominates de- 
pends on the reaction temperature. 

Compared to amine cures, the pot life of anhydride cures is long, and the reaction 
produces a low exotherm. Long-time, elevated-temperature cures up to 392°F (200°C) 
are necessary if ultimate properties are desired. Overall, compared to amine-cured sys- 
tems, anhydride cures result in much better chemical resistance for the final cured 
resin product. 

From a processing standpoint, the environmental and industrial hygiene aspects of 
amine- or anhydride-cured epoxy resins are much better than the hygiene problems 
associated with unsaturated polyester resin processing. In all cases, proper protective 
clothing (coat, gloves, and goggles) must be worn while working with these resins. 
Amine and anhydride chemicals are generally quite corrosive to the skin and may cause 

Forms of Composite Materials 
and Fabrication Techniques 

Discontinuous fiber reinforcement. The reaction injection molding (RIM) process in- 
volves bringing together two components of a thermosetting polymeric resin system in 
a mixing head and injecting the reacting mixture into a closed mold before reaction is 
complete, as illustrated in Figure 8.25. The resin system then cures in the mold at a rel- 
atively low pressure of 50 psi (345 kPa). The timing of the curing reaction is very im- 
portant because the reaction must occur at the moment the mold cavity is filled. Close 
process control is required. Because the process involves low-viscosity intermediates, 
complex parts can be fabricated using the RIM method. 


8 Plastics 

FIGURE 8.25 

The reaction injection 
molding (RIM) process 

component A 

component B 

^ cylinder 

Reinforcement (glass, fiber, or flake) can be added to one of the resin components 
prior to mixing if increased flexural modulus, thermal stability, and, in some instances, 
a special surface finish is desired in the final molded product. This process, reinforced 
reaction injection molding (RRIM), is shown in Figure 8.26. 

Structural reaction injection molding (SRIM) and resin transfer molding (RTM) 
are similar to RRIM, except that the reinforcement is placed directly into the mold 
prior to the injection of the resin. In SRIM, the reinforcement is typically a preform of 
reinforcement fibers or mat of nonwoven fibers. In RTM, as shown in Figure 8.27, a 
catalyzed resin is pumped directly into the mold cavity containing the reinforcement. 

FIGURE 8.26 

The reinforced reaction 
injection molding 
(RRIM) process 


component A 

component B 



8.5 Fiber-Reinforced Polymeric Composites 


FIGURE 8.27 

The resin transfer 
molding (RTM) process 


Dry Reinforcement 

The resin system is such that it cures without heat. The advantages of RTM are that, 
because no mixing head is involved, a relatively low investment is needed for equip- 
ment and tooling. Furthermore, large FRP parts and parts containing inserts and cores 
can be fabricated using the RTM process. RIM, RRIM, SRIM, and RTM processing 
are widely used in the automotive and aerospace industries. 

Wet lay-up and vacuum bagging. Imbedding plies of glass, carbon, and/or polyaramid 
plain-weave fabric or fibrous mat into an uncured liquid resin and allowing the liquid 
resin to solidify (cure) while being constrained by a mold or form is a common pro- 
cessing technique used in the pleasure boat building industry. A typical arrangement of 
the plies used in this technique, called the wet lay-up process, is shown in Figure 8.28. 
Related to this wet lay-up process is the vacuum bagging method of fabricating com- 
posite parts and shapes. The principle of vacuum bagging is quite simple. The shape to be 
fabricated is prepared by a room temperature wet lay-up procedure as just described. The 
part to be fabricated is usually assembled over a form or shape of the desired (complex 
and/or contoured) part. The assembly, like the lay-up arrangement shown in Figure 8.28, 
is then placed in an airtight disposable plastic "bag" fitted with a vacuum tube fitting or 
stem. If the air is sealed off and then evacuated from it, the bag will automatically close in 
on the wet laid-up plies of fiber and liquid (uncured) resin and consolidate these plies by 

FIGURE 8.28 

Arrangement of plies in 
wet lay-up assembly 

u a d n u u u a u u u u o~q 
n n n n n n w n n rrri n n rn 

VBF - Impermeable vacuum bag film 

B - Conformable nonwoven bleeder/breather fabric 

P - Perforated release film 

C - Fiber reinforced resin composite part 

S - Pressure sensitive flexible sealant 


8 Plastics 

the action of atmospheric pressure. This composite assembly is then allowed to solidify 
(cure) at room or elevated temperature. After this cure time, the vacuum bag, bleeder 
ply, and resin-absorber material are removed from the assembly and discarded, leaving 
the fabricated composite part ready for subsequent finishing or treatment. 

A variation of the wet lay-up method is the spray-up process, where a spray gun si- 
multaneously sprays catalyzed resin and chops continuous glass yarn into specific 
lengths. As shown in Figure 8.29, chopped fibers enter the spray nozzel of the spray gun, 
and the materials are comixed and sprayed onto an open-cavity mold. The mold usually 
is faced with a smooth coating of already cured resin called a gel-coat or a thermoplas- 
tic shell. This forms the outer surface of the structure being fabricated. When the 
sprayed-on fiber-reinforced resin cures, the part is removed from the mold. The laminar 
structure formed is composed of an aesthetically acceptable or otherwise finished outer 
skin. Adhered to and backing up this skin is the cured fiber-reinforced resin. Open-mold 
processing of this type is used extensively in bathtub and shower stall applications. 

Unidirectional-fiber resin prepregs. Fiber-reinforced composite materials are com- 
monly used in the form of a prepreg. Prepregs are typically side-by-side aligned fiber 
yarns that have been impregnated by a B-staged resin matrix (meaning that it has been 
deliberately partially cured). Unidirectional-fiber composite prepregs are commercially 
available in the form of rolls, tapes, and sheets. One drawback is that these prepregs 
must be kept frozen, below 32°F (0°C), for shipping and storage before use. They also 
have a relatively short shelf life. If not properly stored, the B-stage resins will cure 
slowly at room temperature, and their function will be destroyed. 

Prepreg material is used to fabricate structures by plying together lay-ups of these 
resin-impregnated unidirectional fibers. The lay-ups can be designed to have different 
desired mechanical properties depending upon the geometrical arrangement or assem- 
bly of the reinforcing fibers in the cured lay-up. Some typical unidirectional-fiber ply 
arrangements are shown in Figure 8.30. Mechanically, these unidirectional (0°, 0°), 
cross-ply (0°, 90°), and quasi-isotropic (0°, +45°, 90°, -45°, 0°) plied laminates will 

FIGURE 8.29 ^^ 

The spray-up process f J \ — 

8.5 Fiber-Reinforced Polymeric Composites 


FIGURE 8.30 

Various arrangements 
of unidirectional-fiber 
ply laminates 



(0°,+45 o ,-45 o ,90°) 

have planar anisotropic properties. Their flexural stiffness will always be higher in the 
longitudinal direction of the fibers. Other forms of B-stage resin-impregnated fiber 
forms are commercially available (e.g., fabrics and fibrous mats). The numerous B- 
stage precomposite forms and types of fiber are all available to the composite materi- 
als design engineer in the construction of a fiber-reinforced composite structure. 

320 8 Plastics 

Filament winding. Filament winding is a fiber-reinforced composite processing pro- 
cedure commonly used to fabricate tubular (hollow) and cylindrical tank or bottle- 
like structures. The apparatus used in the filament winding process is shown in Figure 
8.31a. Basically, filamentary yarns are fed off a spool that is mounted on a creel. The 
yarn is immersed in a catalyzed, but still liquid, resin bath, where the yarn is impreg- 
nated with the resin. After squeezing out excess resin, the resin-impregnated yarn is 
wound onto a rotating mandrel in a controlled and directed manner. A computer sys- 
tem and control arm guide the yarn back and forth across the mandrel in a predeter- 
mined pattern. The computer controls the type of wind pattern and the number of 
layers of yarn filaments to be laid down on the mandrel surface. Two types of wind 
patterns are possible: circumferential and helical, (as Figure 8.31b shows). In the cir- 
cumferential or hoop wind, the yarn is wound in a continuous manner in close prox- 
imity alongside itself. No crossover of the yarn occurs during the lay-down of a given 
layer, and the lay-down pattern can thus be considered to be at a zero wind angle. The 
wind proceeds back and forth across the mandrel until the desired number of layers is 
accomplished. In the helical wind, the yarn is permitted to cross over itself and tra- 
verses the length of the mandrel at a prescribed angle (e.g., 10°, 30°, 45°). Again, the 
wind proceeds back and forth across the surface of the rotating mandrel until the de- 
sired number of layers is formed. 

In practice, combinations of hoop and helical wind are usually performed to fab- 
ricate a part. The desired lay-down sequence is programmed on the computer. While 
the desired (yarn) filament-wound resin composite is being formed on the mandrel, 
heating lamps can be focused on the resin/fiber mass to affect partial cure of the resin 
during this lay-down step. Once the desired winding pattern is completed, the man- 
drel with its wound fiber/resin composite outer surface is left rotating. Rotation and 
heat-lamp curing continue until the resin material is in a rigid enough state that the 
rotation can stop and the cylindrical part and mandrel can be removed from the fila- 
ment winding machine. Postcuring of the wound composite and mandrel can then be 
accomplished by placing the assembly in an oven. After final curing, the mandrel is 
removed from the core of the assembly. To facilitate this, the mandrel form is gener- 
ally made with a slight taper along its length so that the mandrel can easily be 
slipped out of an end, leaving the desired filamentary composite cylindrical "shell." 
The composite part can then be machined and/or post-treated to the desired condition 
or form. 

Pultrusion processing. Pultrusion is a fiber-reinforced resin processing technique 
that is readily adaptable to the continuous manufacture of constant cross-sectional lin- 
ear composite shapes. Rods, I beams, angles, channels, and hollow tubes and pipes are 
commonly produced by pultrusion processing. Pultrusion is a linear-oriented process- 
ing method whereby yarns of reinforcing fiber are continuously immersed in and im- 
pregnated with a catalyzed fluid resin. As the term pultrusion indicates, these 
resin-impregnated continuous-fiber yarns are concurrently pulled through an elongated 
heated die designed so that the fiber/resin composite mass exiting the die is sufficiently 
cured and retains the cross-sectional shape of the die. The apparatus used in the pul- 
trusion process is shown in Figure 8.32a. In practice, prescribed lengths of the formed 

8.5 Fiber-Reinforced Polymeric Composites 


FIGURE 8.31 

The filament winding 
process: (a) apparatus; 
(b) wind patterns 


Yarn spools 
on a creel 

Hoop wind 






Multiple helical 


Hoop and helical 


piece can be cut using an in-line cutoff wheel. Pultrusion is, therefore, adaptable to 
low-cost, continuous production of constant cross-sectional composite shapes. The 
process of pultrusion is critically controlled by the resin system used (e.g., unsaturated 
polyester, epoxy, and vinyl ester resins), the temperature and temperature profile of the 
heated die, and the rate of pulling through the die. 

In the manufacture of pultruded shapes, such as those shown in Figure 8.32b, al- 
though the core cross section of the composite is linear oriented, there is often a need 
to wrap the outer surface of the composite with a webbing (nonwoven or woven tape) 
of fibrous material. This serves to consolidate the pultruded shape and gives a much 
more durable outer surface to the finished part. In this instance, thin veils of non- 
woven or woven fabric tapes are fed into the entrance of the die along with the resin- 
impregnated continuous-fiber yarns. This assembled mass of fibers and resin proceeds 
to be pulled through the die as just described. The manufacture of hollow pultruded 
shapes is common, and a special die is then required. A shaped insert or "torpedo" is 
fitted at the die entrance and extends partway into it. The fluid resin-impregnated fibers 
entering the die are now constrained by this center-core obstruction. With the proper 


8 Plastics 

FIGURE 8.32 

The pultrusion process: 
(a) apparatus; (b) cross- 
sectional designs 







Yarn spools 
on a creel 




Structural beams 







pipe and 

pulling speed, die temperature profile, and catalyzed resin formulation, the shape of 
the insert is retained as the desired hollow cross section of the part exits the die. 

Engineering Design with Composite 

In the development of commercial products, there are many considerations. The par- 
ticular field of organic polymer engineering composites is no exception. It is impera- 
tive for the engineer to have an integrated understanding of the design, materials 
behavior, processing, and service performance behavior of composite materials in 
order to develop a successful product. This integrated approach is diagrammed in Fig- 
ure 8.33. Thus far, this section has reviewed some of the materials and processing as- 
pects of fiber-reinforced organic polymer composites. The engineering design and final 
application aspects of composite materials are covered next. 

8.5 Fiber-Reinforced Polymeric Composites 


FIGURE 8.33 

Model of technical base 
for engineered 
composite materials 
product development 

First, however, in order to carry out an engineering design with organic poly- 
mer composites, the engineer must recognize and understand their advantages and 
limitations. Some of the advantages and disadvantages of carbon and polyaramid 
fiber-reinforced polymeric composites are as follows. 

Advantages of carbon fibers. 

1. High stiffness-to-weight and strength-to-weight ratios 

2. High compressive strength 

3. Excellent fatigue resistance 

4. Good wear resistance (self-lubricating) and low friction coefficient 

5. Mechanical vibration damping ability better than metals 

6. Excellent creep resistance 

7. Corrosion resistance (when not in contact with metals) 

8. Some (directional) electrical and thermal conductivity 

9. Very low (to slightly negative) directional thermal expansion coefficient 

10. Very broad engineering design versatility 

11. Broad processing versatility 

324 8 Plastics 

12. Less energy required to manufacture engineering composite structures than to 
fabricate with metals 

Advantages of polyaramld fibers. 

1. High stiffness-to-weight and strength-to-weight ratios 

2. Excellent fatigue resistance 

3. Excellent corrosion resistance 

4. Good vibration damping properties 

5. Better impact resistance than carbon fiber composites 

6. Electrically insulating 


1. Limited service temperature 

2. Moisture sensitivity/swelling/distortion 

3. Anisotropic properties 

4. Low compression strength (polyaramid fiber) 

5. Bimetallic corrosion (carbon fiber) 

6. Relatively high cost of advanced fibers 

With these features and limitations in mind, the design engineer can proceed to 
create unique products. In the composites field, it is not appropriate to think only of 
using composite materials as a materials replacement for existing products. New prod- 
ucts that take advantage of the unique properties of composite materials can also be 
conceived. Many of these new product concepts involve exploiting the remarkable 
specific strengths and specific elastic moduli of the "advanced" fiber- reinforced com- 
posites. The design engineer can choose from a multitude of reinforcing fiber types 
and fiber geometry arrangements, as well as from a variety of matrix materials. He or 
she has the freedom to mix in the design specification two or more diverse fiber types, 
as well as the freedom to directionally place the reinforcing fibers. All these degrees 
of freedom of choice are available so that the desired final component can be de- 
signed and fabricated. Fiber-reinforced organic polymer engineering composites are, 
therefore, capable of being used to create what can be referred to as integral design 
engineering material structures (IDEMS). Through computer-aided design (CAD) 
and finite-element stress analysis (FEM) techniques, new products are developed in 
computer-model form. In creating the actual fabricated product, the other facets of the 
integrated materials system manufacturing operation come into play (see Figure 8.33). 
Some specific areas in the design of organic polymer composites are discussed next. 

Cutting, hole drilling, and machining. Although composite parts and structures are 
process molded to the near-finished state, machining, drilling, and trimming are often 
required as final steps. Therefore, the assembly and the finishing of the fabricated part 
are important in the creation of a final commercial product. There is always the possi- 
bility of damaging the composite material in these finishing post-treatments. Delamina- 

8.5 Fiber-Reinforced Polymeric Composites 325 

tion, edge fraying, matrix cracking, or crazing leading to weak spots in the composite 
material structure are all possible. Great care must be taken to maintain the compos- 
ite's structural integrity and appearance. 

Post-treatment of fiber-reinforced composites involves different tooling and pro- 
cedures compared to what is done for metal or plastics. The abrasiveness of the fiber 
and the possibility of fragmentation of the matrix resin are two factors to consider. 
Composites are machined, cut, and trimmed more easily using processes similar to 
grinding or abrasive cutting rather than conventional metal-cutting techniques. Also, 
the method used is dictated by the type of fiber reinforcement. Glass fiber, carbon 
fiber, and especially polyaramid fiber composites all require their own procedures. 
For example, the cutting of polyaramid-fiber-reinforced composites is difficult be- 
cause the fiber is so tough and does not cleave or cut in a brittle, fracture mode. Pol- 
yaramid fibers undergo a process called fibrillation when "damaged" by the drilling, 
cutting, or machining tool. Fuzzy edge cuts or fiber-filled drill holes are produced 
when conventional machining and drilling tools are used. For polyaramid and for 
other fiber-reinforced composite materials for that matter, water-jet cutting, laser cut- 
ting, and diamond wire cutting are often used to achieve an acceptable edge profile 
to the final machine-finished parts. For carbon-fiber-reinforced composites, the ther- 
mal effects due to laser cutting, machining, and drilling can be a deterrent because 
the carbon fibers are thermally conductive. A weakened, charred, heat-damaged zone 
may surround the laser-cut edge. In summary, great care must be taken in the finish- 
ing post-treatments of fiber-reinforced composite materials. 

Adhesive and mechanical joining. Adhesives are the principal means of joining com- 
posite materials to themselves and other materials of construction (metals, plastics, 
wood). The reasons for this are numerous. Most importantly, adhesive bonds are 
uniquely capable of distributing stress and can easily be joined into contoured shapes. 
In mechanical joining, hole drilling is required, which can lead to delamination of the 
composite and a stress concentration at the point of joining. The transfer of load from 
one material to another without creating large stress concentrations is the ultimate goal 
of materials joining. This can be achieved better by adhesive joining. Adhesives can 
often be incorporated into the structural laminar shape being fabricated as a one-step 
manufacturing process. Metal strips, layers, and/or fittings can easily be adhesively 
"molded" in the manufactured structure during the composite processing stage (e.g., 
wet lay-up, filament winding, RIM, RRIM, and so on). Adhesive joining techniques 
lend themselves to the creation of integrally designed structures as described previ- 
ously. The various adhesive joint designs (lap shear, butt tensile, scarf joints, and so 
on) were discussed in Chapter 4. 

Structural adhesives are available in various forms and types. Most common are 
the two-package epoxy resins. These formulated products are very similar to the epoxy 
matrix resins used to create the fiber-reinforced composite materials themselves. Usu- 
ally, these two-package products consist of part A, the epoxy resin prepolymer, and 
part B, the curative (such as a primary amine or a polyamide/amine). Fillers, thicken- 
ers, reactive diluents, tackifiers, and other processing aids such as silicone compounds 
to improve the moisture durability of the adhesive are added to the final formulation. 

326 8 Plastics 

These two-part adhesives are mixed just before being applied to the surfaces of the 
parts to be joined. The assembly is then placed in a compression mold, platen press, or 
vacuum bagging arrangement, where heat may be applied to consolidate the layers 
being joined and cure the adhesive. There are also some one-package paste adhesives 
that are formulated with a latent curative; the curative reacts only at high temperature. 

Another useful form of adhesive is the film adhesive. Film adhesives are used ex- 
tensively in the aerospace industry. Here, adhesives exist in the form of sheets. These 
sheets are malleable, are drapable, and can be cut using shears to the desired size and 
shape. These films are then placed between the surfaces to be joined and are cured 
under consolidation pressure and elevated temperature. Like the one-package adhe- 
sives, these adhesives are formulated with a high-temperature-reacting latent curative. 
Film adhesives, like the fiber-reinforced epoxy prepregs described earlier, must be 
stored at low temperature and kept frozen until ready to use. 

Also used in bonding composite materials are the acrylic adhesives. Acrylic ad- 
hesives having different flexibilities are available. They cure at room temperature by a 
free-radical polymerization reaction. One feature of acrylic adhesives is that cure can 
be achieved by first coating the free-radical catalyst on the surfaces to be bonded. This 
"catalyst-primed" surface can then be stored until it is ready for bonding. An uncat- 
alyzed acrylic adhesive is then coated onto the catalyst-primed surface. The surfaces 
to be joined are then mated under contact pressure and allowed to cure, undisturbed, 
at room temperature. Acrylic adhesives can produce bonds that are very oil resistant. 

Finally, it is important that the surfaces to be joined be clean and free of oils, 
greases, and loose surface material layers. This is especially necessary when joining 
composite materials to metals. Vapor degreasing, followed by a chemically alkaline 
cleaning bath, is normally used for surface treating metals prior to adhesive bonding. 

Sandwich-panel construction. Structural sandwich-panel construction consists of 
face sheets made up of fiber-reinforced laminar composite material (or metal sheet) 
adhesively bonded to both sides of a core material. This concept is illustrated in Fig- 
ure 8.34. The principle behind sandwich construction is that the core material spaces 
the facings away from the symmetric center of the panel. Therefore, in flexure, the 
faces or outer skins of the panel are in tension or compression. This construction leads 
to the reinforcement in the faces, which resists the bending of the panel. The columnar 
strength of the honeycomb core material then provides the shear and compression 
strength of this unique panel structure. Above all, the adhesive must be strong and 
have a high enough shear and peel strength to withstand these shear stresses. 

Sandwich construction leads to the use of panels that give the highest stiffness-to- 
weight ratio of any material design. Sandwich-panel construction is used extensively 
in aircraft and aerospace applications, where the core materials are generally honey- 
combed in geometric shape. Honeycomb cores can be made of thin metal (aluminum 
or titanium) or of fiber-reinforced resin sheet (e.g., thin sheet of resin-impregnated 
glass, carbon, or polyaramid mat). The manufacture of honeycomb core by the expan- 
sion process is shown in Figure 8.35. Manufacturing honeycomb core involves coating 
discrete strips of adhesive onto sheets of core material. The specially coated core ma- 
terial is then cured under compression to form a "log" or block of core material. The 

8.5 Fiber-Reinforced Polymeric Composites 


FIGURE 8.34 

Structural sandwich 
panel construction 
(Courtesy Strong, A. B. 
Fundamentals of 
Materials, Methods, 
and Applications. 
Dearborn, Michigan: 
Society of 
Engineers, 1989) 

Face sheet 

Honeycomb (Metal, composite, or paper) 

Film adhesive 

Face sheet 

log must then be cut to the desired core height and subsequently expanded to form the 
final core material. In some instances, the core material is dipped into a resin solution 
so that the core structure can be consolidated or stiffened. Another method of making 
honeycomb is the direct corrugation process. In some less demanding stiffness and 
compression applications, a rigid foam core material can be used. Rigid foam and 

FIGURE 8.35 

Manufacture of 
honeycomb core by the 
expansion process 

1 . Adhesive 
strips are 
coated onto 


2. Plies from step 1 are laid 
to form a block. 

3. Block is cured under 
heat and compression. 



u u 




1 2 3 4 5 6. 

coat index 

4. Expansion leads to 

formation of honeycomb cone. 


8 Plastics 

Kraft-paper-based honeycomb core panels are often used in truck cargo bed panels and 
in door panels. 

Painting and coating. Standard coating methods can be used for painting or coating 
fiber-reinforced composite structures. In all cases, the surface of the composite must 
be thoroughly prepared before the final coating is applied. Surface cleaning, sanding, 
abrading, filling in surface grooves/blemishes, and a solvent wipe must be carried out 
before the paint sealer and final paint finish are applied. Paint sealers and the final 
paint coating must be dried/cured at temperatures below the cure temperature of the 
composite part. Drying with infrared heaters can be troublesome as the heat location 
and temperature cannot be properly controlled using this technique. Epoxy and 
polyurethane-based surface coatings are especially useful in the painting of composite 


Kaverman, R. D. "Reinforced Plastics and Compos- 
ites." In Michael L. Berins, ed., SPI Plastics Engi- 
neering Handbook. New York: Van Nostrand 
Reinhold, 1991. 

Mayer, Rayner M. Design with Reinforced Plastics. 
Design Council, K128 Haymarket, London SWIY 
450. Bournemouth, England: Bourne Press Ltd., 

Schwartz, Mel M., ed. Composite Materials Hand- 
book, 2nd ed. New York: McGraw-Hill, 1992. 

Strong, A. B. Fundamentals of Composites Manu- 
facturing: Materials, Methods, and Applications. 
Dearborn, Michigan: Society of Manufacturing En- 
gineers, 1989. 

Review Questions 


1. What are plastics and why are they called poly- 

2. What is a monomer"? 

3. Are all polymers artificial? Give examples. 

4. Why is the strength of polymers lower than that 
of metals? 

5. Why is the electrical conductivity of polymers 
lower than that of metals? 

6. When did polymers start to gain widespread ap- 
plication and why? 

7. How are polymers classified based on their 
temperature characteristics? 

8. What is meant by chemical families of poly- 
mers? Give examples. 

9. What are the main characteristics of a thermo- 
plastic polymer? 

10. Does a thermoplastic polymer have a fixed 
melting temperature? Why? 

11. What is meant by shaping memoryl 

12. What are the main characteristics of a ther- 
mosetting polymer? 

13. How do molecules of a thermosetting polymer 
differ from those of a thermoplastic polymer? 

Chapter 8 Review Questions 


14. Compare the properties of plastics with those of 
metals. How do the differences affect the de- 
sign of plastic products? 

15. How can we have different polymers starting 
from the same monomer? 

16. List four polymers that belong to the ethenic 
group. Discuss their properties and applica- 

17. What are the main applications of polyacetals? 

18. What is cellophane and how is it produced? 

19. What is the major disadvantage of cellulose ni- 

20. What are the major applications for cellulose 

21. What is the chief limitation of nylons? 

22. What are the major characteristics of pheno- 

23. How are polyimides manufactured? 

24. List the common applications for epoxies. 

25. Discuss the properties of polyurethanes and list 
some of their applications. 

26. What property characterizes silicones? Suggest 
suitable applications to make use of that prop- 

27. Explain how natural rubber is processed. 

28. Why are additives compounded with polymers? 

29. List some fillers. Why are they added to poly- 

30. What happens when too much filler is added? 

31. How does the addition of plasticizers affect the 
properties of a polymer? 

32. List some of the lubricants used when process- 
ing polymers. 

33. What are the mechanisms for coloring poly- 

34. Are all polymers cast in the same manner? 

35. What are the design features of parts produced 
by blow molding? 

36. Using sketches, explain the injection molding 

37. What is the chief limitation of injection mold- 

38. What kinds of polymers are usually processed 
by compression molding? 

39. List some advantages of the compression mold- 
ing process. 

40. What is the main difference between compres- 
sion molding and transfer molding? 

41. Explain briefly the operating principles of rota- 
tional molding. 

42. List examples of plastic products that are man- 
ufactured by extrusion. 

43. What is the coextrusion process? Why is it used 
in industry? 

44. What are the design features of parts produced 
by thermoforming? Give examples. 

45. What are the products of the calendering 

46. What is the major problem experienced when 
machining plastics? 

47. Using sketches, explain the process of hot-plate 

48. Describe thermal staking. 

49. Explain how ultrasonics are employed in weld- 
ing and assembling plastic parts. 

50. Do all plastics render themselves suitable for 
ultrasonic welding? Explain. 

51. What are the basic components of ultrasonic 
welding equipment? 

52. Using sketches, show some designs of ultra- 
sonic-welded joints. List the characteristics of 

53. Explain the sequence of operations involved in 
open-mold processing of reinforced polymers. 

54. What are the similarities and differences be- 
tween extrusion and pultrusion of polymers? 

55. What are the design features of parts manufac- 
tured by filament winding? 


8 Plastics 

56. Explain briefly the nature of FRP composites. 

57. How can you predict the properties of a com- 
posite? Provide a quantitative equation. 

58. List some of the fibers used as inforcement in 
FRP composites. 

59. Briefly discuss the various matrix resins for 
FRP composites indicating their advantages, 
disadvantages, and limitations. 

60. Why is vacuum bagging used in the modified 
version of the wet lay-up method? 

61. What should we be careful about when using 
fiber resin prepregs? 

62. What are the advantages of sandwich panels? 

Design Pxojects__ 

The current products of a company involve dif- 
ferent fruit preserves in tin cans, each containing 
8 ounces (about 250 g). The company uses 
250,000 tin cans annually, and each costs 13 
cents. Because their machines are almost obso- 
lete and the cost of tin is rising every year, the 
company is considering replacing the tin cans 
with plastic containers. Design plastic containers 
to serve this goal, taking into account the plastic- 
processing method to be used. Also, make a fea- 
sibility study for the project. 

Design a plastic cup that has a capacity of 8 
ounces (about 250 g) of water. Assume the an- 
nual production volume is 20,000 pieces. 

Design a high-quality plastic pitcher that has a 
capacity of 32 ounces (about 1 kg) of liquid. As- 


sume the annual production volume is 15,000 

Design a wheel for a bicycle so that it can be 
produced by injection molding instead of sheet 
metal forming. The diameter is 24 inches (600 
mm), and a load of 100 pounds (about 45 kg) is 
applied, through the axle, at its center. Assume 
the annual production volume is 100,000 wheels. 

A trash container that has a capacity of 1 cubic 
foot (0.027 m 3 ) is made of sheet metal and can 
withstand an axial compressive load of 110 
pounds (50 kg). Redesign it so that it can be 
made of plastic. Assume the annual production 
volume is 20,000 pieces. 

Chapter 9 

yslcs of 

Metal Cutting 


Metal cutting can be defined as a process during which the shape and dimen- 
sions of a workpiece are changed by removing some of its material in the form 
of chips. The chips are separated from the workpiece by means of a cutting 
tool that possesses a very high hardness compared with that of the workpiece, 
as well as certain geometrical characteristics that depend upon the conditions 
of the cutting operation. Among all of the manufacturing methods, metal cut- 
ting, commonly called machining, is perhaps the most important. Forgings and 
castings are subjected to subsequent machining operations to acquire the pre- 
cise dimensions and surface finish required. Also, products can sometimes be 
manufactured by machining stock materials like bars, plates, or structural sec- 

Machining comprises a group of operations that involve seven basic chip- 
producing processes: shaping, turning, milling, drilling, sawing, broaching, and 
grinding. Although one or more of these metal-removal processes are performed 
at some stage in the manufacture of the vast majority of industrial products, the 
basis for all these processes (i.e., the mechanics of metal cutting) is yet not fully 
or perfectly understood. This is certainly not due to the lack of research but 
rather is caused by the extreme complexity of the problem. A wide variety of fac- 
tors contribute to this complexity, including the large plastic strains and high 
strain rates involved, the heat generated and high rise in temperature during ma- 
chining, and, finally, the effect of variations in tool geometry and tool material. It 
seems, therefore, realistic to try to simplify the cutting operation by eliminating 



9 Physics of Metal Cutting 


Two-dimensional cutting 
using a prismatic, 
wedge-shaped tool 


as many of the independent variables as possible and making appropriately im- 
plicit assumptions if an insight into this complicated process is to be gained. In 
fact, we are going to take this approach in discussing the cutting tools and the 
mechanics of chip formation. We are going to consider two-dimensional cutting, 
in which a prismatic, wedge-shaped tool with a straight cutting edge is employed, 
as shown in Figure 9.1, and the direction of motion of the tool (relative to the 
workpiece) is perpendicular to its straight cutting edge. In reality, such condi- 
tions resemble the case of machining a plate or the edge of a thin tube and are 
referred to as orthogonal cutting. 


Figure 9.2 clearly illustrates that the lower surface of the tool, called the flank, makes 
an angle \j/ with the newly machined surface of the workpiece. This clearance angle is 
essential for the elimination of friction between the flank and the newly machined sur- 
face. As can also be seen in Figure 9.2, there is an angle a between the upper surface, 
or face, of the tool along which chips flow and the plane perpendicular to the machined 


Tool angles in two- 
dimensional cutting 

Tool angle 


9.1 Cutting Angles 


surface of the workpiece. It is easy to realize that the angle a indirectly specifies the 
slope of the tool face. This angle is known as the rake angle and is necessary for shov- 
eling the chips formed during machining operations. The resistance to the flow of the 
removed chips depends mainly upon the value of the rake angle. As a consequence, the 
quality of the machined surface also depends on the value of the rake angle. In addi- 
tion to these two angles, there is the tool angle (or wedge angle), which is the angle 
confined between the face and the flank of the tool. Note that the algebraic sum of the 
rake, tool, and clearance angles is always equal to 90°. Therefore, it is sufficient to de- 
fine only two of these three angles. In metal-cutting practice, the rake and clearance 
angles are the ones that are defined. 

As you may expect, the recommended values for the rake and clearance angles are 
dependent upon the nature of the metal-cutting operation and the material of the work- 
piece to be machined. The choice of proper values for these two angles results in the 
following gains: 

1. Improved quality of the machined surface 

2. A decrease in the energy consumed during the machining operation (most of 
which is converted into heat) 

3. Longer tool life as a result of a decrease in the rate of tool wear because the 
elapsed heat is reduced to minimum 

Let us now consider how the mechanical properties of the workpiece material af- 
fect the optimum value of the rake and clearance angles. Generally, soft, ductile met- 
als require tools with larger positive rake angles to allow easy flow of the removed 
chips on the tool face, as shown in Figure 9.3. In addition, the higher the ductility of 
the workpiece material, the larger the tool clearance angle that is needed in order to re- 
duce the part of the tool that will sink into the workpiece (i.e., reduce the area of con- 
tact between the tool flank and the machined workpiece surface). On the other hand, 
hard, brittle materials require tools with smaller or even negative rake angles in order 
to increase the section of the tool subjected to the loading, thus enabling the tool to 
withstand the high cutting forces that result. Figure 9.4 illustrates tools having zero and 
negative rake angles required when machining hard, brittle alloys. In this case, the 
clearance angle is usually taken as smaller than that recommended when machining 
soft, ductile materials. 


Positive rake angle 
required when 
machining soft, ductile 

rake angle 



9 Physics of Metal Cutting 


Zero and negative rake 
angles required when 
machining hard, brittle 

rake angle 

rake angle. 






Mechanics of Chip Formation 

There was an early attempt by Reuleaux at the beginning of the twentieth century to 
explain the mechanics of chip formation. He established a theory that gained popular- 
ity for many years; it was based on assuming that a crack would be initiated ahead of 
the cutting edge and would propagate in a fashion similar to that of the splitting of 
wood fibers, as shown in Figure 9.5. Thanks to modern research that employed high- 
speed photography and quick stopping devices capable of freezing the cutting action, 
it was possible to gain a deeper insight into the process of chip formation. As a result, 
Reuleaux's theory collapsed and proved to be a misconception; it has been found that 
the operation of chip formation basically involves shearing of the workpiece material. 
Let us now see, step by step, how that operation takes place. 

The stages involved in chip removal are shown in Figure 9.6. When the tool is set 
at a certain depth of cut (see Figure 9.6a) and is then pushed against the workpiece, the 
cutting edge of the tool and the face start to penetrate the workpiece material. The sur- 
face layer of the material is compressed; then pressure builds up and eventually ex- 
ceeds the elastic limit of the material. As a result of the intense shear stress along the 
plane N-N, called the shear plane, plastic deformation takes place, and the material of 
the surface layer has no option but to flow along the face of the tool without being sep- 
arated from the rest of the workpiece (see Figure 9.6b). With further pushing of the 
tool, the ultimate tensile strength is exceeded, and a little piece of material (a chip) is 


misconception of the 
mechanics of chip 


9.2 Chip Formation 



Stages in chip removal: 
(a) tool set at a certain 
depth of cut set; (b) 
workpiece penetration; 
(c) chip separation 




A new chip 




separated from the workpiece by slipping along the shear plane (see Figure 9.6c). This 
sequence is repeated as long as the tool continues to be pushed against the workpiece, 
and the second, third, and subsequent chips are accordingly separated. 

Types of Chips 

The type of chip produced during metal cutting depends upon the following factors: 

1. The mechanical properties (mainly ductility) of the material being machined 

2. The geometry of the cutting tool 

3. The cutting conditions used (e.g., cutting speed) and the cross-sectional area of 
the chip 

Based on these factors, the generated chips may take one of the forms shown in Fig- 
ure 9.7. Following is a discussion of each type of chip. 

Continuous chip. When machining soft, ductile metals such as low-carbon steel, cop- 
per, and aluminum at the recommended cutting speeds (which are high), plastic flow 
predominates over shearing (i.e., plastic flow continues, and shearing of the chip never 
takes place). Consequently, the chip takes the form of a continuous, twisted ribbon (see 
Figure 9.7a). Because the energy consumed in plastically deforming the metal is even- 
tually converted into heat, coolants and lubricants must be used to remove the gener- 
ated heat and to reduce friction between the tool face and the hot, soft chip. 

Discontinuous chips. When machining hard, brittle materials such as cast iron or 
bronze, brittle failure takes place along the shear plane before any tangible plastic flow 
occurs. Consequently, the chips take the form of discontinuous segments with irregu- 
lar shape (see Figure 9.7b). As no plastic deformation is involved, there is no energy 
to be converted into heat. Also, the period of time during which a chip remains in con- 


Types of machining 
chips: (a) continuous, 
twisted ribbon; (b) 
discontinuous, irregular 
segments; (c) sheared, 
short ribbons 






Physics of Metal Cutting 

tact with the face of the tool is short, and. therefore, the heat generated due to friction 
is very small. As a result, the tool does not become hot, and lubricants and coolants are 
not required. 

Sheared chips. When machining semiductile materials with heavy cuts and at rela- 
tively low cutting speeds, the resulting sheared chips have a shape that is midway be- 
tween the segmented and the continuous chips (see Figure 9.7c). They are usually 
short, twisted ribbons that break every now and then. 

The Problem of the Built-Up Edge 

When machining highly plastic, tough metals at high cutting speeds, the amount of 
heat generated as a result of plastic deformation and friction between the chip and the 
tool is large and results in the formation of a built-up edge, as shown in Figure 9.8. The 
combination of the resulting elevated temperature with the high pressure at the tool 
face causes localized welding of some of the chip material to the tool face (see Figure 
9.8a). The welded material (chip segment) becomes an integral part of the cutting tool, 
thus changing the values of the cutting angles. This certainly increases friction, lead- 
ing to the buildup of layer upon layer of chip material. This newly formed false cut- 
ting edge (see Figure 9.8b) is referred to as the built-up edge. The cutting forces also 
increase, the built-up edge breaks down, and the fractured edges adhere to the ma- 
chined surface (see Figure 9.8c). The harmful effects of the built-up edge are increased 
tool wear and a very poorly machined workpiece. The manufacturing engineer must 
choose the proper cutting conditions to avoid the formation of a continuous chip with 
a built-up edge. 

The Cutting Ratio 

As can be seen in Figure 9.9. during a cutting operation, the workpiece material just 
ahead of the tool is subjected to compression, and, therefore, the chip thickness be- 
comes greater than the depth of cut. The ratio of t /t is called the cutting ratio (r c ) and 


Stages in the formation 
of the built-up edge: (a) 
localized welding; (b) 
false cutting edge; (c) 
flawed surface 




Broken chips 

sticking to the 

newly machined 



9.2 Chip Formation 



Geometry of a chip with 
respect to depth of cut 

can be obtained as follows: 

sin (|) 

r _*o _ t s sin $ 

t t s cos (()) - a) cos (§ - a) 


By employing trigonometry and carrying out simple mathematical manipulation, we 
can obtain the following equation: 

tan <b = 

r c cos a 
1 - r c sin a 


Equation 9.2 is employed in obtaining the value of the shear angle § when the rake 
angle a, the depth of cut, and the final thickness of the chips are known. In experi- 
mental work, the chip thickness is either measured directly with the help of a ball- 
ended micrometer or obtained from the weight of a known length of chip (of course, 
the density and the width of the chip must also be known). 

Let us now study the relationship between velocities. Considering the constancy 
of mass and assuming the width of the chip to remain constant, it is easy to see that 


V x t n = V, x t 

V t c 

In other words, 

V r = Vr r = 

V sin (j) 
cos (()) - a) 


We can now draw the velocity triangle because we know the magnitudes and directions 
of two velocities, V and V c . The shear velocity, V v , which is the velocity with which the 
metal slides along the shear plane, can then be determined. Based on the velocity tri- 
angle shown in Figure 9.10 and applying the sine rule, the following can be stated: 




sin (90 - (j» + a) sin (90 - a) sin ty 


9 Physics of Metal Cutting 

FIGURE 9.10 

Velocity triangle and 
kinematics of the chip- 
removal process 

(90 - <t> + a) 

(0 -a) 

(90 - a) 

This equation can also take the form 

V Vs _ v c 

cos ((}) - a) cos a sin <j) 


cos a 


cos ((() - a) 


Shear Strain During Chip Formation 

The value of the shear strain is an indication of the amount of deformation that the 
metal undergoes during the process of chip formation. As can be seen in Figure 9.11, 
the parallelogram abda' will take the shape abed' due to shearing. The shear strain can 
be expressed as follows: 

a'n d'n ., . 

y = h = cot (J) + tan (<|) - a) 

an an 


The shear strain rate can be obtained from Equation 9.5 as follows: 

a'n 1 d'n 1 

y = x — + x — 

an At an At 

a'd' 1 

x — 

an At 

FIGURE 9.11 

Shear strain during chip 

Rake angle 




Final shape of material 
, just after deformation 
^ y (broken line 

Original shape of 
material just before 
machining (hatched) 

9.3 Cutting Forces 





= K 

Y = 



where an is the thickness of the shear zone. Experimental results have indicated that 
the thickness of the shear zone is very small. Consequently, it can easily be concluded 
that the process of chip formation takes place at an extremely high strain rate. This 
finding is very important, especially for strain-rate-sensitive materials, where the 
strength and ductility of the material are markedly affected. 


Theory of Ernst and Merchant 

In order to simplify the problem, let us consider the two-dimensional, idealized cut- 
ting model of continuous chip formation. In this case, all the forces lie in the same 
plane and, therefore, form a coplanar system of forces. Walter Ernst and Eugene M. 
Merchant, both eminent American manufacturing scientists, based their analysis of 
this system of forces on the assumption that a chip acts as a rigid body in equilib- 
rium under the forces acting across the chip-tool interface and the shear plane. As 
Figure 9.12 shows, the cutting edge exerts a certain force upon the workpiece. The 
magnitude of that force is dependent upon many factors, such as the workpiece ma- 
terial, the conditions of cutting, and the values of the cutting angles. 

FIGURE 9.12 

Cutting force diagram 
according to Ernst and 

340 9 Physics of Metal Cutting 

By employing simple mechanics, the force can be resolved into two perpendicu- 
lar components, F c and F,. As can be seen in Figure 9.12, F c acts in the direction of 
tool travel and is referred to as the cutting component, whereas F, acts normal to that 
direction and is known as the thrust component. The resultant tool force can alterna- 
tively be resolved into another two perpendicular components, F s and F n . The first 
component, F s , acts along the shear plane and is referred to as the shearing force; the 
second component, F n , acts normal to it and causes compressive stress to act on the 
shear plane. Again, at the chip-tool interface, the components of the resultant force 
that acts on the chip are F and TV. Notice from the figure that F represents the fric- 
tion force that resists the movement of the chip as it slides over the face of the tool, 
while N is the normal force. The ratio between F and TV is actually the coefficient of 
friction at the chip-tool interface. Because each two components are perpendicular, it 
is clear from Euclidean geometry that the point of intersection of each two compo- 
nents must lie on the circumference of the circle that has the resultant force as a di- 
ameter. The cutting force diagram of Figure 9.12 lets us express F s , F n , F, and TV in 
terms F c and F, as follows: 

F s = F c cos § - F, sin (J) (9.6) 

F n = F c sin <J) + F, cos (J) (9.7) 

F = F c sin a + F t cos a (9.8) 

N = F c cos a - F, sin a (9.9) 

The preceding equations can be used to determine different unknown parameters 
that affect the cutting operations. For instance, the coefficient of friction at the chip- 
tool interface can be obtained as follows: 

F F,. sin cl + F, cos a _i 

u. = — = — - - = tan B 

TV F c cos a- F t sin a 

Dividing both the numerator and denominator by cos a, we obtain 

F, + F,tana (9>10) 

F c - F, tan a 

The shear force F s is of particular importance as it is used for obtaining the mag- 
nitude of the mean shear strength of the material along the shear plane and during the 
cutting operation. This is equal to the mean shear stress acting through the shear plane 
and can be computed as follows: 

' A, 

where A s , the area of the shear plane, equals A chip /sm ((), where A chip is the cross- 
sectional area of the chip. Therefore, 

x _ [F c cos (() - F, sin <t>]sin ()) 


9.3 Cutting Forces 341 

Experimental work has indicated that the mean shear stress, calculated from 
Equation 9.11, is constant for a given metal over a wide variation in the cutting con- 
ditions. This can be explained by the fact that the strain rate at which metal cutting 
occurs is sufficiently high to be the only factor that affects the shear strength for a 
given material. Therefore, the cutting speed, amount of strain, or temperature do not 
have any appreciable effect on the value of the mean shear stress of the metal being 

Ernst and Merchant extended their analysis and studied the relationship between 
the shear angle and the cutting conditions. They suggested that the shear angle always 
takes the value that reduces the total energy consumed in cutting to a minimum. Be- 
cause the total work done in cutting is dependent upon and is a direct function of the 
component F c of the cutting force, they developed an expression for F c in terms of (J) 
and the constant properties of the workpiece material. Next, that expression was dif- 
ferentiated with respect to <J) and then equated to zero in order to obtain the value § for 
which F c and, therefore, the energy consumed in cutting is a minimum. Following is 
the mathematical treatment of this problem. 

From Figure 9.12, we can see that 

F s = R cos (4> + p - a) (9.12) 


cos (§ + p - a) 

F s = x s A s = x s 


sin § 

R = T * Achi P x (9.13) 

sin § cos (<|) + P - a) 

Again, it can be seen from Figure 9. 1 2 that 

F c = /?cos(p-<x) (9.14) 

Hence, from Equations 9.13 and 9.14, 

F = TA hiD x cos (p - a) (9lS) 

sin § cos ((() + P - a) 

Differentiating Equation 9.15 with respect to ({) and equating the outcome to zero, we 
obtain the condition that will make F c minimal. This condition is given by the follow- 
ing equation: 

20 + P - a = ^ (9.16) 

342 9 Physics of Metal Cutting 

It was found that the theoretical value of ty obtained from Equation 9.16 agreed well 
with the experimental results when cutting polymers, but this was not the case when 
machining aluminum, copper, or steels. 

Theory of Lee and Shaffer 

The theory of American manufacturing scientists E. Lee and Bernard W. Shaffer is 
based on applying the slip-line field theory to the two-dimensional metal-cutting prob- 
lem. A further assumption is that the material behaves in a rigid, perfectly plastic man- 
ner and obeys the von Mises yield criterion and its associated flow rule. After 
constructing the slip-line field for that problem, it was not difficult for Lee and Shaf- 
fer to obtain the relationship between the cutting parameters and the shear angle. The 
result can be given by the following equation: 

<|> + f3 - a = -j (9.17) 

In fact, neither of the preceding theories quantitatively agrees with experimental re- 
sults. However, the theories yield linear relationships between ()) and ((3 - a), which is 
qualitatively in agreement with the experimental results. 

Cutting Energy 

We can see from the previous discussion that it is the component F c that determines 
the energy consumed during machining because it acts along the direction of relative 
tool travel. The power consumption P,„ (i.e., the rate of energy consumption during 
machining) can be obtained from the following equation: 

P m =F c xV (9.18) 

where V is the cutting speed. 

The rate of metal removal during machining Z m is also proportional to the cutting 
speed and can be given by 

Z m =A xV (9.19) 

where A , the cross-sectional area of the uncut chip, equals t times the width of the 
chip. Now, the energy consumed in removing a unit volume of metal can be obtained 
from Equations 9.18 and 9.19 as follows: 

n P,„ F c x V F c 

P ( = — = — = — (9.20) 

Zm A t) x V A„ 

In Equation 9.20, P c , a parameter that indicates the efficiency of the process, is 
commonly known as the specific cutting energy and also sometimes is called the 
unit horsepower. Unfortunately, the specific cutting energy for a given metal is not 
constant but rather varies considerably with the cutting conditions, as we will see 

9.4 Oblique Versus Orthogonal Cutting 



Until now, we have simplified the metal-cutting process by considering only orthogo- 
nal cutting. In this type of cutting, the cutting edge of the tool is normal to the direc- 
tion of relative tool movement, as shown in Figure 9.13a. It is actually a 
two-dimensional process in which each longitudinal section (i.e., parallel to the tool 
travel) of the tool and chip is identical to any other longitudinal section of the tool and 
chip. The cutting force is, therefore, also two-dimensional and can be resolved into two 
components, both lying within the plane of the drawing. Although this approach facil- 
itated the analysis of chip formation and the mechanics of metal cutting, it is seldom 
used in practice because it applies only when turning the end face of a thin tube in a 
direction parallel to its axis. 

The more common type (or model) of cutting used in the various machining op- 
erations is oblique cutting. In this case, the cutting edge of the tool is inclined to (i.e., 
not normal to) the relative tool travel, as can be seen in Figure 9.13b. It is a three- 
dimensional problem in which the cutting force can be resolved into three perpendic- 
ular components, as indicated in Figure 9.14. The magnitudes of these components can 
be measured by means of a special apparatus that is mounted either in the workholder 
or toolholder and is known as a dynamometer. As you may expect, the tool geometry 

FIGURE 9.13 

Types of cutting: (a) 
orthogonal; (b) oblique 



FIGURE 9.14 

Components of the 
cutting force 


344 9 Physics of Metal Cutting 

is rather complicated and will be discussed later. For now, let us see the effect of each 
of the cutting force components on the oblique cutting operations. 

Forces in Oblique Cutting 

Following is a discussion of the three components referred to as F c , F f , and F r in Fig- 
ure 9.14: 

1. F c is the cutting force and acts in the direction where the cutting action takes place. 
It is the highest of the three components and results in 99 percent of the energy con- 
sumed during the process. The horsepower due to this force, hp c , can be given by 
the following equation: 

F c xV c 

hp -=l^ <9 - 21) 

In Equation 9.21, F c is in pounds and V c is in feet per minute. Consequently, the 
appropriate conversion factors must be used if the horsepower is to be obtained in 
SI units. 

2. Ff is the feed force (or longitudinal force in turning). The term feed means the 
movement of the tool to regenerate the cutting path in order to obtain the machined 
surface. This force amounts to only about 40 percent of the cutting force. The 
horsepower required to feed the tool, hp f , can be given as follows: 

F f x V f 

hpf = t?, 7- (9.22) 

Ff 550x60 

The horsepower given by Equation 9.22 amounts to only 1 percent of the total 
power consumed during cutting. 

3. F r is the thrust force (or radial force in turning) and acts in the direction of the 
depth of cutting. This force is the smallest of the three components and amounts to 
only 20 percent of the cutting force or, in other words, 50 percent of the feed force. 
This component does not result in any power consumption as there is no tool move- 
ment along the direction of the depth of cut. 

These components of the cutting force are measured only in scientific metal- 
cutting research. The manufacturing engineer is, however, interested in determining 
beforehand the motor horsepower required to perform a certain job in order to be able 
to choose the right machine for that job. Therefore, use is made of the concept of unit 
horsepower, which was mentioned previously. Experimentally obtained values of unit 
horsepower for various common materials are compiled in tables ready for use. The 
total cutting horsepower can be obtained from the following equation: 

hp c = unit hp x rate of metal removal x correction factor (9.23) 

where the rate of metal removal is in cubic inches per minute and the correction fac- 
tor is introduced to account for the tool geometry and and the variation in feed. 

9.4 Oblique Versus Orthogonal Cutting 


Table 9.1 indicates the unit horsepower values for various ferrous metals and al- 
loys having different hardness numbers. Table 9.2 provides the unit horsepower values 
for nonferrous metals and alloys. Figure 9.15a through c indicates the different correc- 
tion factors for the unit horsepower to account for variations in the cutting conditions. 

The cutting horsepower is not of practical importance by itself. Its significance is 
that it is used in computing the motor horsepower. Obviously, the motor horsepower 

TABLE 9.1 

Unit horsepower values for ferrous metals and alloys 

Brinnei Hardness Number 


Metals and 







































































































































Plain cast iron 







Alloy cast iron 







Malleable iron 







Cast steel 







Source: Turning Handbook of High-Efficiency Metal Cutting, 7950, courtesy Carboloy Inc., a Seco Tools Company. 


9 Physics of Metal Cutting 

TABLE 9.2 

Unit horsepower values 
for nonferrous metals 
and alloys 

Nonferrous Metals and Alloys 


Unit Horsepower 








Free machining 















Hard (rolled) 





Zinc alloy 

(die cast) 


Source: Turning Handbook of High-Efficiency Metal Cutting, 1980, courtesy Carboloy Inc., A Seco Tools 

has to be higher than the cutting horsepower as some power is lost in overcoming fric- 
tion and inertia of the moving parts. The following equation can be used for calculat- 
ing the motor horsepower: 

hp m = hp c x 



where r\ is the machine efficiency, which can be taken from Table 9.3. 

The cutting horsepower is used not only in calculating the motor horsepower 
but also for giving a fair estimate of the cutting force component F c by using Equa- 
tion 9.21. This force is very important when studying the vibrations associated with 
metal cutting, as we will see later. The following example illustrates how to estimate 
the cutting force component. 

Example of Estimating Cutting 
Force Component 

During a turning operation, the metal-removal rate (M.R.R.) was found to be 3.6 cubic 
inches per minute. Following are other data of the process: 

Cutting speed: 
Undeformed chip thickness: 
Tool character: 

Estimate the cutting force component F c 

ANSI 1055, HB 250 

300 feet per minute 

0.01 inch 

0-7-7-7-15-15-1/32 (see Section 9.5) 

9.4 Oblique Versus Orthogonal Cutting 


FIGURE 9.15 

Different correction 
factors to account for 
variations in the cutting 
conditions: (a) cutting 
speed; (b) chip 
thickness; (c) rake 
angle (Source: Turning 
Handbook of High- 
Efficiency Metal 
Cutting, 1980, courtesy 
Carboloy Inc., A Seco 
Tools Company) 


200 400 600 800 
Cutting speed (SFPM) 


0.0010.002 0.004 0.010 0.020 0.040 0.100 
Undeformed chip thickness (in.) 


c r 














-20 -10 +10 +20 
True rake angle 

Here is the solution to this example problem: 

spindle hp = M.R.R. x unit hp x correction factor 

The correction factor because of the cutting speed is 0.8, and the correction factor be- 
cause of the undeformed chip thickness is 1 . The true rake angle is 

tan a,™ = cos 15° tan 7° + sin 15° tan 

TABLE 9.3 



Typical overall machine 


Efficiency (%) 

tool efficiencies (except 

milling machines) 

Direct-spindle drive 


One-belt drive 


Two-belt drive 


Geared head 


Source: Turning Handbook of High-Efficiency Metal 
Cutting, 1980, courtesy Carboloy Inc., A Seco Tools 


Physics of Metal Cutting 

where a true = 6° (see Section 9.5). The correction factor because of the true rake angle 
is 0.83, and the unit hp is 0.8 from Table 9.1. Thus, 

spindle hp = 3.6 x 0.8 x 0.8 x 1 x 0.83 = 1.9 hp 

F r x 300 ft/min. 

hp c = 


550 x 60 

„ 1.9x550x60 

F c = — = 209 pounds 

300 F 

Note that the undeformed chip thickness equals feed (inches per revolution) times the 
cosine of the side cutting-edge angle. 


Basic Geometry 

In order for a tool to cut a material, it must have two important characteristics: First, it 
must be harder than that material, and, second, it must possess certain geometrical 
characteristics. The cutting tool geometry differs for different machining operations. 
Nevertheless, it is always a matter of rake and clearance angles. Therefore, we are 
going to limit our discussion, at the moment, to single-point tools for the sake of sim- 
plicity. Other types of tools will be considered when we cover the various machining 

As can be seen in Figure 9.16, the geometry of a single-point cutting tool can be 
adequately described by six cutting angles. These can be shown more clearly by pro- 
jecting them on three perpendicular planes using orthogonal projection, as is done in 
Figure 9.17. Let us now consider the definition of each of the six angles. 

Side cutting-edge angle. The side cutting-edge angle (SCEA) is usually referred to as 
the lead angle. It is the angle enclosed between the side cutting edge and the longitu- 
dinal direction of the tool. The value of this angle varies between 0° and 90°, depend- 
ing upon the machinability, rigidity, and, sometimes, the shape of the workpiece (e.g., 

FIGURE 9.16 

Geometry of a single- 
point cutting tool 

End relief 

9.5 Cutting Tools 


FIGURE 9.17 

Orthogonal projection 
of the cutting angles of 
a single-point tool and 
tool character 

ECEA 20° 

Nose radius 

SCEA 15° 
Top view 

Side rake 8° 

Back rake 2° 

Side relief 6 

Side view 
(dotted lines are not shown) 

[* End relief 6° 
Front view 

Tool character 
2° 8° 6° 6 

Back rake- 
Side rake- 

End relief- 
Side relief - 

20° 15° ±\n. 


Nose radius 

a 90° shoulder must be produced by a 0° SCEA). As this angle increases from 0° to 
15°, the power consumption during cutting decreases. However, there is a limit for in- 
creasing the SCEA, beyond which excessive vibrations take place because of the large 
tool-workpiece interface. On the other hand, if the angle were taken as 0°, the full cut- 
ting edge would start to cut the workpiece at once, causing an initial shock. Usually, 
the recommended value for the lead angle should range between 15° and 30°. 

End cutting-edge angle. The end cutting-edge angle (ECEA) serves to eliminate rub- 
bing between the end cutting edge and the machined surface of the workpiece. Al- 
though this angle takes values in the range of 5° to 30°, commonly recommended 
values are 8° to 15°. 

Side relief and end relief angles. Side and end relief angles serve to eliminate rub- 
bing between the workpiece and the side and end flank, respectively. Usually, the value 
of each of these angles ranges between 5° and 15°. 

Back and side rake angles. Back and side rake angles determine the direction of flow 
of the chips onto the face of the tool. Rake angles can be positive, negative, or zero. It 
is the side rake angle that has the dominant influence on cutting. Its value usually 
varies between 0° and 15°, whereas the back rake angle is usually taken as 0°. 

Another useful term in metal cutting is the true rake angle, which is confined be- 
tween the line of major inclination within the face of the tool and a horizontal plane. 
It determines the actual flow of chips across the face of the tool and can be obtained 

350 9 Physics of Metal Cutting 

from the following equation: 

true rake angle = tan" '(tan a sin X + tan f3 cos X) (9.25) 

where: a is the back rake angle 
(3 is the side rake angle 
X is the lead angle (SCEA) 

As previously mentioned, the true rake angle has a marked effect on the unit horse- 
power for a given workpiece material, and a correction factor has to be used when cal- 
culating the power in order to account for variations in the true rake angle. 

Tool character. The tool angles are usually specified by a standard abbreviation sys- 
tem called the tool character, or the tool signature. As also illustrated in Figure 9.17, 
the tool angles are always given in a certain order: back rake, side rake, end relief, side 
relief, ECEA, and SCEA, followed by the nose radius of the tool. 

Cutting Tool Materials 

Cutting tools must possess certain mechanical properties in order to function ade- 
quately during the cutting operations. These properties include high hardness and the 
ability to retain it even at the elevated temperatures generated during cutting, as well 
as toughness, creep and abrasion resistance, and the ability to withstand high bearing 
pressures. Cutting materials differ in the degree to which they possess each of these 
mechanical properties. Therefore, a cutting material is selected to suit the cutting con- 
ditions (i.e., the workpiece material, cutting speed or production rate, coolants used, 
and so on). Following is a survey of the commonly used cutting tool materials. 

Plain-carbon steel. Plain-carbon steel contains from 0.8 to 1.4 percent carbon, has no 
additives, and is subjected to heat treatment to increase its hardness. Plain-carbon steel 
is suitable only when making hand tools or when soft metals are machined at low cut- 
ting speeds as it cannot retain its hardness at temperatures above 600°F (300°C) due to 
tempering action. 

Alloy steel. The carbon content of alloy steel is similar to that of plain-carbon steel, 
but it contains alloying elements (in limited amounts). Tools made of alloy steel must 
be heat treated and are used only when machining is carried out at low cutting speeds. 
The temperature generated as a result of cutting should not exceed 600°F (300°C) to 
avoid any tempering action. 

High-speed steel. High-speed steel (HSS) is a kind of alloy steel that contains a cer- 
tain percentage of alloying elements, such as tungsten (18 percent), chromium (4 per- 
cent), molybdenum, vanadium, and cobalt. High-speed steel is heat treated by heating 
(at two stages), cooling by employing a stream of air, and then tempering it. Tools 
made of HSS can retain their hardness at elevated temperatures up to 1 100°F (600°C). 
These tools are used when relatively high cutting speeds are required. Single-point 
tools, twist drills, and milling cutters are generally made of high-speed steel, except 
when these tools are required for high-production machining. 

9.5 Cutting Tools 351 

Cast hard alloys. Cast hard alloys can be either ferrous or nonferrous and contain 
about 3 percent carbon, which, in turn, reacts with the metals to form very hard car- 
bides. The carbides retain their hardness even at a temperature of about 1650°F 
(900°C). Because such a material cannot be worked or machined, it is cast in ceramic 
molds to take the form of tips that are mounted onto holders by brazing or by being 
mechanically fastened. 

Sintered cemented-carbide tips. Sintered cemented carbide was developed to elimi- 
nate the main disadvantage of the hard cast alloys: brittleness. Originally, the compo- 
sition of this material involved about 82 percent very hard tungsten carbide particles 
and 18 percent cobalt as a binder. Sintered cemented carbides are always molded to 
shape by the powder metallurgy technique (i.e., pressing and sintering, as was ex- 
plained in Chapter 7). As it is impossible to manufacture the entire tool out of ce- 
mented carbide because of the strength consideration, only tips are made of this 
material; these tips are brazed or mechanically fastened to steel shanks that have the 
required cutting angles. 

Cemented carbides used to be referred to as Widia, taken from the German ex- 
pression "Wie Diamant," meaning diamondlike, because they possess extremely high 
hardness, reaching about 90 Re, and they retain such hardness even at temperatures of 
up to 1850°F (1000°C). Recent developments involve employing combinations of 
tungsten, titanium, and tantalum carbides with cobalt or nickel alloy as binders. The re- 
sult is characterized by its low coefficient of friction and high abrasion resistance. 
Tools with cemented-carbide tips are recommended whenever the cutting speeds re- 
quired or the feed rates are high and are, therefore, commonly used in mass produc- 
tion. Recently, carbide tips have been coated with nitrites or oxides to increase their 
wear resistance and service life. 

Ceramic tips. Ceramic tips consist basically of very fine alumina powder, A1 2 3 , 
which is molded by pressing and sintering. Ceramics have almost the same hardness 
as cemented carbides, but they can retain that hardness up to a temperature of 2200°F 
(1100°C) and have a very low coefficient of thermal conductivity. Such properties 
allow for cutting to be performed at speeds that range from two to three times the cut- 
ting speed used when carbide tips are employed. Ceramic tips are also characterized by 
their superior resistance to wear and to the formation of crater cavities. They require 
no coolants. Their toughness and bending strength are low, which must be added to 
their sensitivity to creep loading and vibration. Therefore, ceramic tips are recom- 
mended only for finishing operations (small depth of cut) at extremely high cutting 
speeds of up to 180 feet per minute (600 m/min.). Following are the three common 
types of ceramic tips: 

1. Oxide tips, consisting mainly of aluminum oxide, have a white color with some 
pink or yellow tint. 

2. Cermet tips, including alumina and some metals such as titanium or molybdenum, 
are dark gray in color. 

3. Tips that consist of both oxides as well as carbides are black in color. 

352 9 Physics of Metal Cutting 

Ceramic tips should not be used for machining aluminum because of their affinity to 

Diamond. Diamond pieces are fixed to steel shanks and are used in precision cutting 
operations. They are recommended for machining aluminum, magnesium, titanium, 
bronze, rubber, and polymer. When machining metallic materials, a mirror finish can 
be obtained. 

Tool Wear 

There are two interrelated causes for tool wear: mechanical abrasion and thermal ero- 
sion. Although these two actions take place simultaneously, the role of each varies for 
various cutting conditions. Mechanical wear is dominant when low cutting speeds are 
used or when the workpiece possesses high machinability. Thermal wear prevails when 
high cutting speeds are used with workpieces having low machinability. Thermal wear 
is due to diffusion, oxidation, and the fact that the mechanical properties of the tool 
change as a result of the high temperature generated during the cutting operation. 

The face of the cutting tool is subjected to friction caused by the fast relative 
motion of the generated chips onto its surface. Similarly, the flanks are also subjected 
to friction as a result of rubbing by the workpiece. Although the tool is harder than 
the workpiece, friction and wear will take place and will not be evenly distributed 
over the face of the tool. Wear is localized in the vicinity of the cutting edge and re- 
sults in the formation of a crater. There are different kinds of tool wear: 

1. Flank wear 

2. Wear of the face that comes in contact with the removed chip 

3. Wear of the cutting edge itself 

4. Wear of the nose 

5. Wear and formation of a crater 

6. Cracks in the cutting edges occurring during interrupted machining operations 
such as millings 

Tool Life 

Tool life is defined as the length of actual machining time beginning at the moment 
when a just-ground tool is used and ending at the moment when the machining opera- 
tion is stopped because of the poor performance of that tool. Different criteria can be 
used to judge the moment at which the machining operation should be stopped. It is 
common to consider the tool life as over when the flank wear reaches a certain amount 
(measured as the length along the surface generated due to abrasion starting from the 
tip). This maximum permissible flank wear is taken as 0.062 inch (1.58 mm) in the 
case of high-speed steel tools and 0.03 inch (0.76 mm) for carbide tools. 

The tool life is affected by several variables, the important ones being cutting 
speed, feed, and the coolants used. The effect of these variables can be determined ex- 
perimentally and then represented graphically for practical use. It was found by Fred- 
erick W. Taylor that the relationship between tool life and cutting speed is exponential. 
It can, therefore, be plotted on a logarithmic scale so that it takes the form of a straight 

9.6 Machinability 


FIGURE 9.18 

Relationship between 
tool life and cutting 
speed on a log-log 

Tool life, min. (log scale) 

line, as shown in Figure 9.18. In fact, this was the basis for establishing an empirical 
formula that correlates tool life with cutting speed. A correction factor is also intro- 
duced into the formula to account for the effects of other variables. The original for- 
mula had the following form: 

VT n = c 


where: n is a constant that depends upon the tool material (0. 1 for HSS, 0.20 for 
carbides, and 0.5 for ceramic tools) 

c is a constant that depends upon the cutting conditions (e.g., feed) 

T is the tool life measured in minutes 

V is the cutting speed in feet per minute 

Equation 9.26 is very useful in obtaining the tool life for any cutting speed if the tool 
life is known at any other cutting speed. 


Machinability Defined 

Machinability is a property characterizing the material of the workpiece: It is the ease 
with which that material can be machined. In order to express machinability in a quan- 
titative manner, one of the following methods is used: 

1. The maximum possible rate of chip removal 

2. Surface finish of the machined workpiece 

3. Tool life 

4. Energy required to accomplish the cutting operation 

It is clear that the tool life is the most important of these criteria as it plays an impor- 
tant role in maximizing the production while minimizing the production cost. More- 
over, criteria such as surface finish and machining precision depend upon many 
factors, such as the sharpness of the cutting edge, the rigidity of the tool, and the pos- 


Physics of Metal Cutting 

TABLE 9.4 

Machinability indices 
for some metals and 
alloys (using carbide 


Metal or Alloy 

Index (%) 

Steel SAE 1020 (annealed) 


Steel SAE A2340 


Cast iron 


Stainless steel 18-8 (austenitic) 


Tool steel (low tungsten, chrome, and carbon) 






Aluminum alloy 

300 and above 

sibility of formation of a built-up edge. As a consequence, it is the tool life that is most 
suitable as a criterion of machinability. 

Machinability Index 

Because machinability cannot be expressed in an absolute manner, it is appropriate to 
take a highly machinable metal as a reference and express the machinability of any 
other ferrous metal as a percentage of that of the reference metal. The reference metal 
chosen was steel SAE-AISI 1112 because of its superior machinability, which exceeds 
that for any other steel. Such steel is usually referred to as free cutting steel. The 
machinability index can now be given: 

machinability index = 

cutting speed of metal for tool life of 20 minutes 
cutting speed of steel SAE 1 1 1 2 for tool life of 20 minutes 

x 100 (9.27) 

Table 9.4 indicates the machinability index for some commonly used metals and 


Necessary Characteristics 

As previously mentioned, the process of metal cutting results in the generation of a 
large amount of heat and a localized increase in the temperature of the cutting tool. 
This effect is particularly evident when machining ductile metals. Accordingly, 
coolants are required to remove any generated heat, to lower the temperature of the 
cutting tool, and, consequently, to increase the tool service life. In order to fulfill such 
conditions and function properly, a cutting fluid must possess certain characteristics: 

1. The cutting fluid must possess suitable chemical properties (i.e., to be appropriate 
from the point of view of chemistry), must not react with the workpiece material 
or cause corrosion in any component of the machine tool, and should not promote 

9.7 Cutting Fluids 355 

the formation of rust or spoil the lubricating oil of the machine bearing and slides 
whenever it comes in contact with that oil. 

2. The cutting fluid must be chemically stable (i.e., must not change its properties 
with time). 

3. No poisonous gases or fumes should evolve during machining so that there is no 
possibility of problems regarding the safety or health of the workers. 

4. The lubricating and cooling properties of the cutting fluid must be superior. 

5. The fluid used should be cheap and should be recycled by a simple filtration 

Types of Cutting Fluids 

The following discussion involves the different kinds of cutting fluids that are used in 
industry to satisfy the preceding requirements. 

Pure oils. Mineral oils such as kerosene or polar organic oils such as sperm oil, lin- 
seed oil, or turpentine can be used as cutting fluids. The application of pure mineral 
oils is permissible only when machining metals with high machinability, such as free 
cutting steel, brass, and aluminum. This is a consequence of their poor lubricating and 
cooling properties. Although the polar organic oils possess good lubricating and cool- 
ing properties, they are prone to oxidation, give off unpleasant odors, and tend to gum. 

Mixed oils. Mineral oils are mixed with polar organic oils to obtain the advantages of 
both constituents. In some cases, sulfur or chlorine is added to enable the lubricant to 
adhere to the tool face, giving a film of lubricant that is tougher and more stable. The 
oils are then referred to as sulfurized or chlorinated oils. The chlorinated oils have the 
disadvantage of the possible emission of chlorine gas during the machining operation. 

Soluble oils. Soluble oils are sometimes called water-miscible fluids or emulsifiable 
oils. By blending oil with water and some emulsifying agents, soapy or milky mix- 
tures can be obtained. These liquids have superior cooling properties and are recom- 
mended for machining operations requiring high speeds and low pressures. 
Sometimes, extreme-pressure additives are blended with the mixtures to produce 
emulsions with superior lubricating properties. 

Water solutions. A solution of sodium nitrate and trinolamine in water can be em- 
ployed as a cutting fluid. Caustic soda is also used, provided that the concentration 
does not exceed 5 percent. If the concentration of the solution exceeds this limit, the 
paint of the machine and the lubricating oil of the slides may be affected. 

Synthetic fluids. Synthetic fluids can be diluted with water to give a mixture that 
varies in appearance from clear to translucent. Extreme-pressure additives like sulfur 
or chlorine can be added to the mixture so that it can be used for difficult machining 

356 9 Physics of Metal Cutting 


When we feel cold in winter, our jaws and teeth may start to chatter. A similar phe- 
nomenon occurs when the cutting tool and workpiece are exposed to certain unfavor- 
able cutting conditions and dynamic characteristics of the machine tool structure. The 
analysis of this chatter phenomenon is an extremely complex task. However, thanks to 
the work of the late Professor Stephen A. Tobias of the University of Birmingham in 
England, we are able to understand how vibrations of the cutting tool initiate and how 
they can be minimized. Left without remedy, these vibrations result in breakage of the 
cutting tool (especially if it is ceramic or carbide) and poor surface quality. They may 
also cause breakage of the entire machine tool. Two basic types of vibrations are gen- 
erated during machining: forced vibrations and self-excited vibrations. 

Forced vibrations take place as a result of periodic force applied within the ma- 
chine tool structure. This force can be due to an imbalance in any of the machine tool 
components or interrupted cutting action, such as milling, in which there is a periodic 
engagement and disengagement between the cutting edges and the workpiece. The fre- 
quency of these forced vibrations must not be allowed to come close to the natural fre- 
quency of the machine tool system or any of its components; otherwise, resonance 
(vibrations with extremely high amplitude) takes place. The remedy in this case is to 
try to identify any possible source for the imbalance of the machine tool components 
and eliminate it. In milling machines, the stiffness and the damping characteristics of 
the machine tool are controlled so as to keep the forcing frequency away from the nat- 
ural frequency of any component and/or the natural frequency of the system. 

Self-excited vibrations, or chatter, occur when an unexpected disturbing force, 
such as a hard spot in the workpiece material or sticking friction at the chip-tool inter- 
face, causes the cutting tool to vibrate at a frequency near the natural frequency of the 
machine tool. As a result, resonance takes place, and a minimum excitation produces 
an extremely large amplitude. Such conditions drastically reduce tool life, result in 
poor surface quality, and may cause damage to either the workpiece or the machine 
tool or both. This unfavorable condition can be eliminated, or at least reduced, by con- 
trolling the stiffness and the damping characteristics of the system. This is usually 
achieved by selecting the proper material for the machine bed (cast iron has better 
damping characteristics than steel), by employing dry-bolted joints as energy dissipa- 
tors where the vibration energy is absorbed in friction, or by using external dampers or 
absorbers. Advanced research carried out at the University of Birmingham in England 
indicated the potentials of employing layers of composites as a means to safeguard 
against the occurrence of chatter. 


Our goal now is to find out the operating conditions (mainly the cutting speed) that 
maximize the metal-removal rate or the tool life. These two variables are in opposition 
to each other; a higher metal-removal rate results in a shorter tool life. Therefore, some 

9.9 Economics of Metal Cutting 


FIGURE 9.19 

Relationship between 
cost per piece and 
cutting speed 


Optimum cutting 

speed for minimum 


Cutting speed 

trade-off or balance must be made in order to achieve either minimum machining cost 
per piece or maximum production rate, whichever is necessitated by the production 


Figure 9.19 indicates how to construct the relationship between the cutting 
speed and the total cost per piece for a simple turning operation. The total cost is 
composed of four components: machining cost, idle-time (nonproductive) cost, tool 
cost, and tool-change cost. An increase in cutting speed obviously results in a re- 
duction in machining time and, therefore, lower machining cost. This is accompa- 
nied by a reduction in tool life, thus increasing tool and tool-change costs. As can 
be seen in Figure 9.19, the curve of the cost per piece versus the cutting speed has 
a minimum that corresponds to the optimum cutting speed for the minimum cost 

per piece. 

The relationship between the production time per piece and the cutting speed can 
be constructed in the same manner, as shown in Figure 9.20. There is also a minimum 
for this curve that corresponds to the optimum cutting speed for the maximum pro- 
ductivity (minimum time per piece). Usually, this value is higher than the maximum 
economy speed given in Figure 9.19. Obviously, a cutting speed between these two 
limits (and depending upon the goals to be achieved) is recommended. 


9 Physics of Metal Cutting 

FIGURE 9.20 

Relationship between 
production time per 
piece and cutting speed 

Tool -change 
time per piece 

Idle time 

.Optimum cutting speed for 
j maximum production 

Cutting speed 

Review Questions 

1. How can the complex process of metal cutting 
be approached? 

2. Define the rake angle and the clearance angle 
in two-dimensional cutting. 

3. Why are the angles in Question 2 required? 

4. What is the upper surface of the tool called? 

5. What is the lower surface of the tool called? 

6. What are the cutting variables that affect the 
values of the rake and clearance angles? 

7. List some drawbacks if the cutting angles are 
not properly chosen. 

8. When should the rake angle be taken as a posi- 
tive value? 

9. When should the rake angle be taken as a nega- 
tive value? 

10. Can orthogonal cutting actually take place? 

11. Use sketches to explain the stages involved in 
the formation of chips during machining. 

12. Use sketches to illustrate the different types of 
machining chips and explain when and why we 
can expect to have each of these types. 

13. Explain the stages involved in the formation of 
the built-up edge. 

14. Does the built-up edge have useful or harmful 

15. What is meant by the shear angle? 

16. What is meant by the cutting ratio? 

17. Derive an expression for the shear strain that 
takes place during orthogonal cutting. 

18. Draw a sketch of the cutting force diagram pro- 
posed by Ernst and Merchant. 

19. How can the relationship between the shear and 
rake angles be expressed according to Ernst and 

20. On what basis have Lee and Shaffer developed 
their theory? 

Chapter 9 Problems 






Derive an expression for the specific energy 
during two-dimensional cutting. 

Illustrate the difference between orthogonal and 
oblique cutting. 

What are the components of the cutting force in 
oblique cutting? How do you compare their 
magnitudes with each other? 

Define the unit horsepower. 

Describe fully the geometry of single-point cut- 
ting tools. 

Explain the effect of each of the cutting angles 
in oblique cutting on the mechanics of the 

List the different cutting tool materials and enu- 
merate the advantages, disadvantages, and ap- 
plications of each. 

What are the two main causes for tool wear? 

List the different kinds of tool wear. 

30. Define tool life. 

31. What is the relationship between tool life and 
cutting speed? 

32. Define machinability and explain how it is 
quantitatively expressed by the machinability 

33. What are the necessary characteristics of cut- 
ting fluids? 

34. List the different types of cutting fluids and 
provide the advantages and limitations of each. 

35. What are the causes for forced vibrations dur- 
ing machining? 

36. How can forced vibrations be minimized? 

37. What is chatter and why does it occur? 

38. How can we eliminate chatter? 

39. What trouble can vibrations cause during ma- 

40. Use sketches to explain how the value of 
the optimum cutting speed can be obtained 
for maximum economy and for maximum 





In a turning operation, the diameter of the work- 
piece is 2 inches (50 mm), and it rotates at 360 
revolutions per minute. How long will a carbide 
tool last (n = 0.3) under such conditions if an 
identical carbide tool lasted for 1 minute when 
used at 1000 feet per minute (305.0 m/min.)? 

Determine the increase in the tool life of a car- 
bide tip as a result of a decrease in the cutting 
speed of 25, 50, and 75 percent. 
When turning a thin tube at its edge, the follow- 
ing conditions were observed: 

Depth of cut: 
Chip thickness: 
Back rake angle: 
Cutting speed: 

0.125 inch 

0.15 inch 


300 ft/min. 

Calculate the 

a. Cutting ratio 

b. Shear angle 

c. Chip velocity 

A geared-head lathe is employed for machining 
steel AISI 1055, BHN 250. The cutting speed is 
400 feet per minute, and the rate of metal re- 
moval is 2.4 cubic inches per minute. If the tool 
used has the character 0-7-7-7-15-15-1/32, esti- 
mate the following: 

The energy consumed in machining per unit 


The power required at the motor 

The tangential component of the cutting 



Physics of Metal Cutting 

Neglect the correction factor for the undeformed 
chip thickness. 

5. A 5-hp, 2-V, belt-driven lathe is to be used for 
machining brass under the following conditions: 

Cutting speed: 

Rate of metal removal: 

600 ft/min. 
7.2 in.7min 

SCEA of the tool: 


Neglect the effect of chip thickness. Does this 
lathe have enough power for the required job? 

Design Prpiecl„ 


Prepare a computer program that determines the optimum cutting speed that results in 
maximum productivity. The program should be interactive, the input being workpiece 
material, tool material, and depth of cut. Assume the time for changing the tool is 60 
seconds and the time to return the tool to the beginning of the cut is 20 seconds. Take 
the workpiece material to be 

a. Steel 1020 

b. Brass 

c. Aluminum 

d. Stainless steel 

Chapter 10 



This chapter will focus on the technological aspects of the different machining 
operations, as well as the design features of the various machine tools em- 
ployed to perform those operations. In addition, the different shapes and 
geometries produced by each operation, the tools used, and the work-holding 
devices will be covered. Special attention will be given to the required workshop 
calculations that are aimed at estimating machining parameters such as cut- 
ting speeds and feeds, metal-removal rate, and machining time. 

Machine tools are designed to drive the cutting tool in order to produce the 
desired machined surface. For such a goal to be accomplished, a machine tool 
must include appropriate elements and mechanisms capable of generating the 
following motions: 

1. A relative motion between the cutting tool and the workpiece in the direc- 
tion of cutting 

2. A motion that enables the cutting tool to penetrate into the workpiece until 
the desired depth of cut is achieved 

3. A feed motion that repeats the cutting action every round or every stroke 
to ensure continuation of the cutting operation 


362 10 Machining of Metals 


The Lathe and Its Construction 

A lathe is a machine tool used for producing surfaces of revolution and flat edges. 
Based on their purpose, construction, number of tools that can simultaneously be 
mounted, and degree of automation, lathes, or more accurately, lathe-type machine 
tools, can be classified as follows: 

1. Engine lathes 

2. Toolroom lathes 

3. Turret lathes 

4. Vertical turning and boring mills 

5. Automatic lathes 

6. Special-purpose lathes 

In spite of the diversity of lathe-type machine tools, there are common features with 
respect to construction and principles of operation. These features can be illustrated by 
considering the commonly used representative type, the engine lathe, which is shown 
in Figure 10.1. Following is a description of each of the main elements of an engine 

Lathe bed. The lathe bed is the main frame, a horizontal beam on two vertical sup- 
ports. It is usually made of gray or nodular cast iron to damp vibrations and is made 
by casting. It has guideways that allow the carriage to slide easily lengthwise. The 
height of the lathe bed should be such that the technician can do his or her job easily 
and comfortably. 

Headstock. The headstock assembly is fixed at the left-hand side of the lathe bed and 
includes the spindle, whose axis is parallel to the guideways (the slide surface of the 
bed). The spindle is driven through the gearbox, which is housed within the headstock. 
The function of the gearbox is to provide a number of different spindle speeds (usually 
6 to 18 speeds). Some modern lathes have headstocks with infinitely variable spindle 
speeds and that employ frictional, electrical, or hydraulic drives. 

The spindle is always hollow (i.e., it has a through hole extending lengthwise). 
Bar stocks can be fed through the hole if continuous production is adopted. Also, the 
hole has a tapered surface to allow the mounting of a plain lathe center, such as the one 
shown in Figure 10.2. It is made of hardened tool steel. The part of the lathe center that 
fits into the spindle hole has a Morse taper, while the other part of the center is coni- 
cal with a 60° apex angle. As explained later, lathe centers are used for mounting long 
workpieces. The outer surface of the spindle is threaded to allow the mounting of a 
chuck, a faceplate, or the like. 

Tailstock. The tailstock assembly consists basically of three parts: its lower base, an 
intermediate part, and the quill. The lower base is a casting that can slide on the lathe 

10.1 Turning Operations 


FIGURE 10.1 

An engine lathe 
(Courtesy of Clausing 
Industrial, Inc., 
Kalamazoo, Michigan) 

Headstock assembly 

Tool post 
Compound rest 

Tailstock quill 

Tailstock assembly 
Lead screw 
Feed rod 

bed along the guideways, and it has a clamping device so that the entire tailstock can 
be locked at any desired location, depending upon the length of the workpiece. The in- 
termediate part is a casting that can be moved transversely so that the axis of the tail- 
stock can be aligned with that of the headstock. The third part, called the quill, is a 
hardened steel tube that can be moved longitudinally in and out of the intermediate 
part as required. This is achieved through the use of a handwheel and a screw, around 
which a nut fixed to the quill is engaged. The hole in the open side of the quill is ta- 
pered to allow the mounting of lathe centers or other tools like twist drills or boring 
bars. The quill can be locked at any point along its travel path by means of a clamping 

Carriage. The main function of the carriage is to mount the cutting tools and gener- 
ate longitudinal and /or cross feeds. It is actually an H-shaped block that slides on the 
lathe bed between the headstock and tailstock while being guided by the V-shaped 

FIGURE 10.2 

A plain lathe center 


364 10 Machining of Metals 

guideways of the bed. The carriage can be moved either manually or mechanically by 
means of the apron and either the feed rod or the lead screw. 

The apron is attached to the saddle of the carriage and serves to convert the rotary 
motion of the feed rod (or lead screw) into linear longitudinal motion of the carriage 
and, accordingly, the cutting tool (i.e., it generates the axial feed). The apron also pro- 
vides powered motion for the cross slide located on the carriage. Usually, the tool post 
is mounted on the compound rest, which is, in turn, mounted on the cross slide. The 
compound rest is pivoted around a vertical axis so that the tools can be set at any de- 
sired angle with respect to the axis of the lathe (and that of the workpiece). These var- 
ious components of the carriage form a system that provides motion for the cutting 
tool in two perpendicular directions during turning operations. 

When cutting screw threads, power is provided from the gearbox to the apron by 
the lead screw. In all other turning operations, it is the feed rod that drives the carriage. 
The lead screw goes through a pair of half nuts that are fixed to the rear of the apron. 
When actuating a certain lever, the half nuts are clamped together and engage with the 
rotating lead screw as a single nut that is fed, together with the carriage, along the bed. 
When the lever is disengaged, the half nuts are released and the carriage stops. On the 
other hand, when the feed rod is used, it supplies power to the apron through a worm 
gear. This gear is keyed to the feed rod and travels with the apron along the feed rod, 
which has a keyway extending along its entire length. A modern lathe usually has a 
quick-change gearbox located under the headstock and driven from the spindle through 
a train of gears. It is connected to both the feed rod and the lead screw so that a vari- 
ety of feeds can easily and rapidly be selected by simply shifting the appropriate 
levers. The quick-change gearbox is employed in plain turning, facing, and thread- 
cutting operations. Because the gearbox is linked to the spindle, the distance that the 
apron (and the cutting tool) travels for each revolution of the spindle can be controlled 
and is referred to as the feed. 

The Turret Lathe 

A turret lathe is similar to an engine lathe, except that the conventional tool post is re- 
placed with a hexagonal (or octagonal) turret that can be rotated around a vertical axis 
as required. Appropriate tools are mounted on the six (or eight) sides of the turret. The 
length of each tool is adjusted so that, by simply indexing the turret, any tool can be 
brought into the exactly desired operating position. These cutting tools can, therefore, 
be employed successively without the need for dismounting the tool and mounting a 
new one each time, as is the case with conventional engine lathes. This results in an 
appreciable saving in the time required for setting up the tools. Also, on a turret lathe, 
a skilled machinist is required only initially to set up the tools. A laborer with limited 
training can operate the turret lathe thereafter and produce parts identical to those that 
can be manufactured when a skilled machinist operates the lathe. Figure 10.3 illus- 
trates a top view of a hexagonal turret with six different tools mounted on its sides. 
Sometimes, the turret replaces the tailstock and can be either vertical (i.e., with a hor- 
izontal axis) or horizontal (i.e., with a vertical axis). In this case, four additional tools 
can be mounted on the square tool post, sometimes called a square turret, thus allow- 

10.1 Turning Operations 


FIGURE 10.3 

Top view of a hexagonal 
turret with six different 


ing twelve machining operations to be performed successively. Turret lathes always 
have work-holding devices with quick-release (and quick-tightening) mechanisms. 

Specifying a Lathe 

It is important for a manufacturing engineer to be able to specify a lathe in order to 
place an order or to compare and examine contract bids. The specifications of a lathe 
should involve data that reveal the dimensions of the largest workpiece to be machined 
on that lathe. They also must include the power consumption, as well as information 
that is needed for shipping and handling. Table 10.1 indicates an example of how to 
specify a lathe. 

Tool Holding 

Tools for turning operations are mounted in a toolholder (tool post). On an engine 
lathe, it is located on the compound rest. More than one cutting tool (up to four) can 
be mounted in the toolholder in order to save the time required for changing and set- 
ting up each tool should only one tool be mounted at a time. In all turning operations, 
the following conditions for holding the tools must be fulfilled: 

1. The tip of the cutting edge must fully coincide with the level of the lathe axis. 
This can be achieved by using the pointed edge of the lathe center as a basis for 
adjustment, as shown in Figure 10.4. Failure to meet this condition results in a 
change in the values of the cutting angles from the desired ones. 

2. The centerline of the cutting tool must be horizontal. 

3. The tool must be fixed tightly along its length and not just on two points. 

4. A long tool-overhang should be avoided in order to eliminate any possibility for 
elastic strains and vibrations. 


10 Machining of Metals 

TABLE 10.1 

Example of 
specifications of a lathe 

/ Model 

Example \ 

Maximum swing over bed (largest diameter 

12 in. (300 mm) 

of workpiece) 

Maximum swing over carriage (largest 

8 in. (200 mm) 

diameter over carriage) 

Hole through spindle 

0.75 in. (19 mm) 

Height of centers 

6 in. (150 mm) 

Turning length 

24 in. (600 mm) 

Thread on spindle nose 

Taper in spindle and tailstock sleeves 

3 Morse 

21 spindle speeds 

20-2000 rev/min. 

Metric threads 

2-6 mm 


4-28 teeth 

Feeds per revolution 

0.0002-0.008 in. (0.05-0.2 mm) 

Power required 

1.6 kW 

Net weight 

1 ton 

Floor space requirement 

64/36/56 in. 


(1600/900/1400 mm) 

Lathe Cutting Tools 

The shape and geometry of lathe cutting tools depend upon the purpose for which they 
are employed. Turning tools can be classified into two main groups: external cutting 
tools and internal cutting tools. 

Types of tools. Each of these groups includes the following types of tools: 

1. Turning tools. Turning tools can be either finishing or rough turning tools. Rough 
turning tools have small nose radii and are employed when deep cuts are made. 
Finishing tools have larger nose radii and are used when shallower cuts are made 
in order to obtain the final required dimensions with good surface finish. Rough 
turning tools can be right-hand or left-hand tools, depending upon the direction of 
feed. They can have straight, bent, or offset shanks. Figure 10.5 illustrates the dif- 
ferent kinds of turning tools. 

FIGURE 10.4 

A simple method for 
tool setup 


10.1 Turning Operations 


FIGURE 10.5 

Different kinds of 
turning tools 

Right-hand Left-hand 

Rough turning tools 


Finishing tools 

2. Facing tools. Facing tools are employed in facing operations for machining fiat 
side or end surfaces. As can be seen in Figure 10.6, there are tools for machining 
both left and right side surfaces. These side surfaces are generated through the use 
of cross feed, contrary to turning operations, where longitudinal feed is used. 

3. Cutoff tools. Cutoff tools, which are sometimes called parting tools, serve to sepa- 
rate the workpiece into parts and/or machine external annular grooves, as shown in 
Figure 10.7. 

4. Thread-cutting tools. Thread-cutting tools have either triangular, square, or trape- 
zoidal cutting edges, depending upon the cross section of the desired thread. Also, 
the plane angles of these tools must always be identical to those of the thread 
forms. Thread-cutting tools have straight shanks for external thread cutting and 
bent shanks for internal thread cutting. Figure 10.8 illustrates the different shapes 
of thread-cutting tools. 

FIGURE 10.6 

Different kinds of 
facing tools 

FIGURE 10.7 

Cutoff tools 




10 Machining of Metals 

FIGURE 10.8 

Different shapes of 
thread-cutting tools 




5. Form tools. As shown in Figure 10.9, form tools have edges specially manufactured 
to take a form that is opposite to the desired shape of the machined workpiece. 

Internal and external tools. The types of internal cutting tools are similar to those of 
the external cutting tools. They include tools for rough turning, finish turning, thread 
cutting, and recess machining. Figure 10.10 illustrates the different types of internal 
cutting tools. 

Carbide tips. As previously mentioned, a high-speed steel tool is usually made in 
the form of a single piece, contrary to cemented carbides or ceramics, which are 
made in the form of tips. The tips are brazed or mechanically fastened to steel 
shanks. Figure 10.11 shows an arrangement that includes a carbide tip, a chip 
breaker, a seat, a clamping screw (with a washer and a nut), and a shank. As its name 
suggests, the function of a chip breaker is to break long chips every now and then, 
thus preventing the formation of very long, twisted ribbons that may cause problems 
during the machining operation. As shown in Figure 10.12, the carbide tips (or ce- 
ramic tips) have different shapes, depending upon the machining operations for 
which they are to be employed. The tips can either be solid or have a central through 
hole, depending upon whether brazing or mechanical clamping is employed for 
mounting the tip on the shank. 

FIGURE 10.9 

Form tools 




FIGURE 10.10 

Different types of 
internal cutting tools 


Recess or 
groove making 


nm\ u 



10.1 Turning Operations 


FIGURE 10.11 

A carbide tip fastened 
to a toolholder 


FIGURE 10.12 

Different shapes of 
carbide tips 






Methods of Supporting Workpieces 
in Lathe Operations 

Some precautions must be taken when mounting workpieces on a lathe to ensure 
trouble-free machining. They can be summarized as follows: 

1. It is recommended that an appropriate gripping force that is neither too high nor too 
low be used. A high gripping force may result in distortion of the workpiece after 
the turning operation, whereas a low gripping force causes either vibration of the 
workpiece or slip between the workpiece and the spindle (i.e., the rotational speed, 
or rpm, of the workpiece will be lower than that of the spindle). 

2. The workpiece must be fully balanced, both statically and dynamically, by em- 
ploying counterweights and the like if necessary. 

3. The cutting force should not affect the shape of the workpiece or cause any perma- 
nent deformation. A manufacturing engineer should calculate the cutting force using 
his or her knowledge of metal cutting (Chapter 9) and then check whether or not such 
a force will cause permanent deformation by using stress analysis. Such calculations 
are very important when machining slender workpieces (i.e., those with high length- 
to-diameter ratios). Whenever it becomes evident that the cutting force will cause 
permanent deformation, the machining parameters must be changed to reduce the 
magnitude of the force (e.g., use a smaller depth of cut or lower feed). 

Following is a brief discussion of each of the work-holding methods employed in 
lathe operations. 

Holding the workpiece between two centers. The workpiece is held between two 
centers when turning long workpieces like shafts and axles having length-to-diameter 
ratios higher than 3 or 4. Before a workpiece is held, each of its flat ends must be pre- 
pared by drilling a 60° center hole. The pointed edges of the live center (mounted in 
the tailstock so that its conical part rotates freely with the workpiece) and the dead 
center (mounted in the spindle hole) are inserted in the previously drilled center holes. 


10 Machining of Metals 

FIGURE 10.13 

Holding the workpiece 
between two centers 
during turning 


(screwed on the 

spindle nose) 

As shown in Figure 10.13, a driving dog is clamped on the left end of the workpiece 
by means of a tightening screw. The tail of the lathe dog enters a slot in the driving- 
dog plate (or faceplate), which is screwed on the spindle nose. 

When very long workpieces having length-to-diameter ratios of 10 or more are 
turned between centers, rests must be used to provide support and prevent sagging of 
the workpiece at its middle. Steady rests are clamped on the lathe bed and thus do not 
move during the machining operation; follower rests are bolted to and travel with the 
carriage. A steady rest employs three adjustable fingers to support the workpiece. 
However, in high-speed turning, the steady rest should involve balls and rollers at the 
end of the fingers where the workpiece is supported. A follower rest has only two fin- 
gers and supports the workpiece against the cutting tool. A steady rest can be used as 
an alternative to the tailstock for supporting the right-hand end of the workpiece. Fig- 
ure 10.14 illustrates a steady rest used to support a very long workpiece. 

Holding the workpiece in a chuck. When turning short workpieces and/or when per- 
forming facing operations, the workpiece is held in a chuck, which is screwed on the 
spindle nose. A universal, self-centering chuck has three jaws that can be moved sep- 

FIGURE 10.14 

A steady rest used to 
support a very long 

Three adjustable 

10.1 Turning Operations 


arately or simultaneously in radial slots toward its center to grip the workpiece or away 
from its center to release the workpiece. This movement is achieved by inserting a 
chuck wrench into a square socket and then turning it as required. Four-jaw chucks are 
also employed; these are popular when turning complex workpieces and those having 
asymmetric shapes. Magnetic chucks (without jaws) are used to hold thin, fiat work- 
pieces for facing operations. There are also pneumatic and hydraulic chucks, and they 
are utilized for speeding up the processes of loading and unloading the workpieces. 
Figure 10.15 shows how a workpiece is held in a chuck. 

Mounting the workpiece on a faceplate. A faceplate is a large circular disk with ra- 
dial plain slots and T-slots in its face. The workpiece can be mounted on it with the 
help of bolts, T-nuts, and other means of clamping. The faceplate is usually employed 
when the workpiece to be gripped is large or noncircular or has an irregular shape and 
cannot, therefore, be held in a chuck. Before any machining operation, the faceplate 
and the workpiece must be balanced by a counterweight mounted opposite to the 
workpiece on the faceplate, as shown in Figure 10.16. 

Using a mandrel. Disklike workpieces or those that have to be machined on both 
ends are mounted on mandrels, which are held between the lathe centers. In this case, 
the mandrel acts like a fixture and can take different forms. As Figure 10.17 shows, a 

FIGURE 10.15 

Holding the workpiece 
in a chuck 



Large bevel gear 
with spiral scrol 
on the other side 

FIGURE 10.16 

Mounting the workpiece 
on a faceplate 





10 Machining of Metals 

FIGURE 10.17 

Mounting the workpiece 
on a mandrel 



(for the dog) 


y; // ;; ;/ // // /> 



Mandrel having 
invisible slope 


mandrel can be a truncated conical rod with an intangible slope on which the work- 
piece is held by the wedge action. A split sleeve that is forced against a conical rod is 
also employed. There are also some other designs for mandrels. 

Holding the workpiece in a chuck collet. A chuck collet consists of a three-segment 
split sleeve with an external tapered surface. The collet can grip a smooth bar placed 
between these segments when a collet sleeve, which is internally tapered, is pushed 
against the external tapered surface of the split sleeve, as shown in Figure 10.18. 

Lathe Operations 

The following sections focus on the various machining operations that can be per- 
formed on a conventional engine lathe. It must be born in mind, however, that modern 
computerized numerically controlled (CNC) lathes have more capabilities and can do 
other operations, such as contouring, for example. Following are the conventional 
lathe operations. 

Cylindrical turning. Cylindrical turning is the simplest and the most common of all 
lathe operations. A single full turn of the workpiece generates a circle whose center 
falls on the lathe axis; this motion is then reproduced numerous times as a result of the 
axial feed motion of the tool. The resulting machining marks are, therefore, a helix 
having a very small pitch, which is equal to the feed. Consequently, the machined sur- 
face is always cylindrical. 

The axial feed is provided by the carriage or compound rest, either manually or 
automatically, whereas the depth of cut is controlled by the cross slide. In roughing 
cuts, it is recommended that large depths of cuts, up to 1/4 inch (6 mm) depending 
upon the workpiece material, and smaller feeds be used. On the other hand, very fine 
feeds, smaller depths of cut, less than 0.05 inch (0.4 mm), and high cutting speeds are 

FIGURE 10.18 

Holding the workpiece 
in a chuck collet 


10.1 Turning Operations 


FIGURE 10.19 

Equations applicable to lathe operations 


Cutting Speed 

Machining Time 

Material -removal Rate 

N (rpm) 


V = it(D +2d)N 

where L = *. workpi e C e + allowance 
i.e., length of the workpiece plus 


MRR = tt(0 + d)N-f-d 

f (feed) 


V = nDN 

T -k 

MRR = n(D - d)N-f-d 


Feed, f 

max. V = nDN 
min. V = 

.. nDN 

mean y = — - — 

D + allowance 

max. MRR = nDN-f-d 
mean MRR = 


max. V = nDN 
min. V = 


D + allowance 

mean 1/ 

max. MRR = nDN-f-d 

MDD irDN-f-d 
mean MRR = 

Feed, f 

preferred for finishing cuts. Figure 10.19 indicates the equations used to estimate the 
different machining parameters in cylindrical turning. 

Facing. The result of a facing operation is a flat surface that is either the entire end 
surface of the workpiece or an annular intermediate surface like a shoulder. During a 
facing operation, feed is provided by the cross slide, whereas the depth of cut is con- 
trolled by the carriage or compound rest. Facing can be carried out either from the pe- 
riphery inward or from the center of the workpiece outward. It is obvious that the 
machining marks in both cases take the form of a spiral. Usually, it is preferred to 
clamp the carriage during a facing operation as the cutting force tends to push the tool 
(and, of course, the whole carriage) away from the workpiece. In most facing opera- 
tions, the workpiece is held in a chuck or on a faceplate. Figure 10.19 also indicates 
the equations applicable to facing operations. 


10 Machining of Metals 

Groove cutting. In cutoff and groove-cutting operations, only cross feed of the tool is 
employed. The cutoff and grooving tools that were previously discussed are employed. 

Boring and internal turning. Boring and internal turning are performed on the inter- 
nal surfaces by a boring bar or suitable internal cutting tool. If the initial workpiece is 
solid, a drilling operation must be performed first. The drilling tool is held in the tail- 
stock, which is then fed against the workpiece. 

Taper turning. Taper turning is achieved by driving the tool in a direction that is not 
parallel to the lathe axis but inclined to it with an angle that is equal to the desired 
angle of the taper. Following are the different methods used in taper turning: 

1. One method is to rotate the disk of the compound rest with an angle equal to half the 
apex angle of the cone, as is shown in Figure 10.20. Feed is manually provided by 
cranking the handle of the compound rest. This method is recommended for the taper 
turning of external and internal surfaces when the taper angle is relatively large. 

2. Special form tools can be used for external, very short, conical surfaces, as shown 
in Figure 10.21. The width of the workpiece must be slightly smaller than that of 
the tool, and the workpiece is usually held in a chuck or clamped on a faceplate. In 

FIGURE 10.20 

Taper turning by 
rotating the disk of the 
compound rest 



FIGURE 10.21 

Taper turning by 
employing a form tool 




10.1 Turning Operations 


this case, only the cross feed is used during the machining process, and the carriage 
is clamped to the machine bed. 

The method of offsetting the tailstock center, as shown in Figure 10.22, is em- 
ployed for the external taper turning of long workpieces that are required to have 
small taper angles (less than 8°). The workpiece is mounted between the two cen- 
ters; then the tailstock center is shifted a distance S in the direction normal to the 
lathe axis. This distance can be obtained from the following equation: 

5 = 

L(D - d) 


where: L is the full length of the workpiece 

D is the largest diameter of the workpiece 
d is the smallest diameter of the workpiece 
i is the length of the tapered surface 

4. A special taper-turning attachment, such as the one shown in Figure 10.23, is used 
for turning very long workpieces, when the length is larger than the full stroke of 
the compound rest. The procedure followed in such cases involves complete disen- 
gagement of the cross slide from the carriage, which is then guided by the taper- 
turning attachment. During this process, the automatic axial feed can be used as 
usual. This method is recommended for very long workpieces with a small cone 
angle (8° through 10°). 

Thread cutting. For thread cutting, the axial feed must be kept at a constant rate, 
which is dependent upon the rotational speed (rpm) of the workpiece. The relationship 
between both is determined primarily by the desired pitch of the thread to be cut. 

As previously mentioned, the axial feed is automatically generated when cutting a 
thread by means of the lead screw, which drives the carriage. When the lead screw rotates 

FIGURE 10.22 

Taper turning by 
offsetting the tailstock 


10 Machining of Metals 

FIGURE 10.23 

Taper turning by 
employing a special 


a single revolution, the carriage travels a distance equal to the pitch of the lead screw. Con- 
sequently, if the rotational speed of the lead screw is equal to that of the spindle (i.e., that 
of the workpiece), the pitch of the resulting cut thread is exactly equal to that of the lead 
screw. The pitch of the resulting thread being cut, therefore, always depends upon the ratio 
of the rotational speeds of the lead screw and the spindle: 

pitch of lead screw 

rpm of workpiece 

desired pitch of workpiece rpm of lead screw 

= spindle-to-carriage gearing ratio 


This equation is useful in determining the kinematic linkage between the lathe spindle 
and the lead screw and enables proper selection of the gear train between them. 

In thread-cutting operations, the workpiece can be either held in a chuck or 
mounted between two lathe centers for relatively long workpieces. The form of the 
tool used must exactly coincide with the profile of the thread to be cut (i.e., triangular 
tools must be used for triangular threads, and so on). 

Knurling. Knurling is basically a forming operation in which no chips are produced. 
It involves pressing two hardened rolls with rough filelike surfaces against the rotat- 
ing workpiece to cause plastic deformation of the workpiece metal, as shown in Fig- 
ure 10.24. Knurling is carried out to produce rough, cylindrical (or conical) surfaces 
that are usually used as handles. Sometimes, surfaces are knurled just for the sake of 
decoration, in which case there are different knurl patterns to choose from. 

Cutting Speeds and Feeds 

The cutting speed, which is usually given in surface feet per minute (SFM), is the 
number of feet traveled in the circumferential direction by a given point on the surface 
(being cut) of the workpiece in one minute. The relationship between the surface speed 

10.1 Turning Operations 


FIGURE 10.24 

The knurling operation 

and the rpm can be given by the following equation: 
SFM = kDN (see Table 1 0. 1 ) 
where: D is the diameter of the workpiece in feet 
TV is the rpm 

The surface cutting speed is dependent upon the material being machined as well 
as the material of the cutting tool and can be obtained from handbooks and informa- 
tion provided by cutting-tool manufacturers. Generally, the SFM is taken as 100 when 
machining cold-rolled or mild steel, as 50 when machining tougher metals, and as 200 
when machining softer materials. For aluminum, the SFM is usually taken as 400 or 
above. There are also other variables that affect the optimal value of the surface cut- 
ting speed. These include the tool geometry, the type of lubricant or coolant, the feed, 
and the depth of cut. As soon as the cutting speed is decided upon, the rotational speed 
(rpm) of the spindle can be obtained as follows: 

/V = 



The selection of a suitable feed depends upon many factors, such as the required 
surface finish, the depth of cut, and the geometry of the tool used. Finer feeds will pro- 
duce better surface finish, whereas higher feeds reduce the machining time during which 
the tool is in direct contact with the workpiece. Therefore, it is generally recommended 
to use high feeds for roughing operations and finer feeds for finishing operations. Again, 
recommended values for feeds, which can be taken as guidelines, are found in hand- 
books and in information booklets provided by cutting-tool manufacturers. 

Design Considerations for Turning 

When designing parts to be produced by turning, the product designer must consider 
the possibilities and limitations of the turning operation as well as the machining cost. 
The cost increases with the quality of the surface finish, with the tightness of the tol- 
erances, and with the area of the surface to be machined. Therefore, it is not recom- 
mended that high-quality surface finishes or tighter tolerances be used in the product 
design unless they are required for the proper functioning of the product. Figure 10.25 


10 Machining of Metals 

FIGURE 10.25 

Design considerations 
for turning: (a) reduce 
area of surface to be 
machined; (b) reduce 
number of operations 
required; (c) provide 
allowance for tool 
clearance; (d) opt for 
machining external over 
internal surfaces; 
(e) opt for through 
boring over alternatives 




>*■ LJ 



Not recommended Preferred 


Preferred Not recommended 



Better design 

Less recommended 


graphically depicts some design considerations for turning. Here are the guidelines to 
be followed: 

1. Try to reduce the area of the surfaces to be machined, especially when a large num- 
ber of parts is required or when the surfaces are to mate with other parts (see Fig- 
ure 10.25a). 

2. Try to reduce the number of operations required by appropriate changes in the de- 
sign (see Figure 10.25b). 

3. Provide an allowance for tool clearance between different sections of a product (see 
Figure 10.25c). 

4. Always keep in mind that machining of exposed surfaces is easier and less expen- 
sive than machining of internal surfaces (see Figure 10.25d). 

10.2 Shaping and Planing Operations 379 

5. Remember that through boring is easier and cheaper than other alternatives (see 
Figure 10.25e). 


Planing, shaping, and slotting are processes for machining horizontal, vertical, and in- 
clined flat surfaces, slots, or grooves by means of a lathe-type cutting tool. In all these 
processes, the cutting action takes place along a straight line. In planing, the workpiece 
(and the machine bed) is reciprocated, and the tool is fed across the workpiece to re- 
produce another straight line, thus generating a flat surface. In shaping and slotting, the 
cutting tool is reciprocated, and the workpiece is fed normal to the direction of tool 
motion. The difference between the latter two processes is that the tool path is hori- 
zontal in shaping and it is vertical in slotting. Shapers and slotters can be employed in 
cutting external and internal keyways, gear racks, dovetails, and T-slots. Shapers and 
planers have become virtually obsolete because most shaping and planing operations 
have been replaced by more productive processes such as milling, broaching, and abra- 
sive machining. The use of shapers and planers is now limited to the machining of 
large beds of machine tools and the like. 

In all three processes, there are successive alternating cutting and idle return 
strokes. The cutting speed is, therefore, the speed of the tool (or the workpiece) in the 
direction of cutting during the working stroke. The cutting speed may be either con- 
stant throughout the working stroke or variable, depending upon the design of the 
shaper or planer. Let us now discuss the construction and operation of the most com- 
mon types of shapers and planers. 

Horizontal Push-Cut Shaper 

Construction. As can be seen in Figure 10.26, a horizontal push-cut shaper consists 
of a frame that houses the speed gearbox and the quick-return mechanism that trans- 
mits power from the motor to the ram and the table. The ram travel is the primary mo- 
tion that produces a straight-line cut in the working stroke, whereas the intermittent 
cross travel of the table is responsible for the cross feed. The tool head is mounted at 
the front end of the ram and carries the clapper box toolholder. The toolholder is piv- 
oted at its upper end to allow the tool to rise during the idle return stroke in order not 
to ruin the newly machined surface. The tool head can be swiveled to permit the ma- 
chining of inclined surfaces. 

The workpiece can be either bolted directly to the machine table or held in a vise 
or other suitable fixture. The cross feed of the table is generated by a ratchet and pawl 
mechanism that is driven through the quick-return mechanism (i.e., the crank and the 
slotted arm). The machine table can be raised or lowered by means of a power screw 
and a crank handle. It can also be swiveled in a universal shaper. 

Quick-return mechanism. As can be seen in Figure 10.27, the quick-return mecha- 
nism involves a rotating crank that is driven at a uniform angular speed and an oscil- 
lating slotted arm that is connected to the crank by a sliding block. The working stroke 
takes up an angle (of the crank revolution) that is larger than that of the return stroke. 


10 Machining of Metals 

FIGURE 10.26 

Design features of a 
horizontal push-cut 

Feed screw 

(to control 

depth of cut) 

Tool slide 



for adjusting 
table height 

Because the angular speed of the crank is constant, it is obvious that the time taken by 
the idle return stroke is less than that taken by the cutting stroke. In fact, it is the main 
function of the quick-return mechanism to reduce the idle time during the machining 
operation to a minimum. 

Now, let us consider the average speed (s) of the tool during the cutting stroke. It 
can be determined as a function of the length of the stroke and the number of strokes 
per minute as follows: 

2LN . . , . , . . , 

5 = in tt/min. (m/min.) 


where: L is the length of stroke in feet (m) 

TV is the number of strokes per minute 
C is the cutting time ratio 

Note that the cutting time ratio is 

cutting time 


C = 

total time for one crank revolution 

_ angle corresponding to cutting stroke 


It is also obvious that the total number of strokes required to machine a given surface 
can be given by the following equation: 



10.2 Shaping and Planing Operations 


FIGURE 10.27 

Details and working 
principles of the quick- 
return mechanism 

Length of stroke 


where: W is the total width of the workpiece 

/is the cross feed (e.g., inches per stroke) 
Therefore, the machining time T is n/N. After mathematical manipulation, it can be 
given as follows: 

2LN (10.6) 

T = 


382 10 Machining of Metals 

Next, the metal-removal rate (MRR) can be given by the following equation: 

MRR = rx/xLxyV(in. 3 /min.) (10.7) 

Vertical Shaper 

The vertical shaper is similar in construction and operation to the push-cut shaper, the 
difference being that the ram and the tool head travel vertically instead of horizontally. 
Also, in this type of shaper, the workpiece is mounted on a round table that can have 
a rotary feed whenever desired to allow the machining of curved surfaces (e.g., spiral 
grooves). Vertical shapers, which are sometimes referred to as slotters, are used in in- 
ternal cutting. Another type of vertical shaper is known as a keyseater because it is 
specially designed for cutting keyways in gears, cams, pulleys, and the like. 


A planer is a machine tool that does the same work as the horizontal shaper but on work- 
pieces that are much larger than those machined on a shaper. Although the designs of 
planers vary, most common are the double-housing and open-side constructions. In a 
double-housing planer, two vertical housings are mounted at the sides of the long, heavy 
bed. A cross rail that is supported at the top of these housings carries the cutting tools. 
The machine table (while in operation) reciprocates along the guide ways of the bed and 
has T-slots in its upper surface for clamping the workpiece. In this type of planer, the 
table is powered by a variable-speed dc motor through a gear drive. The cross rail can 
be raised or lowered as required, and the inclination of the tools can be adjusted as well. 
In an open-side planer, there is only one upright housing at one side of the bed. This con- 
struction provides more flexibility when wider workpieces are to be machined. 

Planing and Shaping Tools 

Planing and shaping processes employ single-point tools of the lathe type, but heavier 
in construction. They are made of either high-speed steel or carbon tool steel with car- 
bide tips. In the latter case, the machine tool should be equipped with an automatic lift- 
ing device to keep the tool from rubbing the workpiece during the return stroke, thus 
eliminating the possibility of breaking or chipping the carbide tips. 

The cutting angles for these tools depend upon the purpose for which the tool is 
to be used and the material being cut. The end relief angle does not usually exceed 4°, 
whereas the side relief varies between 6° and 14°. The side rake angle also varies be- 
tween 5° (for cast iron) and 15° (for medium-carbon steel). 


Drilling involves producing through or blind holes in a workpiece by forcing a tool 
that rotates around its axis against the workpiece. Consequently, the range of cutting 
from this axis of rotation is equal to the radius of the required hole. In practice, two 
symmetrical cutting edges that rotate about the same axis are employed. 

10.3 Drilling Operations 


Drilling operations can be carried out by using either hand drills or drilling ma- 
chines. The latter differ in size and construction. Nevertheless, the tool always rotates 
around its axis while the workpiece is kept firmly fixed. This is contrary to drilling on 
a lathe. 

Cutting Tools for Drilling Operations 

In drilling operations, a cylindrical rotary-end cutting tool, called a drill, is employed. 
The drill can have one or more cutting edges and corresponding flutes that are straight 
or helical. The function of the flutes is to provide outlet passages for the chips gener- 
ated during the drilling operation and also to allow lubricants and coolants to reach the 
cutting edges and the surface being machined. Following is a survey of the commonly 
used types of drills. 

Twist drill. The twist drill is the most common type of drill. It has two cutting edges 
and two helical flutes that continue over the length of the drill body, as shown in Fig- 
ure 10.28. The drill also consists of a neck and a shank that can be either straight or ta- 
pered. A tapered shank is fitted by the wedge action into the tapered socket of the 
spindle and has a tang that goes into a slot in the spindle socket, thus acting as a solid 
means for transmitting rotation. Straight-shank drills are held in a drill chuck that is, 
in turn, fitted into the spindle socket in the same way as tapered-shank drills. 

As can be seen in Figure 10.28, the two cutting edges are referred to as the lips 
and are connected together by a wedge, which is a chisel-like edge. The twist drill also 
has two margins that allow the drill to be properly located and guided while it is in op- 
eration. The tool point angle (TPA) is formed by the two lips and is chosen based on 
the properties of the material to be cut. The usual TPA for commercial drills is 118°, 
which is appropriate for drilling low-carbon steels and cast irons. For harder and 
tougher metals, such as hardened steel, brass, and bronze, larger TPAs (130° or 140°) 

FIGURE 10.28 

A twist drill 



Tool point 








10 Machining of Metals 

give better performance. The helix angle of the flutes of a twist drill ranges between 
24° and 30°. When drilling copper or soft plastics, higher values for the helix angle are 
recommended (between 35° and 45°). Twist drills are usually made of high-speed 
steel, although carbide-tipped drills are also available. The sizes of twist drills used in 
industrial practice range from 0.01 inch to V/i inches (0.25 up to 80 mm). 

Core drill. A core drill consists of the chamfer, body, neck, and shank, as shown in 
Figure 10.29. This type of drill may have three or four flutes and an equal number of 
margins, which ensures superior guidance, thus resulting in high machining accuracy. 
The figure also shows that a core drill has a flat end. The chamfer can have three or 
four cutting edges, or lips, and the lip angle may vary between 90° and 120°. Core 
drills are employed for enlarging previously made holes and not for originating holes. 
This type of drill promotes greater productivity, high machining accuracy, and superior 
quality of the drilled surfaces. 

Gun drill. A gun drill is used for drilling deep holes. All gun drills are straight-fluted, 
and each has a single cutting edge. A hole in the body acts as a conduit to trans- 
mit coolant under considerable pressure to the tip of the drill. As can be seen in Fig- 
ure 10.30, there are two kinds of gun drills: the center-cut gun drill used for drilling 
blind holes and the trepanning drill. The latter has a cylindrical groove at its center, 
thus generating a solid core that guides the tool as it proceeds during the drilling op- 

Spade drill. A spade drill is used for drilling large holes of 3V 2 inches (90 mm) or 
more. The design of this type of drill results in a marked saving in tool cost as well as 
in a tangible reduction in tool weight that facilitates its ease of handling. Moreover, 
this drill is easy to grind. Figure 10.31 shows a spade drill. 

Saw-type cutter. A saw-type cutter, like the one illustrated in Figure 10.32, is used 
for cutting large holes in thin metal. 

Drills made in combination with other tools. An example is a tool that involves both 
a drill and a tap. Step drills and drill and countersink tools are also sometimes used in 
industrial practice. 

Cutting Speeds and Feeds in Drilling 

We can easily see that the cutting speed varies along the cutting edge. It is always 
maximum at the periphery of the tool and is equal to zero on the tool axis. Never- 
theless, we consider the maximum speed because it is the one that affects the tool 

FIGURE 10.29 

A core drill 



n _■ Neck 

Body 1 i Shank 


10.3 Drilling Operations 


FIGURE 10.30 

Gun drills: (a) 
trepanning gun drill; (b) 
center-cut gun drill 

Cutting Cutting fluid 
edge passage 

Shape of the 
resulting hole 



FIGURE 10.31 

A spade drill 



of the 

resulting hole 

FIGURE 10.32 

A saw-type cutter 


10 Machining of Metals 

wear and the quality of the machined surface. The maximum speed must not exceed 
the permissible cutting speed, which depends upon the material of the workpiece as 
well as the material of the cutting tool. Data about permissible cutting speeds in 
drilling operations can be found in handbooks. The rotational speed of the spindle 
can be determined from the following equation: 

N = 




where: N is the rotational speed of the spindle (rpm) 
D is the diameter of the drill in feet (m) 
CS is the permissible cutting speed in ft/min. (m/min.) 

In drilling operations, feeds are expressed in inches or millimeters per revolution. 
Again, the appropriate value of feed to be used depends upon the metal of the work- 
piece and drill material and can be found in handbooks. Whenever the production rate 
must be increased, it is advisable to use higher feeds rather than increase the cutting 

Other Types of Drilling Operations 

In addition to conventional drilling, there are other operations that are involved in the 
production of holes in industrial practice. Following is a brief description of each of 
these operations. 

Boring. Boring involves enlarging a hole that has already been drilled. It is similar to 
internal turning and can, therefore, be performed on a lathe, as previously mentioned. 
There are also some specialized machine tools for carrying out boring operations. 
These include the vertical boring mill, the jig boring machine, and the horizontal bor- 
ing machine. 

Counterboring. As a result of counterboring, only one end of a drilled hole is en- 
larged, as is illustrated in Figure 10.33a. This enlarged hole provides a space in which 
to set a bolt head or a nut so that it will be entirely below the surface of the part. 

Spot facing. Spot facing is performed to finish off a small surface area around the 
opening of a hole. As can be seen in Figure 10.33b, this process involves removing a 
minimal depth of cut and is usually performed on castings or forgings. 

Countersinking. As shown in Figure 10.33c, countersinking is done to accommodate 
the conical seat of a flathead screw so that the screw does not appear above the surface 
of the part. 

FIGURE 10.33 

Operations related to 

(a) counterboring; 

(b) spot facing; 

(c) countersinking 

10.3 Drilling Operations 


FIGURE 10.34 

Details of a reamer 

Fluted section 




Rake angle 

Tool angle 

Cutting angles 
of a tooth 

Reaming. Reaming is actually a "sizing" process, by which an already drilled hole is 
slightly enlarged to the desired size. As a result of a reaming operation, a hole has a very 
smooth surface. The cutting tool used in this operation is known as a reamer. As shown 
in Figure 10.34, a reamer has a fluted section, a neck, and a shank. The fluted section in- 
cludes four zones: the chamfer, the starting taper, the sizing zone, and the back taper. 
The chamfer or bevel encloses an angle that depends upon the method of reaming and 
the material being cut. This is a consequence of the fact that this angle affects the mag- 
nitude of the axial reaming force. The larger the chamfer angle, the larger the required 
reaming force. Table 10.2 indicates some recommended values of the chamfer angle for 
different reaming conditions. The starting taper is the part of the reamer that actually re- 
moves chips. Figure 10.34 also shows that each tooth of that part of the reamer has a cut- 
ting edge as well as rake, relief, and tool (or lip) angles. The sizing zone guides the 
reamer and smooths the surface of the hole. Finally, the back taper serves to reduce fric- 
tion between the reamer and the newly machined surface. 

Reamers are usually made of hardened tool steel. Nevertheless, reamers that are 
used in mass production are tipped with cemented carbides in order to increase the tool 
life and the production rate. 

Tapping. Tapping is the process of cutting internal threads. The tool used is called a 
tap. As shown in Figure 10.35, it has a boltlike shape with four longitudinal flutes. 
Made of hardened tool steel, taps can be used for either manual or machine cutting of 

TABLE 10.2 

Recommended values 
of the chamfer angle of 

Metal to Be Reamed 


Cast Iron 

Soft Metals 

Manual reaming 
Machining reaming 

8°-10 c 

20°-30 c 



10 Machining of Metals 

FIGURE 10.35 

A tap 






threads. In the latter case, the spindle of the machine tool must reverse its direction of 
rotation at the end of the cutting stroke so that the tap can be withdrawn without de- 
stroying the newly cut thread. When tapping is carried out by hand, a set of three taps 
is used for each desired threaded hole size. The three taps differ slightly in size, and 
two of them are actually undersized. The first tap of the set to be used is always a tap- 
per tap; it reduces the torque (and, consequently, the power) required for tapping. 

Design Considerations for Drilling 

Figure 10.36 graphically depicts some design considerations for drilling. Here are the 
guidelines to be followed: 

1. Make sure the centerline of the hole to be drilled is normal to the surface of the 
part. This is to avoid bending and breaking the tool during the drilling operation. As 
previously mentioned, the twist drill has a chisel edge and not a pointed edge at its 
center. This, although it facilitates the process of grinding the tool, causes the tool 
to shift from the desired location and makes it liable to breakage, especially if it is 
not normal to the surface to be drilled. (See Figure 10.36a for examples of poor and 
proper design practice for drilled holes.) 

2. When tapping through holes, ensure that the tap will be in the clear when it appears 
from the other side of the part (see Figure 10.36b). 

3. Remember that it is impossible to tap the entire length of a blind or counterbored 
hole without providing special tool allowance (see Figure 10.36c). 

Classification of Drilling Machines 

Drilling operations can be carried out by employing small portable machines or by 
using the appropriate machine tools. These machine tools differ in shape and size, but 
they have common features. For instance, they all involve one or more twist drills, each 
rotating around its own axis while the workpiece is kept firmly fixed. This is contrary 
to the drilling operation on a lathe, where the workpiece is held in and rotates with the 
chuck. Following is a survey of the commonly used types of drilling machines. 

Bench-type drilling machines. Bench-type drilling machines are general-purpose, 
small machine tools that are usually placed on benches. This type of drilling machine 
includes an electric motor as the source of motion, which is transmitted via pulleys and 
belts to the spindle, where the tool is mounted. The feed is manually generated by low- 

10.3 Drilling Operations 


FIGURE 10.36 

Design considerations 
for drilling: (a) set 
centerline of tool 
normal to surface to be 
drilled; (b) ensure tap is 
clear when it appears 
from other side; 
(c) provide allowance 
when tapping a blind 

To be avoided 





To be avoided 




= ^ in. (6 mm) 


ering a lever handle that is designed to lower (or raise) the spindle. The spindle rotates 
freely inside a sleeve (which is actuated by the lever through a rack-and-pinion sys- 
tem) but does not rotate with the spindle. 

The workpiece is mounted on the machine table, although a special vise is some- 
times used to hold the workpiece. The maximum height of a workpiece to be machined 
is limited by the maximum gap between the spindle and the machine table. 

Upright drilling machines. Depending upon the size, upright drilling machines can be 
used for light, medium, and even relatively heavy jobs. A light-duty upright drilling ma- 
chine is shown in Figure 10.37. It is basically similar to a bench-type machine, the main 
difference being a longer cylindrical column fixed to the base. Along the column is an 
additional sliding table for fixing the workpiece that can be locked in position at any de- 
sired height. The power required for this type of machine is greater than that for a bench- 
type drilling machine as this type is employed in performing medium-duty jobs. 


10 Machining of Metals 

FIGURE 10.37 

An upright drilling 
machine (Courtesy of 
Clausing Industrial, 
Inc., Kalamazoo, 

Drill chuck 





There are also large drilling machines of the upright type. In this case, the ma- 
chine has a box column and a higher power to deal with large jobs. Moreover, gear- 
boxes are employed to provide different rotational spindle speeds as well as axial feed 
motion, which can be preset at any desired rate. 

Multispindle drilling machines. Multispindle drilling machines are sturdily con- 
structed and require high power; each is capable of drilling many holes simultaneously. 
The positions of the different tools (spindles) can be adjusted as desired. Also, the en- 
tire head (which carries the spindles and the tools) can be tilted if necessary. This type 
of drilling machine is used mainly for mass production in jobs having many holes, 
such as cylinder blocks. 

Gang drilling machines. When several separate heads (each with a single spindle) are 
arranged on a single common table, the machine tool is then referred to as a gang 
drilling machine. This type of machine tool is particularly suitable where several op- 
erations are to be performed in succession. 

10.3 Drilling Operations 


Radial drills. Radial drills are particularly suitable for drilling holes in large and 
heavy workpieces that are inconvenient to mount on the table of an upright drilling 
machine. As shown in Figure 10.38, a radial drilling machine has a main column that 
is fixed to the base. The cantileverd guide arm, which carries the drilling head spindle 
and tool, can be raised or lowered along the column and clamped at any desired posi- 
tion. The drilling head slides along the arm and provides rotary motion and axial feed 
motion. The cantilevered guide arm can be swung, thus allowing the tool to be moved 
in all directions according to a cylindrical coordinate system. 

Turret drilling machines. Machine tools that belong in the turret drilling machine cat- 
egory are either semiautomatic or fully automatic. A common design feature is that the 
main spindle is replaced by a turret that carries several drilling, boring, reaming, and 
threading tools. Consequently, several successive operations can be carried out with 
only a single initial setup and without the need for setting up the workpiece again be- 
tween operations. 

Automatic turret drilling machines that are operated by NC or CNC systems (see 
Chapter 14) are quite common. In this case, the human role is limited to the initial 
setup and monitoring. This type of machine tool has advantages over the gang-type 
drilling machine with respect to the space required (physical size of the machine tool) 
and the number of workpiece setups. 

Deep-hole drilling machines. Deep-hole drilling machines are special machines em- 
ployed for drilling long holes like those of rifle barrels. Usually, gun-type drills are 
used and are fed slowly against the workpiece. In this type of machine tool, it is the 

FIGURE 10.38 

A radial drill 








Arm (can swing 
to left or right) 






392 10 Machining of Metals 

workpiece that is rotated, while the drill is kept from rotary motion. A deep-hole 
drilling machine may have either a vertical or a horizontal construction. However, in 
both cases, the common feature is the precise guidance and positive support of the 
workpiece during the drilling operation. 

Jig-boring machines. Jig-boring machines are specially designed to possess high pre- 
cision and accuracy. A machine of this type not only drills the holes but also locates 
them because the table movements are monitored by electronic measuring devices. Jig- 
boring machines are usually employed in the manufacture of forming and molding 
dies, gages, and work-holding devices like jigs and fixtures. 

Work-Holding Devices in Drilling 

During conventional drilling operations, the workpiece must be held firmly on the ma- 
chine table. The type of work-holding device used depends upon the shape and the size 
of the workpiece, the desired accuracy, and the production rate. For low production 
when the accuracy is not very important, conventional machine vises or vises with V- 
blocks (for round work) are used. For moderate production and when accuracy is of 
some importance, jigs are usually employed. A jig is a work-holding device that is de- 
signed to hold a particular workpiece (i.e., it cannot be used for workpieces having dif- 
ferent shapes) and to guide the cutting tool during the drilling operation. This 
eliminates the need for laying out the workpiece prior to machining, thus saving the 
time spent in blueing and scribing when no jigs are employed. The design of jigs and 
fixtures is a separate topic and is beyond the scope of this text. Interested readers are 
referred to the books dealing with tool design and with jig and fixture design that are 
given at the end of this text. 


Milling is a machining process that is carried out by means of a multiedge rotating tool 
known as a milling cutter. In this process, metal removal is achieved by simultaneously 
combining the rotary motion of the milling cutter and linear motions of the workpiece. 
Milling operations are employed in producing flat, contoured, and helical surfaces, as 
well as for thread- and gear-cutting operations. 

Each of the cutting edges of a milling cutter acts as an individual single-point cut- 
ter when it engages with the workpiece metal. Therefore, each of the cutting edges has 
the appropriate rake and relief angles. Because only a few of the cutting edges are en- 
gaged with the workpiece at a time, heavy cuts can be taken without adversely affect- 
ing the tool life. In fact, the permissible cutting speeds and feeds for milling are three 
to four times higher than those for turning or drilling. Moreover, the quality of the sur- 
faces machined by milling is generally superior to the quality of surfaces machined by 
turning, shaping, or drilling. 

A wide variety of milling cutters is available in industry. This, together with the 
fact that a milling machine is a very versatile machine tool, make the milling machine 
the backbone of a machining workshop. 

FIGURE 10.39 

Milling methods: (a) up 
milling; (b) down milling 

10.4 Milling Operations 393 

Milling Methods 

As far as the direction of cutter rotation and workpiece feed are concerned, milling is 
performed by either of the following two methods. 

Up milling (conventional milling). In up milling, the workpiece is fed against the di- 
rection of cutter rotation, as shown in Figure 10.39a. The depth of the cut (and, con- 
sequently, the load) gradually increases on the successively engaged cutting edges. 
Therefore, the machining process involves no impact loading, thus ensuring smoother 
operation of the machine tool and longer tool life. The quality of the machined surface 
obtained by up milling is not very high. Nevertheless, up milling is commonly used in 
industry, especially for rough cuts. 

Down milling (climb milling). In down milling, the cutter rotation coincides with the di- 
rection of feed at the contact point between the tool and the workpiece, as shown in Fig- 
ure 10.39b. The maximum depth of cut is achieved directly as the cutter engages with the 
workpiece. This results in a kind of impact, or sudden loading. Therefore, this method 
cannot be used unless the milling machine is equipped with a backlash eliminator on the 
feed screw. The advantages of this method include higher quality of the machined surface 
and easier clamping of workpieces as the cutting forces act downward. 

Types of Milling Cutters 

Milling cutters come in a wide variety of shapes, each designed to effectively perform 
a specific milling operation. Generally, a milling cutter can be described as a multiedge 
cutting tool having the shape of a solid of revolution, with the cutting teeth arranged 
either on the periphery or on an end face or on both. Following is a survey of the com- 
monly used types of milling cutters. 

Plain milling cutter. A plain milling cutter, as shown in Figure 10.40a, is a disk- 
shaped cutting tool that may have straight or helical teeth. This type of cutter is always 
mounted on horizontal milling machines and is used for machining flat surfaces. 

Face milling cutter. A face milling cutter, like the one in Figure 10.40b, is also used 
for machining flat surfaces. It is bolted at the end of a short arbor that is, in turn, 
mounted on a vertical milling machine. 

Feed *- -* Feed 

(a) (b) 


10 Machining of Metals 

FIGURE 10.40 

Types of milling cutters: 

(a) plain milling cutter; 

(b) face milling cutter 
with inserted teeth; 

(c) plain metal-slitting 
saw cutter; (d) side 
milling cutter; (e) angle 
milling cutter; (f) T-slot 
cutter; (g) end mill 

Straight teeth 

Helical teeth 



Plain metal-slitting saw. Figure 10.40c illustrates a plain metal-slitting saw cutter. 
Notice that it actually involves a very thin plain milling cutter. 

Side milling cutter. A side milling cutter is used for cutting slots, grooves, and 
splines. As can be seen in Figure 10.40d, it is quite similar to the plain milling cutter, 
the difference being that this type has teeth on the sides. As is the case with the plain 
cutter, the cutting teeth can be straight or helical. 

Angle milling cutter. An angle milling cutter is employed in cutting dovetail grooves, 
ratchet wheels, and the like. Figure 10.40e shows a milling cutter of this type. 

T-slot cutter. As shown in Figure 10.40f, a T-slot cutter involves a plain milling cut- 
ter with an integral shaft normal to it. As the name suggests, this type of cutter is used 
for milling T-slots. 

10.4 Milling Operations 395 

End mill cutter. An end mill cutter finds common application in cutting slots, 
grooves, flutes, splines, pocketing work, and the like. As Figure 10.40g indicates, an 
end mill cutter is always mounted on a vertical milling machine and can have two or 
four flutes, which may be straight or helical. 

Form milling cutter. The teeth of a form milling cutter have a shape that is identical 
to the section of the metal to be removed during the milling operation. Examples of 
this type of cutter include gear cutters, gear hobs, and convex and concave cutters. 
Form milling cutters are mounted on horizontal milling machines, as is explained later 
when we discuss gear cutting. 

Materials of Milling Cutters 

The commonly used milling cutters are made of high-speed steel, which is generally 
adequate for most jobs. Milling cutters tipped with sintered carbides or cast nonferrous 
alloys as cutting teeth are usually employed for mass production, where heavier cuts 
and/or high cutting speeds are required. 

Cutting Speeds and Feeds in Milling 

Figure 10.41 indicates methods of estimating the different machining parameters dur- 
ing milling operations. These parameters include the cutting speed, the feed, and the 
metal-removal rate. The cutting speed is the peripheral velocity at any point on the cir- 
cumference of the cutter. The allowable value for the cutting speed in milling is de- 
pendent upon many factors, including the cutter material, material of the workpiece, 
diameter and life of the cutter, feed, depth of cut, width of cut, number of teeth on the 
cutter, and the type of coolant used. The feed in milling operations is the rate of move- 
ment of the cutter axis relative to the workpiece. It is expressed in inches (or mm) per 
revolution or inches (or mm) per minute. It can also be expressed in inches (or mm) 
per tooth, especially for plain and face milling cutters. 

The depth of cut is the thickness of the metal layer that is to be removed in one 
cut. The maximum allowable depth of cut depends upon the material being machined 
and is commonly taken up to 5/16 inch (8 mm) in roughing operations and up to 1/16 
inch (about 1.5 mm) in finishing operations. Another parameter that affects milling op- 
erations is the width of cut, which is the width of the workpiece in contact with the cut- 
ter in a direction normal to the feed. The width of cut should decrease with increasing 
depth of cut to keep the load and power requirement below those that can be met by 
the cutter and the machine tool, respectively. 

Cutting Angles of Milling Cutters 

As previously mentioned, the geometry of any tool is basically a matter of rake and relief 
angles. Figure 10.42 shows the cutting angles of a plain, straight-tooth milling cutter (for 
simplicity). The radial rake angle facilitates the removal of chips and ranges from 10° to 
20°, depending upon the workpiece material to be cut. When machining hard metals with 
carbide-tipped cutters, a negative rake angle of 10° is usually employed. The relief angle 
also depends upon the workpiece material and varies between 12° and 25°. 

O o 

H =5. 


Hi ro 

=> s 

c E 

y £ 




"2 E 

g tJ ts 

€ § 

S E 


10.4 Milling Operations 


FIGURE 10.42 

Cutting angles of a 
plain, straight-tooth 
milling cutter 


Types of Milling Machines 

Several types of milling machines are employed in industry. They are generally 
classified by their construction and design features. They vary from the common 
general-purpose types to duplicators and machining centers that involve a tool mag- 
azine and are capable of carrying out many machining operations with a single 
workpiece setup. Following is a survey of the types of milling machines commonly 
used in industry. 

Plain horizontal milling machine. The construction of the plain horizontal milling ma- 
chine is very similar to that of a universal milling machine (see discussion that fol- 
lows), except that the machine table cannot be swiveled. Plain milling machines 
usually have a column-and knee type of construction and three table motions (i.e., lon- 
gitudinal, transverse, and vertical). The milling cutter is mounted on a short arbor that 
is, in turn, rigidly supported by the overarm of the milling machine. 

Universal milling machine. The construction of a universal milling machine is similar 
to that of the plain milling machine, except that it is more accurate and has a sturdier 
frame and its table can be swiveled with an angle up to 50°. Universal milling ma- 
chines are usually equipped with an index or dividing head that allows for the cutting 
of gears and cams, as is discussed later. Figure 10.43 shows a machine tool of this 

Vertical milling machine. As the name vertical milling machine suggests, the axis of 
the spindle that holds the milling cutter is vertical. Table movements are generally sim- 
ilar to those of plain horizontal milling machines; however, an additional rotary mo- 
tion is sometimes provided for the table when helical and circular grooves are to be 
machined. The cutters used with vertical milling machines are almost always of the 
end-mill type. Figure 10.44 shows a vertical milling machine. 

Duplicator. A duplicator is sometimes referred to as a copy milling machine because 
it is capable of reproducing an exact replica of a model. The machine has a stylus that 
scans the model, at which time counterpoints on the part are successively machined. 
Duplicators were used for the production of large forming dies for the automotive in- 
dustry, where models made of wood, plaster of paris, or wax were employed. Dupli- 
cators are not commonly used in industry now because they have been superseded by 
CAD/CAM systems. 


10 Machining of Metals 

FIGURE 10.43 

A universal milling 
machine (Courtesy of 
Manuel Pereira. 
photography specialist. 
University of 

Machining center. A machining center is comprised of a multipurpose CNC machine 
(see Chapter 14) that is capable of performing a number of different machining 
processes. A machining center has a tool magazine in which many tools are held. Tool 
changes are automatically carried out, and so are functions such as coolant turn-on/off. 
Machining centers are, therefore, highly versatile and can perform a number of ma- 
chining operations on a workpiece with a single setup. Parts having intricate shapes 
can easily be produced with high accuracy and excellent repeatability. 

Universal dividing head. The universal dividing head is an attachment mounted on 
the worktable of a universal milling machine that is employed for cutting gears. The 
function of the dividing head is to index the gear blank through the desired angle each 
time the metal between two successive teeth is removed. Therefore, this attachment is 
sometimes known as an index head. 

Figure 10.45 shows a universal dividing head, which consists of the body, the 
swivel block, the work spindle and its center, the index plate, and the index crank with 
a latch pin. The workpiece (with one of its ends supported by the center of the work 

10.4 Milling Operations 


FIGURE 10.44 

A vertical milling 
machine (Courtesy of 
Manuel Pereira, 
photography specialist. 
University of 

FIGURE 10.45 

A universal dividing 
head (Courtesy of 
Manuel Pereira, 
photography specialist, 
University of 

400 10 Machining of Metals 

spindle) is rotated through the desired angle by rotating the index crank through an 
angle that is dependent upon the desired angle. The index crank is fixed to a shaft that 
is, in turn, attached to a worm-gear reducer with a ratio of 40 to 1 . Consequently, 40 
turns of the index crank result in only one full turn of the workpiece. This index plate 
has six concentric circles of equally spaced holes to assist in measuring and control- 
ling any fraction of revolution in order to crank the correct angle. The following equa- 
tion is used to determine the angle through which the crank is to be rotated in gear 


number of turns of index crank = : (10.9) 

number of teeth of desired gear 

We can see from Equation 10.9 that if the gear to be cut has 20 teeth, the index crank 
should be rotated two full turns each time a tooth space is to be produced. As a conse- 
quence, the workpiece will be rotated each time through an angle equal to 18°. Similarly, 
if the desired gear has 30 teeth, the index crank must be rotated 1 % turns each time. 


Grinding is a manufacturing process that involves the removal of metal by employing 
a rotating abrasive wheel. The wheel simulates a milling cutter with an extremely large 
number of miniature cutting edges. Generally, grinding is considered to be a finishing 
process and is used for obtaining high-dimensional accuracy and superior surface fin- 
ish. Grinding can be performed on flat, cylindrical, or even internal surfaces by em- 
ploying specialized machine tools, referred to as grinding machines. Obviously, 
grinding machines differ in construction as well as capabilities, and the type to be em- 
ployed is determined mainly by the geometrical shape and nature of the surface to be 
ground (e.g., cylindrical surfaces are ground on cylindrical grinding machines). 

Types of Grinding Operations 

Surface grinding. As the name surface grinding suggests, this operation involves 
the grinding of fiat or plane surfaces. Figure 10.46 indicates the two possible varia- 
tions: either a horizontal or a vertical machine spindle. With a horizontal spindle (see 
Figure 10.46a), the machine usually has a planer-type reciprocating table on which the 
workpiece is held. However, grinding machines with vertical spindles can have either 
a planer-type table like that of the horizontal-spindle machine or a rotating worktable. 
Also, the grinding action in this case is achieved by the end face of the grinding wheel 
(see Figure 10.46b), contrary to the case of horizontal-spindle machines, where the 
workpiece is ground by the periphery of the grinding wheel. Figure 10.46 also indi- 
cates the equations used to estimate the different parameters of the grinding operation, 
such as the machining time and the metal-removal rate. During the surface grinding 
operations, heavy workpieces are either held in fixtures or clamped on the machine 
table by strap clamps and the like, whereas smaller workpieces are usually held by 
magnetic chucks. 

10.5 Grinding Operations 


FIGURE 10.46 

Surface grinding: (a) horizontal spindle; (b) vertical spindle 




T - for each travel 

where: L is the length of the workpiece 
A = approach allowance 

where d is the depth of cut 

MRR = W-d-f 
where W is the width of cut 

i I Control of 
Grinding I [j I depth of cut 

feed, f 

r = 

L + 2A 


A = — for W = — up to D 
2 2 

A = s/W(D - W) for W < 
where W is the width of cut 
MRR = W-d-f 

Cylindrical grinding. In cylindrical grinding, the workpiece is held between centers 
during the grinding operation, and the wheel rotation is the source and cause for the 
rotary cutting motion, as shown in Figure 10.47. Cylindrical grinding can be carried 
out by employing any of the following methods: 

1. In the transverse method, both the grinding wheel and the workpiece rotate, and 
longitudinal linear feed is applied so that the entire length can be ground. The 
depth of cut is adjusted by the cross feed of the grinding wheel into the work- 

FIGURE 10.47 

Cylindrical grinding 




Control of 
depth of cut 


10 Machining of Metals 

2. In the plunge-cut method, grinding is achieved through the cross feed of the grind- 
ing wheel, and no axial feed is applied. This method can be applied only when the 
surface to be ground is shorter than the width of the grinding wheel used. 

3. In the full-depth method, which is similar to the transverse method, the grinding 
allowance is removed in a single pass. This method is usually recommended when 
grinding short, rigid shafts. 

Internal grinding. Internal grinding is employed for grinding relatively short holes, as 
shown in Figure 10.48. The workpiece is held in a chuck or a special fixture. Both the 
grinding wheel and the workpiece rotate during the operation, and feed is applied in 
the longitudinal direction. Any desired depth of cut can be obtained by the cross feed 
of the grinding wheel. A variation of this type of grinding is planatery internal grind- 
ing, and it is recommended for heavy workpieces that cannot be held in chucks. In this 
case, the grinding wheel not only spins around its own axis but also rotates around the 
centerline of the hole that is being ground. 

Centerless grinding. Centerless grinding involves passing a cylindrical workpiece, 
which is supported by a rest blade, between two wheels (i.e., the grinding wheel 
and the regulating or feed wheel). The grinding wheel does the actual grinding, 
while the regulating wheel is responsible for rotating the workpiece as well as gen- 
erating the longitudinal feed. This is possible because of the frictional characteris- 
tics of this wheel, which is usually made of rubber-bonded abrasive. As can be seen 
in Figure 10.49, the axis of the regulating wheel is tilted at a slight angle with the 
axis of the grinding wheel. Consequently, the peripheral velocity of the regulating 
wheel can be resolved into two components: workpiece rotational speed and longi- 
tudinal feed. These can be given by the following equations: 

'workpiece "regulating wheel * COS 01 

axial feed = V regu i at i ng wnee i x C x sin ot 


Note that C is a constant coefficient that accounts for the slip between the workpiece 
and the regulating wheel (C = 0.94-0.98). 

FIGURE 10.48 

Internal grinding 




10.5 Grinding Operations 


FIGURE 10.49 

Centerless grinding 


Peripheral velocity 
of regulating wheel 

Peripheral veloc 
of workpiece 




The velocity of the regulating wheel is controllable and is used to achieve any de- 
sired rotational speed of the workpiece. The angle a is usually taken from 1° to 5°; the 
larger the angle, the larger the longitudinal feed will be. When a is taken as 0° (i.e., 
the two axes of the grinding and regulating wheels are parallel), there is no longitudi- 
nal feed of the workpiece. Such a setting is used for grinding short shoulders or heads 
of workpieces having such features. 

Grinding Wheels 

Grinding wheels are composed of abrasive grains having similar size and a binder. The 
actual grinding process is performed by the abrasive grains. Pores between the grains 
within the binder enable the grains to act like separate single-point cutting tools. These 
pores also provide space for the generated chips, thus preventing the wheel from clog- 
ging. In addition, pores assist the easy flow of coolants so that heat generated during 
the grinding process is efficiently and promptly removed. 

Grinding wheels are identified by their shape and size, kind of abrasive, grain size, 
binder, grade (hardness), and structure. 

Shape and size of grinding wheels. Grinding wheels differ in shape and size, de- 
pending upon the purpose for which they are to be used. Various shapes are shown in 
Figure 10.50 and include the following types: 

1. Straight wheels for surface, cylindrical, internal, and centerless grinding 

2. Beveled-face or tapered wheels for grinding threads, gear teeth, and the like 

3. Straight recessed wheels for cylindrical grinding and facing 

4. Abrasive disks for cutoff and slotting operations when thickness is 0.02 to 0.2 
inch (0.5 to 5 mm) 

5. Cylindrical, straight, and flaring cups for surface grinding with the end of the 


10 Machining of Metals 

FIGURE 10.50 

Various shapes of 
grinding wheels 


Ml / 



^ i - 

i 1 



<r I h :> 



The main dimensions of a grinding wheel are the outside diameter D, the bore di- 
ameter d, and the height H. These dimensions vary widely, depending upon the grind- 
ing process for which the wheel is to be used. 

Kind of abrasive. Grinding wheels can be made of natural abrasives such as quartz, 
emery, and corundum or of industrially prepared chemical compounds such as alu- 
minum oxide or silicon carbide (carborundum). Generally, silicon-carbide grinding 
wheels are used when grinding low-tensile-strength materials like cast iron, whereas 
aluminum-oxide wheels are employed for grinding high-strength metals such as alloy 
steel and hardened steel. 

Grain size of abrasive used. As you may expect, the grain size of the abrasive parti- 
cles of the wheel plays a fundamental role in determining the quality of the ground sur- 
face obtained. The finer the grains, the smoother the ground surface is. Therefore, 
coarse-grained grinding wheels are used for roughing operations, whereas fine-grained 
wheels are employed in final finishing operations. 

Grade of the bond. The grade of the bond is an indication of the resistance of the 
bond to the pulling off of the abrasive grains from the grinding wheel. Generally, 
wheels having hard grades are used for grinding soft materials, and vice versa. If a 
hard-grade wheel were to be used for grinding a hard material, the dull grains would 
not be pulled off from the bond quickly enough, thus impeding the self-dressing 
process of the surface of the wheel and finally resulting in a clogged wheel and a burnt 
ground surface. The cutting properties of all grinding wheels must be restored period- 
ically by dressing with a cemented-carbide roller or a diamond tool to give the wheel 
the exact desired shape and remove all worn abrasive grains. 

Structure. Structure refers to the amount of void space between the abrasive grains. 
When grinding soft metals, large void spaces are needed to facilitate the flow of the re- 
moved chips. 

10.6 Sawing Operations 


FIGURE 10.51 

Standard marking system for grinding wheels 




symbol indicating 

exact kind of 

(use optional) 

Aluminium oxide - A 
Silicon carbide 

Standard Marking System Chart 


Coarse Medium/ Fine Very fine 








Dense to 

























private marking 

to identify wheel 

(use optional) 

V — vitrified 






Medium Hard 


Grade scale 

Binder. Abrasive particles are bonded together in many different ways. These include 
the use of vitrified bond, silicate, rubber, resinoid, shellac, or oxychloride. The vitrified 
bond is the most commonly used binder. 

Standard marking system. The standard marking system shown in Figure 10.51 is 
employed for distinguishing grinding wheels by providing all the preceding parameters 
in a specific sequence. 


Parting or cutoff operations can be performed on machine tools such as engine lathes, 
milling machines, and grinding machines. When cutting off is a basic operation in a 
large-volume production line, special sawing machines are required to cope with the 
production volume. 

Types of Sawing Teeth 

The cutting tool may take different forms, depending upon the type of sawing ma- 
chine used. The tool can be a blade, a circular disk, or a continuous band. However, 
all these tools are multiedged with several cutting edges (i.e., teeth) per inch. As can 
be seen in Figure 10.52, teeth can be straight, claw, or buttress. Each tooth, irrespec- 
tive of its form, must have a rake and a relief angle. Also, teeth are offset in order to 
make the kerf wider than the thickness of each individual tooth. This facilitates easy 


10 Machining of Metals 

FIGURE 10.52 

Types of sawing teeth 






Saw set 

(Top view 

movement of the saw blade in the kerf, thus reducing the friction and the generated 
heat. The maximum thickness is usually referred to as the saw set and is equal to the 
width of the resulting kerf. When selecting the cutting speed and the number of teeth 
per inch, several factors have to be taken into consideration, such as the cutting tool 
material, the material of the workpiece, the tooth form, and the type of lubricant 
(coolant) used. 

Sawing Machines 

Sawing machines differ in shape, size, and construction, depending upon the purpose 
for which they are to be used. They can be classified into three main groups. 

Reciprocating saw. In a reciprocating saw, a relatively large hacksaw blade is me- 
chanically reciprocated. Depending upon the construction of the saw, the cutting 
blade may be either horizontal or vertical. Each cycle has a working (cutting) stroke 
as well as an idle stroke. Consequently, this type of saw is considered to be a low- 
productivity saw and is used only in small shops with low-to-moderate production 

FIGURE 10.53 

Basic idea of a band 


Movable pulley 

to adjust tension 

in the band 

Band blade 

Driving pulley 

10.7 Broaching Operations 


Circular saw. The cutting tool in a circular saw is a circular disk, with the cutting 
teeth uniformly arranged on its periphery. It looks like the slitting cutter used with 
milling machines. Although it is highly efficient, it can only be used for parting or cut- 
off operations of bar stocks or rolled sections. 

Band saw. The highly flexible and versatile band saw employs a continuous-band 
sawing blade. As can be seen in Figure 10.53, the band-saw blade is mounted on two 
pulleys, one of which is the source of power and rotation. Each machine has a flash- 
welding attachment that is used to weld the edges of the band-saw blade together 
after adjusting the length, thus forming a closed band. Band saws can be used for 
contouring and for large-volume cutoff operations. Loading and unloading of the bar 
stock as well as length adjustment are done automatically (by special attachments in 
the latter case). 


Broaching is a metal-removing operation in which a multiedge cutting tool, like that 
shown in Figure 10.54, is used. In this operation, only a thin layer or limited amount 
of metal is removed. Broaching is commonly used to generate internal surfaces or 
slots, like those shown in Figure 10.55, that are very difficult to produce otherwise. 
However, it can also be used for producing intricate external surfaces that require tight 

FIGURE 10.54 

A broaching tool 










FIGURE 10.55 

Different shapes 
produced by broaching 

This indicates 
the original 
surface before 

408 10 Machining of Metals 

Broaching Machines 

A broaching machine is simply comprised of a sturdy frame (or bed), a device for locat- 
ing and clamping the workpiece, the cutting tool, and a means for moving the cutting 
tool (or the workpiece). The commonly used types of broaching machines are as follows: 

1. Pull-type machines, in which the broaching tool is withdrawn through the initial 
hole in the tightly clamped workpiece 

2. Push-type machines, in which the broaching tool is pushed to generate the 
required surface 

3. Sutface-broaching machines, where either the tool or the workpiece move to 
generate the desired surface 

4. Continuous-broaching machines, where the workpiece moves continually over a 
fixed broaching tool in a straight or circular path 

Advantages and Limitations 
of Broaching Operations 

It is important to know the advantages and limitations of broaching in order to make full 
use of the potential of this operation. The advantages include the high cutting speed and 
high cycling time, the close tolerances and superior surface quality that can easily be 
achieved, and the fact that both roughing and finishing are combined in the same stroke 
of the broaching tool. Nevertheless, this operation can be performed only on through 
holes or external surfaces and cannot be carried out on blind holes, for example. Also, 
broaching involves only light cuts and, therefore, renders itself unsuitable for operations 
where the amount of metal to be removed is relatively large. Finally, the high cost of 
broaching tools and machines, together with the expensive fixturing, make this opera- 
tion economically unjustifiable unless a large number of products are required. 


The need for nontraditional machining processes came as a result of the shortcomings 
and limitations of the conventional, mechanical, chip-generating processes. Whereas 
conventional processes can be applied only to soft and medium-hard materials, very 
fine features of extremely hard materials can be produced using the nontraditional ma- 
chining processes. There are a variety of nontraditional processes, and each has its own 
set of advantages and fields of application. Following is a brief discussion of each of 
the nontraditional processes commonly used in industry. 

Ultrasonic Machining 

Ultrasonic machining is particularly suitable for machining hard, brittle materials be- 
cause the machining tool does not come in contact with the workpiece. They are sep- 
arated by a liquid (vehicle) in which abrasive grains are suspended. Equal volumes of 

10.8 Nontraditional Machining Operations 409 

water and very fine grains of boron oxide are mixed together to produce the desired 
suspension. Ultrasonic energy applied to the tool results in high-frequency mechanical 
vibrations (20 to 30 kHz). These vibrations impart kinetic energy to the abrasive 
grains, which, in turn, impact the workpiece and abrade it. The machining tool must be 
made of a tough ductile material such as copper, brass, or low-carbon steel so that it 
will not be liable to fretting wear or abrasion, as is the case with the workpiece. Ultra- 
sonic machining is employed mainly in making holes with irregular cross sections. 
Both through and blind holes can be produced by this method. 

Abrasive-Jet Machining 

In abrasive-jet machining, liquids in which abrasive particles are suspended and 
pumped under extremely high pressure out from a nozzle. The resulting jet stream is 
then employed in processes like deburring, drilling, and cutting of thin sheets and sec- 
tions. The process is particularly advantageous when cutting glass and sheets of com- 
posites. The shortcomings of this process involve the problems associated with using 
high-pressure pumps and the relatively slow feed rate employed. 

Chemical Machining 

Chemical machining involves attacking the surfaces of the workpiece to be machined 
with a chemical etch that reacts with the metal and dissolves the resulting chemical 
compound. The procedure consists of first covering the surfaces of the workpiece that 
are not to be machined with neoprene rubber or enamel and then dipping the work- 
piece into a basin of the appropriate chemical etch. Very fine details can be etched by 
this method, and the quality of the machined surface is high and free from any chips. 
A further advantage of this process is that it does not result in any work-hardening. 

Electrochemical Machining 

The mechanism with which electrochemical machining (ECM) takes place is recipro- 
cal to that of the electroplating process, although similar equipment is used in both 
cases. In electrochemical machining, the workpiece is connected to the anode, while 
the cathode is connected to a copper ring that is used as the machining tool. Low- 
voltage, high-amperage direct current is used, and an electrolyte is pumped into the 
small gap between the workpiece and the copper ring. As is the case in electroplating, 
the amperage plays an important role in determining the rate of metal transfer from the 
anode to the cathode (i.e., the rate of metal removal during the electrochemical ma- 
chining process). Electrochemical machining can be applied to all electrically conduc- 
tive metals, including hardened alloy steel and tungsten, and it is particularly 
advantageous for machining thin sheets of nickel and titanium. 

Electrodischarge Machining 

Electrodischarge machining (EDM) is used for producing parts having intricate 
shapes, and it can be applied to all metallic materials, whether they are ductile or brit- 
tle. It cannot, however, be used with ceramics, plastics, or glass. Metal removal takes 


10 Machining of Metals 

FIGURE 10.56 

The EDM process 

Dielectric liquid 
(coolant) in 

Dielectric liquid 
2 *- out 


place as a result of an electric arc between the electrode and the workpiece, which 
are kept apart. A dielectric liquid like kerosene is pumped through the small gap of 
about 0.02 inch (0.5 mm) between the electrode and the workpiece, as shown in Fig- 
ure 10.56. The dielectric liquid also acts as a coolant and a flushing medium to whip 
away the removed metallic dust. 

The electrode is usually made of a material that can easily be shaped, such as cop- 
per, brass, graphite, or a copper-tungsten mixture. The electrode must be given a shape 
that fits exactly into the desired final cavity. Consequently, intricately shaped parts can 
easily be produced by this method, which has gained widespread industrial application 
in the manufacture of tools, metal-forming and forging dies, plastic, and die-casting 

FIGURE 10.57 

The concept of wire 

Chapter 10 Review Questions 


FIGURE 10.58 

Some stamping dies 
cut by wire EDM 
(Courtesy of Capitol 
Concept and 
Engineering, a Member 
of Synergis 
Technologies Group, 
Grand Rapids, 

molds. Generally, it can be stated that the quality of the machined surface is dependent 
upon the number of sparks (electric arc sparks) per second (they range between 3000 
and 10,000). 

A new version of EDM is shown in Figure 10.57, wherein the conventional elec- 
trode is replaced by a tensioned wire of copper or tungsten that is guided by a CNC 
system to trace any desired contour. This process has revolutionized the tool and die- 
making industry. Whereas sharp corners are avoided in tools manufactured by conven- 
tional processes to prevent breakage or cracking during subsequent heat treatment, 
wire EDM can cut heat-treated steels directly to the desired shape. Therefore, large 
dies having intricate shapes and sharp corners can be produced by this technique. Fig- 
ure 10.58 shows some stamping dies that were made by wire EDM. 

Review Questions 


1. What are the three motions necessary to gener- 
ate a surface during machining operations? 

2. List six different types of lathes. 

3. What are the main elements of an engine lathe? 

4. Why is the spindle of an engine lathe hollow 
and why does it have a Morse taper? 

5. Discuss briefly the construction of the tailstock. 

6. What is the main function of the carriage? 

7. How does the carriage receive its motion? 

8. Use sketches to explain the difference between 
a turret lathe and an engine lathe. 

9. How do you specify a lathe? 

10. What are the conditions for proper tool hold- 


10 Machining of Metals 

11. Use sketches to illustrate the following: turning 31. 
tools, facing tools, cutoff tools, thread-cutting 

tools, form tools. 32. 

12. What precautions should be taken when sup- 
porting a workpiece during lathe operations? 33 

13. When should a workpiece be held between two 

14. When is it necessary to hold a workpiece in a 34. 

15. When would a workpiece be mounted on a 35. 
faceplate? 36. 

16. How would you hold a disklike workpiece that 
has to be machined on both sides? 

17. What do the machining marks look like in 37. 
cylindrical turning? 

18. How is the axial feed provided and how is the 38. 
depth of cut controlled in cylindrical turning? 

19. What do the machining marks look like in fac- 39. 
ing operations? 

20. What are the suitable work-holding devices for 40. 
facing operations? 

21. List three methods that can be employed to 41. 
generate a tapered surface. 42. 

22. What provides the feed during thread cutting? 

23. Describe knurling. 43. 

24. What are the variables that affect the optimal 44. 
value of the surface cutting speed? 45 

25. Discuss the considerations that must be taken 
into account when designing turned parts. 

26. What is the difference between shaping and 46. 

27. What kind of surfaces can be produced by 47. 
shaping and planing operations? 

28. Explain the working principles of the quick- 
return mechanism. 48. 

29. List the commonly used types of drills and dis- 
cuss the applications of each. 49. 

30. What is meant by the tool point angle? How 

does the workpiece material affect the optimal 50. 

value for this angle? 

List some other hole-making operations and 
discuss the applications of each. 

Discuss the considerations that must be taken 
into account when designing drilled parts. 

List the various types of drilling machines and 
discuss the characteristics and fields of applica- 
tion of each. 

What is a jig? When is the use of jigs recom- 

Define milling. 

Why are the permissible cutting speeds in 
milling four times higher than those for turn- 

Differentiate between up milling and down 

List the various types of milling cutters and dis- 
cuss the applications of each. 

List the various types of milling machines and 
discuss the applications of each. 

What is the function of a universal dividing 

Define grinding. 

List the types of grinding operations and dis- 
cuss the applications of each. 

Of what are grinding wheels composed? 

How can grinding wheels be identified? 

List the different types of sawing machines and 
discuss the constructional features as well as 
the fields of application of each. 
Use sketches to illustrate the types of teeth of 
saw blades. 

Define broaching. When is the use of this 
process recommended? Discuss the advantages 
and limitations of broaching operations. 

Why did the need arise for nontraditional ma- 
chining operations? 

How is ultrasonic energy employed to machine 

Explain the working principles of abrasive-jet 

Chapter 10 Problems 


51. When is the use of chemical machining recom- 

52. Discuss the advantages and limitations of elec- 
trochemical machining. 

53. Explain the working principles of electrodis- 
charge machining. Discuss the advantages and 
limitations of this process. 

54. Discuss the advantages of wire EDM. 



1. It is required to maintain a cutting speed of 120 
feet per minute (37 m/min.) in a turning opera- 
tion. If the initial workpiece diameter is 3.25 
inches (82 mm) and the depth of cut is 0. 1 inch 
(2.5 mm), calculate the rpm of the spindle dur- 
ing the third and sixth passes. 

2. A 24-inch-diameter (600-mm) part with a 6- 
inch-diameter (150-mm) hole in the center is to 
be faced starting at the outside. The rotational 
speed of the spindle is 7 revolutions per second, 
the depth of cut is 0.15 inch (3.75 mm), and the 
feed is 0.01 inch per revolution (0.25 mm/rev). 
Calculate the machining time as well as the 
maximum and minimum rate of metal removal. 

3. Two thousand bars that are 3.25 inches (81 mm) 
in diameter and 12 inches (300 mm) long are to 
be turned down to 2.75-inch (69-mm) diame- 
ters. Heavy cuts followed by a light finishing 
cut are to be used. For finishing, the feed is 
0.005 inch (0.125 mm), the cutting speed is 300 
feet per minute (90 m/min.), and the depth of 
cut is 0.07 inch (1.75 mm). Two roughing cuts 
(two passes) are required, where the cutting 
speed is only 200 feet per minute (60 m/min.) 
and the feed is 0.01 inch (0.25 mm). Calculate 
the overall production time when the time taken 
to return the tool to the beginning of cut is 1 5 
seconds and the load/unload time is 2 minutes. 

4. A bronze bushing is 2 inches (50 mm) in diam- 
eter, is 3 inches (75 mm) long, and has a central 
hole of 1.25 inches (31.25 mm). It is to be pro- 
duced on a lathe, starting with a solid bar stock 


having a 2-inch (50-mm) outer diameter. Esti- 
mate the production time per piece. Take the 
feed as 0.01 inch per revolution (0.25 mm/rev) 
and the cutting speed as 200 feet per minute (60 
m/min.). Assume any missing data. 

A workpiece having a length of 3 inches (75 
mm) is to be taper-turned by offsetting the tail- 
stock. If the maximum diameter of the work- 
piece is l'/g inches (28.125 mm) and the 
minimum diameter is 1.0 inch (25 mm), calcu- 
late the amount of offset. 

A 10-inch-diameter (250-mm) part having a 4- 
inch-diameter (100-mm) hole is to be bored for 
4 inches (100 mm) of its length to a diameter of 
4.4 inches (110 mm). A depth of cut of 0.08 
inch (2 mm) is to be used with a feed of 0.004 
inch (0. 1 mm) and a cutting speed of 330 feet 
per minute (100 m/min.). If it takes 15 seconds 
to return the tool to the starting point and set the 
depth of cut, calculate the time required to com- 
plete this job. 

A part is to be tapered in such a manner as to 
have the following dimensions: 

Total length: 
Tapered length: 
Large diameter: 
Small diameter: 

6 inches (150 mm) 
1 '/2 inches (62.5 mm) 
1.0 inch (25 mm) 
0.625 inch (16 mm) 

Calculate the tailstock offset. 

8. How far must the tailstock be offset to cut a 
0.5-inch-per-foot (41.7-mm/m) taper on an 8- 
inch-long (200-mm) workpiece? 


10 Machining of Metals 

9. In a drilling operation, the desired depth of the 
hole is 1 inch (25 mm), the drill size is 0.4 inch 
(10 mm), the rpm is 100, and the feed is 0.01 
inch (0.25 mm). Calculate the cutting speed and 
estimate the drilling time. 

10. A standard twist drill is used to drill a number 
of 3/8-inch (9.5-mm) through holes in a 5/8- 
inch-thick (16-mm) SAE 1020 steel plate. Cut- 
ting speed is 60 feet per minute (18.3 m/min.), 
and feed is 0.004 inch (0.1 mm). Calculate the 
time required for drilling each hole and the 
metal-removal rate. 

11. It is -required to drill a 1 -inch-deep (25-mm) 
hole in each of 75,000 cast-iron blocks. If the 
rotational speed is 600 rpm and the feed is 
0.002 inch (0.05 mm), estimate the required 
working hours. Assume that it takes 30 seconds 
to load and unload the part and that 15 seconds 
must be allowed each time the drill bit is 
changed. Take the number of bit changes as 10. 

12. In a drilling operation, the feed rate is 1 inch 
per minute (25 mm/min.), the cutting speed is 
37.2 feet per minute (12 m/min.), and the di- 
ameter of the hole to be drilled is 0.6 inch (15 
mm). What is the feed? 

13. In milling a step 1/8 by 1/8 inch (3.18 by 3.18 
mm) in a 2-inch-long (50-mm) workpiece, a 
two-fluted 1/2-inch-diameter (12.5-mm) end 
mill is used. The rotational speed is 700 rpm, 
and the feed is 0.006 inch per tooth (0.15 mm 
per tooth). Estimate the milling time and the 
metal-removal rate. 

14. In a face milling operation, the depth of cut is 
1/4 inch (6 mm), and the table moves at 0.2 
inch per second (10 mm/s). The width of the 
workpiece is 2.0 inches (50 mm), and the cutter 
has a diameter of 3.25 inches (81 mm). The ro- 
tational speed of the cutter is 120 rpm, and the 
feed is 0.01 inch per tooth (0.25 mm per tooth). 
If the length of the workpiece is 10 inches (250 
mm), calculate the number of teeth of the cut- 

ter, the metal-removal rate, and the milling 


15. An 18-tooth, 1 -inch-wide (25-mm) HSS cutter 
having a diameter of 4 inches ( 1 00 mm) is to be 
used in slot milling a 10-inch-long (250-mm) 
workpiece. If the desired depth of slot is 0.24 
inch (6 mm), the cutting speed is 93 feet per 
minute (30 m/min.), and the feed is 0.005 inch 
per tooth (0.125 mm per tooth), estimate the 
milling time. 

16. Estimate the machining time in face milling 
given the following data: 


Rotational speed: 
Depth of cut: 

Length of 
Cutter width: 

20 teeth and 4 inches 
(100 mm) in diameter 
300 rpm 

0.25 inch (6 mm) 
0.001 inch per tooth 
(0.025 mm per tooth) 

20 inches (500 mm) 
larger than that of the 

17. The recommended feed for milling a kind of 
steel is 0.01 inch per tooth (0.25 mm per tooth) 
when using a helical milling cutter with 20 
teeth. If the cutting speed is 70 feet per minute 
(22.5 m/min.) and the cutter diameter is 4 
inches (100 mm), calculate the feed rate of the 

18. When gear-cutting processes are to be per- 
formed on a universal milling machine by using 
the indexing head, explain the procedure in 
each case when the gear has the following 
number of teeth: 

a. 20 teeth 

b. 32 teeth 

c. 22 teeth 

d. 15 teeth 

Chapter 11 

Product Cost 


As mentioned in Chapter 1, the production turn cannot continue unless the 
manufactured products are successfully marketed, the fixed and working capi- 
tal is recovered, and a profit is made. Cost plays a vital role in the marketing 
process because it provides the information required to set up the selling price 
of a product. An overpriced product cannot penetrate the market and will even- 
tually lose out to similar but more competitively priced products. Underestima- 
tion of the production cost may result in products sold at a loss and, 
consequently, financial problems for the manufacturing corporation. 

Because our main concern is design for manufacturing and because design 
is an open-ended process that yields more than one workable solution, a logi- 
cal criterion for evaluating these "solutions" or "designs" would certainly be 
the cost required to bring each design into being and manufacture the product. 
Therefore, it is fair to state that cost estimation is initiated by, linked to, and 
follows the product design in order to ensure the profitability of new products. 
Cost is also used to determine the most economical operation or sequence of 
operations for manufacturing a product, and it can be used as a means for es- 
tablishing a cost-reduction program aimed at manufacturing the product so that 
it can be priced more competitively. 



11 Product Cost Estimation 


Costs can be classified in different ways based on their relationship to the production 
volume and the nature of the manufacturing operations. The first, and most logical, 
way to classify costs is to split them into two groups: capital costs and operating costs. 
As the name suggests, capital costs are incurred because of buildings, production ma- 
chinery, and land. It is important to remember, when carrying out cost estimation, that 
buildings and machinery are depreciable (i.e., they tend to lose most of their value with 
time) whereas land is not. Operating costs are "running" costs that reoccur as long as 
the plant is in operation. 

Another way to classify costs is to view them as belonging to one of two cate- 
gories: fixed costs, which are independent of the production volume, and variable 
costs, which are dependent on it. Here are some examples of cost elements in the 
fixed-cost category: 

Depreciation on buildings, machinery, and equipment 

Insurance premiums (fire, theft, flooding, and occupational hazards) 

Property taxes (sometimes states and communities give tax breaks to industrial 
corporations to attract them to a region) 

Interest on investment (money borrowed from a bank, as explained in Chapter 1 ) 

Factory indirect-labor cost (wages of security, personnel, secretarial, clerical, jani- 
torial, and financial staffs) 

Engineering cost (high-level engineering jobs and R&D expenses) 

Cost of rentals, if any (sometimes the building itself is rented, or some equipment 
may be rented for a short term) 

Cost of general supplies (supplies used by the factory indirect-labor force) 

Management and administrative expenses (salaries paid to corporate staff, plus 
legal expenses or salaries paid to legal staff) 

Marketing and sales expenses (salaries and wages paid to marketing and sales 
staffs, transportation and delivery expenses, rentals of warehouses, if any) 

The following cost elements fall in the variable-cost category: 

• Cost of materials 

• Cost of labor (including production supervision) 

• Cost of power (electricity, gas, or fuel oil) and utilities (water, sewer, etc.) 

• Cost of maintenance of production equipment 

The logical way to determine the total cost of a product is to add up all the cost 
elements, as is indicated by the bar diagram in Figure 11.1. 

11.1 Costs: Classification and Terminology 


FIGURE 11.1 

Elements contributing 
to the total cost of a 

Direct labor 

Direct material 



Product manufacturing cost 


In product cost estimation, the use of spreadsheets is extremely important, and the 
student is, therefore, encouraged to learn about and practice using them. The value of 
each cost element can be inserted in a spreadsheet cell, and, as an alternative, the for- 
mula for computing a cost element can be employed in the cell if the value of that el- 
ement is not known. A further advantage of using spreadsheets is the ease with which 
alternative designs can be compared and evaluated from the point of view of cost. Fig- 
ure 11.2 shows a spreadsheet where the specific cost elements and the total cost of 

FIGURE 11.2 

A spreadsheet that compares the cost of four alternative designs 

\. Element of 

\. Product 

Design \Tost 

Alternatives >v 

Direct labor 
cost ($) 

Direct material 
cost ($) 

expenses ($) 

expenses ($) 



cost ($) 

Design 1 






Design 2 






Design 3 






Design 4 






418 11 Product Cost Estimation 

each design are shown in columns that make comparisons and conclusions easy. From 
a quick look at the figure, for example, it is not difficult to realize that design 4 is the 
optimal choice based on cost. 

As easy as it may look, however, it is impossible to carry out the process of de- 
termining the total cost of a product unless rational procedures and analyses are em- 
ployed to overcome two main problems. First, some costs cannot be directly assigned 
or traced to any particular product, but rather are spread over the entire factory; they 
are, therefore, labeled as "indirect" costs. In other words, the problem is how to calcu- 
late the cost share of a product from the salary of a receptionist or secretary. Second, 
we do not actually know the time taken to produce a design because that design has 
not been manufactured. It is the objective of this chapter to provide adequate answers 
to these two problems and to show the student how to independently carry out an en- 
gineering cost analysis for any desired design. 

Now, with our stated goal as product cost estimation that is based on and begins 
after a detailed product design is available, we must develop highly accurate cost esti- 
mates that are suitable for submission on a bid or purchase order. This type of estimate 
is referred to as a detailed estimate and must have a level of accuracy of ±5 percent. 
The American Association of Cost Engineers came up with a list of five types of cost 
estimates, each having a certain level of accuracy, a different approach, and recom- 
mended applications. For example, the first type, a rough estimate, has an accuracy of 
±40 percent and is based on indexing and modifying the cost of existing similar de- 
signs. It is, therefore, recommended for initial feasibility studies that are used to decide 
whether or not a probable profit justifies pursuing a project any further. Other types of 
cost estimates fall between the two extremes of rough to detailed and are, conse- 
quently, recommended for applications that depend upon their level of accuracy. 

Before we attempt to gain a deeper insight into each of the elements that con- 
tribute to the total cost of a product, we must consider some factors that, if overlooked, 
may adversely affect the accuracy and validity of the estimate. For instance, the cost 
estimate cannot be held valid for more than a few months if the inflation rate in the 
country of production is noticeable. Further complications arise when the time taken to 
construct the plant and manufacture the products is so long that initial costs are af- 
fected by inflation (meaning that the money loses its purchasing power). Also, there 
are sometimes uncertain and unforeseen expenses, or contingency factors, a typical ex- 
ample being the escalation of R & D costs when developing new technology for man- 
ufacturing the products. 


Labor can be either direct or indirect: Direct labor is explicitly related to the process 
of building the design, whereas indirect labor involves the work of foremen, stock- 
room keepers, and so on. We will be concerned here with the cost analysis of direct 
labor because the indirect-labor cost is generally covered by factory overhead costs in 
the form of a percentage of the cost of direct-labor hours. At this point, our goal is to 
estimate the labor time for building a design and then to multiply that time by the com- 

11.2 Labor Cost Analysis 419 

bined value of wages and fringe benefits, which is usually called gross hourly cost. 
Note, however, that wages are sometimes not based just on attendance, but also on per- 
formance (i.e., incentives are given when the hourly output exceeds a certain estab- 
lished goal). 

Methods for Measurement of Time 

Although there are quite a few approaches for the measurement and estimation of labor 
time, two methods are well accepted in industry and will, therefore, be covered here. 
The first method is based on time and motion study, a modern subject that was estab- 
lished by the eminent American engineer Frederick W. Taylor of Pennsylvania in the 
early twentieth century. This method, which is favored by industrial engineers, in- 
volves breaking down the manual work of an operation into individual simple motions. 
A typical motion is, for example, "reach and grab" (i.e., the worker stretches his or her 
hand to reach a tool and grab it). The operation is then converted into a tabular form 
that includes the entire sequence of basic motions that comprise the desired manual op- 
eration. Because these basic motions were thoroughly studied by industrial engineers 
and because time measurements were taken and standardized for each basic motion, 
our job is fairly easy. It is just to read, from published data that is readily available, the 
standard time unit for each motion included in the manual operation and insert it into 
a time and motion study table. By summing up all the time values, the total time re- 
quired by an average worker to carry out the operation can be obtained. This time is 
modified by dividing it by the efficiency or the "rating" of the actual worker to account 
for interruptions and fatigue. This approach has the clear advantage of including a 
mechanism for rationalization of the operation by eliminating unnecessary and waste- 
ful motions. The procedure described here is used to estimate the time for a single op- 
eration only and must be carried out for all operations required to produce a design. 
Consequently, our first step (after the design is available) is actually to prepare detailed 
process routing sheets indicating all operations included in the production of the part. 
It is clear that the time and motion study method requires a considerable amount of 
work, but it has usually been found that the effort and time spent are well worthwhile. 
The second method is based on the historical value of time. Time cards for a sim- 
ilar design that has already been built are obtained and studied in order to determine 
the number of "man-hours" required to do the job (a man-hour is a unit indicating the 
output of one person working for one hour). Data analysis using spreadsheets is then 
employed to make the necessary adjustments, taking into consideration such factors as 
the skill level of the workers, the workplace environment, and cost escalation, if any. 

The Learning Curve 

It is a well-known fact that doing a job for the first time requires more time from 
a worker than when doing it for the fifth time, for example. This is evidently due 
to the phenomenon of self-teaching while performing the work, which, in turn, 
leads to a gain in work experience and thus a shorter time for doing that job. This 
is what is usually referred to as the learning curve or product improvement curve. 
As can be seen in Figure 11.3, the learning curve indicates the relationship between 


11 Product Cost Estimation 

FIGURE 11.3 

The learning curve 

<D - — . 


•= c 

C => 

■2. £• 



70% learning curve 

100 -" 

80 - 

\l00 x 0.7 

60 - 

^\70 x 0.7 

40 - 

49 x 0.7 

20 - 


1 1 1 1 1 

1 1 1 

1 1 1 1 1 

1 1 1 

12 3 4 5 6 7 8 

Cumulative production (arbitrary units) 

the production time (say, per unit) and the total cumulative production. Learning 
curves are usually labeled by percentages. In Figure 11.3, for example, we have a 
"70% learning curve." This simply means that whenever the total cumulative pro- 
duction doubles, the production time per unit decreases by 30 percent. This de- 
scription suggests that the learning curve is exponential and can, therefore, be 
expressed by the following equation: 

t = t„ x p" 


where: t is the production time per unit after producing a number of units equal 
to p 

t is the time taken to produce the first unit 

p is the total cumulative production (i.e., total number of units produced) 

n is a constant that depends on the constant percentage reduction char- 
acterizing a learning curve (e.g., -0.5146 for 70 percent curve, -0.322 
for 80 percent curve, -.152 for 90 percent curve) 

Taking the concept of the learning curve into consideration results in reducing the total 
production time (and cost) as estimated by the previously mentioned conventional 
methods, and adjustments have to be made. By simple mathematics, the total produc- 
tion time T is given by 

Substituting the value of / from Equation 11.1, we obtain 
T=t Xp n Xp = t Xp n+1 



11.3 Material Cost Analysis 421 

Labor Laws 

Some legal aspects must be considered when estimating the cost of labor. Federal and 
state laws regulate wages; for instance, minimum wage is $5.15 per hour. Also, the 
number of regular working hours per week is 40 and per day is limited to 8. If either 
(or both) of these is exceeded, a production worker must be paid at a rate equal to 150 
percent of his or her regular hourly wage for the number of working hours that exceed 
the limits. Labor hourly wage rates in the United States, as well as other important rel- 
evant information, are compiled and published by the Bureau of Labor Statistics for 
various industrial sectors and can be obtained from the Department of Labor. 

It is also important to remember that the labor cost is not limited to the money 
spent on wages but must include the fringe benefits paid to workers. Fringe benefits 
differ for different companies and may include any of the following: 

Health, dental, and life insurance premiums 

Expenses of insurance against job injury and hazards 

Holidays, paid vacations, and sick leave when actually taken 

The company's share in pension plans 

Payments to union stewards (if the company is unionized) 

Profit-sharing bonus money (that part of the company's profits paid to workers) 

Fringe benefits can amount to as much as 30 percent or more of wages. It is, therefore, 
fairly common to combine wages with fringe benefits into the so-called gross hourly 
cost in order to avoid repetitive calculations. 


Amount of Material Used 

In order to carry out material cost analysis for a product, the amount of material used 
to manufacture that product first must be determined and then multiplied by the price 
of material, (or materials) in the form of dollars per unit weight or volume. Conse- 
quently, some documents must be available before this cost-estimating operation is ini- 
tiated. These include, for example, the bill of materials, the engineering design 
documents, and printouts of inventory data. The engineer is then in a position to de- 
termine the bill of material required to build the design — a process that is sometimes 
referred to as the quantity survey. It is not difficult to see that the raw material required 
is more than the amount of material indicated in the design blueprints. This difference 
includes but is not limited to the waste during manufacturing, which, in turn, depends 
upon the specific production processes employed. For example, when the part is to be 
produced by machining, the amount of material removed from the stock in the form of 
chips must be added to the amount calculated from the design drawing. When the part 
is to be produced by casting, the material in the risers, sprues, and gating system must 
be added, as well as the material that is removed by machining (the "skin," drilled 
holes, slots, recesses, etc.). The same rule of adding the waste applies to the various 
manufacturing processes of forging, press working, and extrusion. Note that the waste 

422 11 Product Cost Estimation 

is sometimes sold to junk dealers for recycling and the money paid for it must accord- 
ingly be subtracted from the cost of material. In some cases, however, the waste is val- 
ueless and is disposed at a cost that has to be added to the cost of material. In addition 
to waste, losses due to scrap (i.e., defective parts that are functionally obsolete) must 
be added. Again, scrap can be sold at a cost for recycling or may require disposal and 
additional expenditure. Losses may also include "shrinkage," which is a loss caused by 
environmental conditions (e.g., oxidation of steel or decrease in the volume of lumber 
when it dries). 

The amount of raw material required to build the design and calculated according 
to the approach just described is a part of a cost analysis category known as direct ma- 
terial. This category can also include standard purchased items like nuts, bolts, 
springs, and washers. These items have no labor cost in our cost estimate, and the pur- 
chase price is considered as material cost (the company that makes these items has to 
consider the labor cost when estimating the selling price). Direct material also includes 
subcontracted items, which are assemblies or subassemblies manufactured outside the 
company and are supplied by an external subcontractor. Again, this does not include 
any labor cost in our estimate and must be categorized as material cost. 

Determination of the amount of direct material required to build a design can 
sometimes be complicated. Consider, for example, the case when plastic is injection 
molded into a mold that has several dissimilar cavities for producing different parts. 
The sprues and runners form the waste in the injection molding operation. The ques- 
tion is, How do we determine the share of each part from that waste? Let us consider 
dividing the amount of material of the runners equally between the parts in order to get 
the amount of waste for each product. Unfortunately, the results may be totally mis- 
leading, especially when some of these products are very small while others are large. 
In fact, this case is referred to as one of joint material cost and arises whenever there 
is a multiple-product manufacturing process where the tracing of the raw material 
share of each individual product is difficult. A well-accepted approach in this case is to 
agree upon a primary product and attribute most of the untraceable common expendi- 
tures to it. In addition to direct material, there is material that is consumed during the 
process of transforming the raw stock into useful products. It is usually both necessary 
and untraceable to a particular design and is, therefore, referred to as indirect material. 
Typical examples include lubricating oils, soaps, and coolants. As it is clear that direct 
mathematics will not work here, a simpler method, which has gained acceptance in in- 
dustrial cost-estimating practice, is to add this item to factory overhead costs. 

Purchasing Price of Material 

As previously mentioned, the purchasing price of material has to be known, in the 
form of dollars per unit weight or volume, in order to be able to estimate the total 
material cost to build the design. When the material used is contractual (i.e., pur- 
chased specifically for manufacturing a certain product), the actual purchase price 
can be directly employed in estimating the cost of material. However, when the ma- 
terial is taken from inventory, it is difficult to find its value and use it in the cost 
analysis because inventories usually contain various lots of the same material that 
have been purchased at different times at different prices. So, the question is, Which 

11.4 Equipment Cost Analysis 423 

price do we use in cost estimation? In fact, nobody can provide a precise answer to 
this question, and, therefore, a number of approaches have been adopted by different 
schools of thought in industry. Following is a brief summary of some of the com- 
monly used methods. 

First-in-first-out method. The first-in-first-out method is based on following the rule 
of issuing first the material that was purchased first (i.e., having the longest time in 
stock) to the factory for processing and using its purchase price in the cost analysis. A 
clear drawback of this method arises when the time between purchasing and process- 
ing of the material becomes long. The original price may not be a true representation 
of the current value of the material, thus resulting in an inaccurate cost estimate. 

Last-in-first-out method. In the last-in-first-out method, the material that was added 
last to the inventory is issued first to the factory. The material used can be from more 
than one purchased lot, with different costs for those lots — a fact that complicates the 
process of estimating the cost of materials. 

Current-cost method. The approach taken by the current-cost method is to use the 
cost of materials corresponding to the time when the estimate is prepared. Once again, 
material issued from the inventory that would have been purchased earlier or later may 
have a cost different from the current cost. 

Actual-price method. The actual-price method is based on calculating the amount of 
money originally spent to purchase the material used. If the material issued from the 
inventory belongs to the same lot, then the calculations are easy and simply take the 
original purchase price of material. However, when the material is taken from two (or 
more) lots having different purchase prices, an average or equivalent cost has to be 
used in the cost-estimating process. Here is the applicable equation: 


. 1 1 Cfit 

'-'equivalent ,; V*'-'*i 

X a, 

i = 1 

where: c, is the cost of material of lot i 

a, is the amount of material taken from lot /' 
i to n are used-lot serial numbers 


The cost of equipment belongs to the fixed-cost category, and the depreciation of ma- 
chinery as well as the interest on investment must be taken into account. It is often 
difficult, however, to get a quote for the cost of equipment, especially during the 
early phase of a feasibility study. Published data about the cost of equipment are. un- 
fortunately, not directly applicable due to inflation and devaluation of the purchasing 
power of the dollar, as well as the difference in capacities or ratings given in the pub- 
lished data and those of the required equipment. Adjustments must, therefore, be 

424 11 Product Cost Estimation 

made to account for factors that affect the validity of the published cost-of-equipment 
data, following are some of the methods used. 

Cost Indexing 

A cost index is an indication of the buying power of money at a particular time for a 
certain category of equipment and machinery. Accordingly, if the cost of equipment 
and the corresponding index are known at some initial time, the current cost can be de- 
termined if the current index is obtained: 

C c = C, (j) (11.5) 

where: C ( . is the current cost 

C, is the cost at some initial time 

I c is the current index 

/,• is the index at the same initial time as C, 

There are many published indexes, and each pertains to a certain area of applica- 
tion (e.g., chemical plants, consumer price). The Marshall and Swift cost index is the 
one to use for industrial equipment. It is readily available and has different values for 
the various kinds of industry. Care must, therefore, be taken not to use the index for 
the paper industry to calculate the cost of a steam turbine, for example. More impor- 
tantly, indexing does not hold true when there is a radical change in the technology. 
This is evidenced by the fact that the prices of many electronic products have actually 
decreased as a result of technology change. Also, further adjustments are sometimes 
needed to account for regional conditions because published indexes are indications of 
national averages. 

Size Effect 

Sometimes, it is possible to only get hold of the cost of a machine similar to the re- 
quired one but having a different size or rating. Corrections must, therefore, be made 
to that cost in order to obtain the cost of the desired machine. Consequently, a mathe- 
matical relationship between the cost and the capacity (size or rating) of capital equip- 
ment must first be established. As you may have guessed, the relationship is not linear 
due to the effect of the economy of scale on engineering design and production. The 
relationship can be given by the following empirical formula, which is usually referred 
to as the six-tenths rule: 

i c \0.6 

where: C 2 is the cost of capital equipment 2 
C, is the cost of capital equipment 1 
S 2 is the capacity (size or rating) of capital equipment 2 
Si is the capacity (size or rating) of capital equipment 1 

11.6 Overhead Costs 425 

Regression Analysis 

Statistical techniques are used for collecting factual data that is, in turn, subjected to 
mathematical analysis and curve fitting in order to establish a relationship between 
cost and the various parameters that affect it. Here is the general equation: 

C=C ^ P?> (11.7) 

j= i 

where: C is the cost 

C (> is a constant 

P, is the parameter i affecting the cost 

w, is a constant exponent for parameter i 

C , Pi, and irij are determined by mathematical and statistical methods. Although the 
formula is limited in scope to the specific equipment or system for which it is devel- 
oped, it has proven to be very useful in cost models because it elaborates the elements 
and parameters that contribute most to the cost. It also leads to the ability to minimize 
and optimize cost using simple mathematical manipulations such as those of differen- 
tial calculus. 


Engineering cost includes salaries for high-level engineering jobs as well as expen- 
ditures (whether salaries or general expenses) for R&D. Usually, the engineering 
cost for a product is considered as part of the factory overheads (or even as part of 
the corporate overheads in some cases). Nevertheless, it is sometimes estimated sep- 
arately based on previous records of existing similar products. In some cases (espe- 
cially when the product is supplied to the federal government), a firm is hired to do 
the engineering on a contractual basis. The contract may specify a lump sum for the 
engineering cost or may involve the true engineering cost plus a negotiated fee or 
profit of, say, 15 percent of the cost. In this latter case, the engineering cost must be 
accurately determined. 


Overhead costs are usually viewed by cost engineers as a burden because such 
costs cannot be directly or specifically related to the manufacturing of any partic- 
ular product or even to a particular category of the company's production. Over- 
head costs can be divided into two main groups: factory overheads and corporate 

426 11 Product Cost Estimation 

Factory Overheads 

Factory overheads include the previously mentioned engineering costs as well as other 
factory expenses that are not related to direct labor or material. An example would be 
the wages paid to personnel for security, safety, shipping and receiving, storage, and 
maintenance. The challenge here is how to calculate the "share" of each different prod- 
uct from these expenses. There are many approaches for charging these expenses to the 
cost of the various products. Following are three bases upon which factory overhead 
costs can be allocated: 

1. The ratio between the direct-labor hours required to manufacture the product and 
the total number of direct-labor hours spent on the factory floor (this ratio, when 
multiplied by the total overhead expenses, yields the share of that product from 
the overhead cost) 

2. The ratio between the material cost of the product and the total cost of material 
consumed on the factory floor (again, the share of a product from overhead cost is 
the product of multiplication of this ratio by the total overhead expenses) 

3. The ratio between the space occupied by the production equipment (e.g., furnace 
or machine tool) and the total area of the factory floor 

The direct-labor-hours method is by far the most commonly used approach for 
allocating factory overhead costs. As is clear, the production volume has a tangible 
effect on the factory overhead cost per product. If the production is reduced to half 
its normal level, for example, without reducing the total overhead expenses, the over- 
head share of a product will automatically double. Consequently, it is always a good 
idea to carry out cost analysis for any potential product at different production vol- 
umes (i.e., percentages of full capacity of production lines). Note that increasing pro- 
ductivity results in a reduced number of direct-labor hours. This is sometimes 
misinterpreted by management, and decisions may be made to reduce the budget for 
maintenance and other factory overhead items. It is the duty of the manufacturing en- 
gineer to eliminate any misinterpretations on the part of management. An alternative, 
in this case, is to use a different basis for allocating the overhead costs and request- 
ing budgetary funds. 

Corporate Overheads 

Corporate overheads basically involve the cost of daily operation of the company be- 
yond the factory floor throughout the year. These expenses include, for example, the 
salaries and fringe benefits of corporate executives as well as those of the business, ad- 
ministrative, and legal staffs. Again, the commonly adopted approach is to obtain an 
overhead rate that is the product of dividing the total corporate overhead expenses by 
the total cost of direct labor. Knowing the direct-labor time and cost for manufacturing 
a product, you can easily calculate the corporate overhead cost using this overhead 
rate. It is worth mentioning that corporations may operate more than one plant from 
the corporate headquarters — a fact that has to be taken into consideration when calcu- 
lating both the corporate and the direct-labor costs in order to obtain the overhead rate. 

11.7 Design to Cost 



The preceding discussions reflect the usual or conventional sequence of preparing the 
design and then costing the product based on the information provided in that design. 
With increasing global competition, however, cost is becoming more and more the dri- 
ving force. Consequently, a need arises for costing a potential product before its design 
is completed or even made. This unusual approach is aimed at continuously improving 
the design in order to manufacture the desired product at a designated price that is 
equal to or less than the market price of the competitor's product. This "reverse" pro- 
cedure is known as design to cost and is gaining popularity in industry, especially with 
newly emerging methodologies such as reengineering. 

The process starts with benchmarking a given product, taking market price and 
quality as the judging criteria. By removing the retail profit, the manufacturing cost 
is obtained. Next, the various overhead rates that are well established in the company 
are employed to remove the different overhead cost items, yielding the prime cost. 
Then comes the difficult task of meticulously breaking down the prime cost among 
components, assemblies, and subassemblies. Favoring one component at the expense 
of another is a big mistake as setting a target cost below reasonable limits will make 
the design of the component virtually impossible. Once the target cost for a compo- 
nent is allocated, design begins using that target cost as an incentive for continuously 
improving the design. If the direct-labor cost, for example, is found to be less than 
the target, this will give some relief in the process of selecting materials. The same 
rule is applicable to subassemblies and assemblies (i.e., if the cost of a component is 
less than the target, this will give more flexibility when designing other components). 
When the design is finalized, it must be subjected to the conventional and accurate 
cost-estimating process. 

Review Questions 


1. Why is cost estimation of vital importance for a 
design engineer? 

2. What role does cost play in process planning? 

3. List two methods for classifying costs. 

4. List some important elements of fixed cost. 

5. List some important elements of variable cost. 

6. What are the two main problems that compli- 
cate cost estimation? 

7. Do all types of cost-estimating methods have 
the same accuracy? Explain why. 

8. Can you rely upon a cost estimate that was 
done last year? Why not? 

9. Assuming that the construction of the plant 
takes a long time, what effects would this have 
on the cost-estimating process? 

10. What is meant by direct labor? 

11. Is there any indirect labor? Explain. 


11 Product Cost Estimation 

12. What is the gross hourly cost? 

13. Explain the difference between the two cases of 
wages based on attendance and wages based on 

14. How can we measure the direct-labor time be- 
fore the product is actually manufactured? 

15. What are the pros and cons of the Industrial En- 
gineering approach for measuring the direct- 
labor time? 

16. Explain the time-card method for estimating 
the direct-labor time. 

17. What is the learning curve? What effects does 
it have on cost-estimating results? 

18. List some important labor laws that must be 
considered when estimating the cost of a 

19. List some common fringe benefits. 

20. What is the quantity survey? 

21. What are the sources of difference between the 
material in a product as indicated by the design 
drawing and the material actually consumed in 
the manufacture of that product? 

22. Explain the term indirect material. 

23. What is the joint material cost? Give examples. 

24. Why is it difficult to get the cost of unit mater- 
ial when the material is issued from inventory? 

25. Explain briefly the different methods used to 
obtain the cost of unit material when the mate- 
rial is issued from inventory. Give the pros and 
cons of each. 

26. What is a cost index? Why is it important in 
cost estimation? 

27. How can you calculate the cost of a machine 
with a known capacity if you know the cost and 
the capacity of another machine? 

28. Show how regression analysis and statistics can 
be employed in cost estimation. 

29. What is meant by engineering cost? How is it 

30. What are the different types of overhead costs? 

31. On what bases are factory overhead costs 

32. In some cases, increasing productivity might 
have an adverse effect on budget allocations. 
Explain how. 

33. Explain the concept of design to cost. 

34. What is the driving force for design to cost? 

35. What is the main problem encountered in the 
procedure of design to cost? 



A number of stock bars, each 3.25 inches (81 mm) in diameter and 12 feet (3.6 m) in 
length, are to be used to produce 2000 bars, each 2.75 inches (69 mm) in diameter and 
12 inches (300 mm) in length. The material cost is $1.05 per pound ($2.11/kg), and the 
density is 0.282 pound per cubic inch (789 kg/m 3 ). The total overhead and other ex- 
pense is $95,000. The total direct-labor expense for the plant is $60,000. Estimate the 
production cost for a piece. 


First, we have to calculate the production time per piece. Consequently, technical pro- 
duction data have to be either obtained or assumed. Following are some assumptions: 

Chapter 11 Problems 


• The facing dimension necessary for a smooth end finish is 1/16 inch (1.6 mm). 

• The width of the cutoff tool is 3/16 inch (4.76 mm). 

• The collet requires 4 inches (100 mm) of length for last-part gripping. 

• Heavy cuts are to be done followed by a light finishing cut: For two rough cuts, 
cutting speed is 200 feet per minute (60 m/min.) and feed is 0.01 inch (0.25 mm); 
for finishing, cutting speed is 300 feet per minute (90 m/min.) and feed is 

0.005 inch (0.125 mm). 

• The time taken to return the tool to the beginning of cut is 15 seconds. 

• Load (setup) and unload time is 1 minute. 

Machining Time 

Position tool to perform cutoff: 
Cutoff time: 

D + a 


radial feed rate 

3.25 + 0.75 


7t x 3.25 

200 x 12 x 0.01 

Position tool to carry out facing operation: 
Facing time: 

x60 = 

D + a 

2 x 71 x 3.25 

radial feed rate 300 x 12 x 0.005 


Position tool to perform first rough cut: 
First rough cut: 

12+ 1/16 + 3/16 + 4/16 

feed rate 

Position tool to perform second rough cut: 
Second rough cut: 
Position tool for finishing: 


feed rate 

Load/unload time per piece (1 minute): 
The total machining time per piece is 

15 seconds 

51 seconds 
15 seconds 

68 seconds 
15 seconds 

320 seconds 

15 seconds 
320 seconds 
15 seconds 

360 seconds 

does not count with 

this piece; it is included 

in the time for next 


60 seconds 

1254 seconds. 

430 11 Product Cost Estimation 

Cost of Labor/Piece 

. , . 12 x 12 

number of pieces produced from a single bar = — — — - — 

total length • . 

= — = 1 1 pieces 


number of stock bars = = 181.8= 182 bars 

total loading time = 2 x 182 x 60 (assuming loading time/bar = 2 minutes) 

share of each piece = = 10.9 — 11 seconds 

F 2000 

total average production time/piece = 1254 + 11 

= 1265 seconds (direct labor) 

1 96S 
cost of labor/piece = x $10/hr 

= $3.52 (assuming a CNC machine is used) 

Cost of Material/Piece 

We have to consider the waste. Assume no scrap as the operation is simple: 

182 x -(3.25) 2 x 12 x 12 x .282 x 1.05 
cost of material/piece = = $32.17 

Cost of Overhead/ Piece 

In this, all other costs are included: 

. , t 95,000 x 100 .„,,„ 
overhead rate = = 158.33% 

158 33 
overhead cost/piece = 3.52 x — ' = $5.57 

Total Cost/Piece 

total cost/piece = direct labor cost + material cost + overhead cost 

= 3.52 + 32.17 + 5.57 = $41.26 

Design Project 


Select a few of the design projects that supplement Chapters 3 through 7, preferably a 
project for each manufacturing process, and then carry out cost estimation for the 
product. You are strongly advised to obtain real values for the different cost elements 
(e.g., material) by contacting industrial companies and obtaining quotations. 

Chapter 12 

for Assembly 


Modern societies are undergoing continuous development, which necessi- 
tates large-scale use of sophisticated products like appliances, automobiles, 
and health-care equipment. Each of these products involves a large number 
of individual components, assemblies, and subassemblies that must be 
brought together and assembled into a final product during the last step of 
the manufacturing sequence. A rational design should, therefore, be con- 
cerned with the ease and cost of assembly, especially when given the fact 
that 70 to 80 percent of the cost of manufacturing a product is determined 
during the design phase. It is for this reason that the concept of design for 
assembly (DFA) emerged. It is simply a process for improving the product de- 
sign for easy and low-cost assembly. In other words, this assembly-conscious 
design approach not only focuses on functionality but also concurrently con- 
siders assemblability. 

Although the use of the term design for assembly is fairly recent, several 
companies can claim, in good faith, that they have developed and have been 
using guidelines for assembly-conscious product design for a long time. For in- 
stance, the General Electric Company published in 1960, for internal use only, 
the Manufacturing Producibility Handbook. It included compiled manufacturing 
data that provided designers in the company with information necessary for 
sound and cost-efficient design. Later, in the 1970s, research institutions and 
research groups started to become more and more interested in the subject 
when the Conference Internationale pour le Recherches de Production (CIRP) 


432 12 Design for Assembly 

established a subcommittee for that purpose and Professor Geoffrey Boothroyd 
began his pioneering research at the University of Massachusetts Amherst. 

The traditional approach for DFA has been to reduce the number of indi- 
vidual components in an assembly and to ensure an easy assembly for the re- 
maining parts through design modifications. When a design is altered in such 
a manner that two components are replaced by just a single one, the logical 
consequence is the elimination of one operation in manual assembly or a 
whole station of an automatic assembly machine. Accordingly, many benefits 
have been credited to the DFA methodology, including simplification of prod- 
ucts, lower assembly costs, reduced assembly (and manufacture) time, and re- 
duced overheads. Recently, the DFA concept has been extended to incorporate 
process capacity and product mix considerations so that products can be de- 
signed to assist in balancing assembly flow, thus eliminating the problem of 
stressing one process too heavily while underutilizing others. Many people now 
are calling for extending DFA over the whole product life cycle, in which case 
environmental concerns would be addressed and designs would be developed 
that facilitate disassembly for service as well as for recycling at the end of the 
life cycle. 

A first step toward a rational design for easy and low-cost assembly is the 
selection of the most appropriate method for assembling the product under 
consideration. The design guidelines for the selected method can then be ap- 
plied to an assembly-conscious design for that product. The next step is the 
use of a quantitative measure to evaluate the design in terms of the ease of 
assembly and to pinpoint the sources of problems so that the design can then 
be subjected to improvement. As many iterations of this evaluation/improve- 
ment process as are necessary can be done in order to achieve an optimal de- 
sign. It is, therefore, essential for us now to discuss the different assembly 
methods currently available. 


As you may expect, there is no single method that is always "better" than other meth- 
ods under all conditions. In other words, each method has its own domain or range 
within which it can most successfully and economically be applied. Factors like the 

12.1 Types and Characteristics of Assembly Methods 433 

number of products assembled per year and the number of individual components in 
an assembly play a major role in determining the range of economical performance of 
an assembly system. Following is a description of the different assembly methods, as 
well as the characteristics of each. 

Manual Assembly 

In manual assembly, the operations are carried out manually with or without the aid of 
simple, general-purpose tools like screwdrivers and pliers. Individual components are 
transferred to the workbench either manually or by employing mechanical equipment 
such as parts feeds or transfer lines and then are manually assembled. This assembly 
method is characterized by its flexibility and adaptability — a direct consequence of the 
very nature of the key element of the system, the human brain. The assembly cost per 
product, however, is virtually constant and is independent of the production volume. 
There is an upper limit to the production volume above which the practicality and fea- 
sibility of the manual assembly method is, to say the least, questionable. This upper 
limit depends upon the number of individual components in an assembly and the num- 
ber of different products assembled. Nevertheless, it is important to remember that the 
capital investment required for this type of assembly system is close to zero. 

Automatic Assembly Using 
Special-Purpose Machines 

In the type of assembly system referred to as fixed automation or the Detroit type, ei- 
ther synchronous indexing machines and automatic feeders or nonsynchronous ma- 
chines where parts are handled by a free-transfer device are used. The system, in both 
cases, should be built to assemble only one specific product. Such is the case with the 
automotive assembly lines in Detroit, where each one is dedicated to the production of 
a specific model of car (and hence the reason this name is given to this type of assem- 
bly system). There is an inherent rigidity in this method of assembly, meaning that 
these systems lack any flexibility to accommodate tangible changes in the design of the 
product. Moreover, a system of this type requires a large-scale capital investment, as 
well as considerable time and engineering work before actual production can be 
started. Also, the individual components must be subjected to strict quality-control in- 
spection before they can be assembled because any downtime due to defective parts 
will result in considerable production and, therefore, cash losses. Nevertheless, a real 
advantage of this assembly system is the decreasing assembly cost per product for in- 
creasing production volume. Naturally, when the production volume increases, the 
share of each product from the capital investment becomes smaller, which makes this 
assembly system particularly appropriate for mass production. It is worth mentioning 
that an underutilized assembly system will simply result in an increase in the assem- 
bly cost per product because the cost of equipment has to be divided between a smaller 
number of products, thus increasing the cost share of each product. 

In order to come up with a more flexible version of the automatic assembly sys- 
tem that can tolerate some minor changes in the design of the product being assem- 
bled, the nonsynchronous machines are fitted with programmable workheads and parts 

434 12 Design for Assembly 

magazines. Thus, the assembly sequence and characteristics can be tailored to match 
the attributes of the modified design. Although this system provides some flexibility, it 
is still considered to be most appropriate for mass production. 

Automatic Assembly Using Robots 

In robotic assembly, the production volume is higher than that of a manual assembly 
system but lower than that of an automatic assembly system that incorporates special- 
purpose machines. It, therefore, fills a gap, in production volume, between these other 
two assembly systems. Robotic assembly systems may take one or more of the fol- 
lowing forms: 

1. A one-arm, general-purpose robot operating at a single workstation that includes 
parts feeders, magazines, and so on. The end effector of the arm is tailored to suit 
the specific operation performed. 

2. Two robotic arms operating at a single workstation. A programmable controller 
(PLC) is employed to simultaneously control and synchronize the motions of the 
two arms. This setup is referred to as a robotic assembly cell and is, in fact, very 
similar to a flexible manufacturing cell. Other supporting equipment like fixtures 
and feeders are also included in the cell. 

3. Multistation robotic assembly system. This system is capable of performing several 
assembly operations simultaneously. It can also perform different assembly opera- 
tions at each station. Accordingly, this robotic assembly system possesses ex- 
tremely high flexibility and adaptability to design changes. On the other hand, a 
production volume that is quite close to that of the automatic assembly mass pro- 
duction system can be achieved using this type of system. 

Comparison of Assembly Methods 

Clearly, manual assembly requires the least capital investment followed by the two 
simplest forms of robotic assembly. On the other hand, compared to the automatic sys- 
tem with special-purpose machines, the multistation robotic assembly system requires 
more capital investment for a large production volume but less capital investment for 
a moderate production volume. A better way of illustrating this comparison is to plot a 
graph indicating the relationship between the assembly cost per product and the annual 
production volume for the three assembly methods. As shown in Figure 12.1, the as- 
sembly cost per product is constant for manual assembly and decreases linearly with 
increasing production volume for automatic assembly using special-purpose machines. 
In the case of robotic assembly, the assembly cost per product also decreases with in- 
creasing production volume but not linearly because the type of system used and its 
physical size depend upon the production volume as well. Figure 12.1 also helps to de- 
termine the range of production volume within which each of the assembly methods is 
cost effective. Consequently, such a graph is a valuable tool for selecting the appro- 
priate assembly method for a specific project. 

12.2 Selection of Assembly Method 


FIGURE 12.1 

Assembly cost per 
product versus annual 
production volume for 
three assembly 



Automatic assembly 

using special- 
purpose machines 

Annual production volume 


Several factors must be taken into consideration by the product designer and the man- 
ufacturer when selecting an assembly method. These factors include the cost of as- 
sembly, the annual production volume (or production rate), the number of individual 
components to be assembled in a product, the number of different versions of a prod- 
uct or products, the availability of labor at a reasonable cost, and, last but not least, the 
payback period. The factors are interactive, and it is impossible to have a single math- 
ematical relationship or a single graph that incorporates them all and indicates an ap- 
propriate range or domain for each assembly method. Usually, a two-variable chart is 
constructed based on fixed specific values for the other variables. 

Figure 12.2 indicates the appropriate ranges of application for each of the var- 
ious assembly methods when there is only one type (or version) of the product to 
be assembled. As can be seen, the two variables, which are pivotal in most cases, 
are the annual production volume and the number of individual components in an 
assembly. Notice that the manual assembly meth od is suit able— for low production 
volumes and a limited number of individual components per assembly. Robotic as- 
sembly is recommended for moderate production, with the one-arm robot being 
more appropriate for assemblies that have less than eight individual components. 
When a large number of assemblies is to be produced, the use of assembly systems 
with special-purpose machines becomes a must. Remember that with an increasing 


12 Design for Assembly 

FIGURE 12.2 

Appropriate ranges of 
application for various 
assembly methods 

to o 

=1 T3 

? E 

2 "> 

Q. 0) 

E « 


40 - 

30 - 

20 - 



1000 2000 3000 

Annual production volume 



number of different types or versions of assemblies, the recommended ranges of ap- 
plication for each assembly method will differ from those shown in Figure 12.2. 
For instance, a multistation robotic assembly system would be more appropriate 
than an automatic assembly system with special-purpose machines for relatively 
high production volumes. The most important point here is that the assembly rate 
of the selected assembly method should not result in any bottleneck but rather 
should ensure trouble-free production. Also, it is always advisable to estimate the 
cost of assembly whenever more than one assembly method is under consideration. 
Assuming that all other factors are comparable, the method that gives the lowest as- 
sembly cost is the one to select. 



We are now in a position to discuss the rules and guidelines to be followed when de- 
signing components for manual assembly. It is important here to emphasize that blind 
adherence to these rules is not recommended. In fact, this approach can result in very 
complex components that are difficult and expensive to manufacture. The use of good 
engineering sense, rational thinking, and accumulated knowledge will ensure that these 
rules are wisely applied. The strategy to adopt when designing products for manual as- 

12.3 Product Design for Manual Assembly 


sembly is to strive to reduce both the assembly time and the skills required of assem- 
bly workers. Here are the guidelines for product design for manual assembly: 

1. Eliminate the need for any decision making by the assembly worker, including his 
or her having to make any final adjustments. Remember that assembly workers are 
usually unskilled and are paid at or close to the minimum wage and it is, therefore, 
not logical or fair to rely on them to make these adjustments. 

2. Ensure accessibility and visibility. It is not logical or fair to require the worker, 
for example, to insert and tighten a bolt in a hole that is not visible or easily 

3. Eliminate the need for assembly tools or special gages by designing the individual 
components to be self-aligning and self-locating. Parts that fit and snap together 
eliminate the need for fasteners, thus resulting in an appreciable reduction in both 
the assembly time and cost. Also, features like lips and chamfers can greatly aid in 
making parts self-locating, as is clearly demonstrated in Figure 12.3, where two 
pins, one having a chamfer and the other without, are being inserted into two iden- 
tical holes during an assembly operation. Obviously, it is far easier and takes less 
time to insert the pin with the chamfer. 

4. Minimize the types of parts by adopting the concept of standardization as a design 
philosophy. Expand the use of standard parts as well as multifunction and multi- 
purpose components. Although more material may be consumed to manufacture 
multipurpose parts, the gains in reducing assembly time and cost will exceed that 

5. Minimize the number of individual parts in an assembly by eliminating excess parts 
and, whenever possible, integrating two or more parts together. Certainly, handling 
one part is far easier than handling two or more. The criteria for reducing the parts 
count per assembly, established by G. Boothroyd and P. Dewhurst (see the refer- 
ences at the end of this book), involve negative answers to the following questions: 

FIGURE 12.3 

Using a chamfer to 
make a part self- 

438 12 Design for Assembly 

• Does the part move relative to all other parts already assembled? 

• Must the part be of a different material or be isolated from other parts already 
assembled? (Only fundamental reasons concerned with material properties are 

• Must the part be separate from all other parts already assembled because other- 
wise necessary assembly or disassembly of other parts would be impossible? 

If the answer to each of these questions is no, then the part can be integrated or 
combined with another neighboring part in the assembly. When applying this rule, 
however, remember that combining two or more parts into a complicated one may 
result in making the part difficult to manufacture. 

6. Avoid or minimize reorienting the parts during assembly. Try to make all motions 
simple by, for example, eliminating multimotion insertions. Avoid rotating or reori- 
enting the assembly as well as releasing and regripping individual components. 
These are wasteful motions and result in increased assembly time and cost. The 
best time to eliminate them is during the design phase. The use of vertical insertion 
(along the Z axis) is ideal, especially when you take advantage of gravity. 

7. Ensure ease of handling of parts from the bulk by eliminating the possibility of 
nesting or tangling them. This is achieved by simple modifications in the design. In 
addition, avoid the use of fragile or brittle materials, as well as flexible parts like 
cords and cables. 

8. Design parts having maximum symmetry in order to facilitate easy orientation and 
handling during assembly. If symmetry is not achievable, the alternative is to de- 
sign for asymmetry that is easily recognizable by the assembly worker. 

Failure to observe the preceding rules may result in serious problems during as- 
sembly in terms of higher assembly costs or jams and delays. Consequently, many 
companies avoid manual assembly and sell their products unassembled. Examples in 
the United States include grills, furniture, and toys. As you may have experienced, 
some of these products are not properly designed for easy assembly, and it takes cus- 
tomers an extremely long time to assemble them. It is no surprise that such faulty de- 
signs do not pay off as they adversely affect the sales of the unassembled products. 



Parts that are designed to be assembled by automatic special-purpose machines must 
possess different geometric characteristics from those of parts to be assembled manu- 
ally. Automatic assembly requires parts that are uniform, are of high quality, and have 
tighter geometric tolerances than those of manually assembled parts. These requirements 
are dictated by the need to eliminate any downtime of the assembly system due to parts 
mismatch or manufacturing defects. As a consequence, problems related to locating and 
inserting parts, though they need to be addressed, are not of primary importance. These 
problems require design changes to ease assembly; by revising the product design, each 

12.4 Product Design for Automatic Assembly 


FIGURE 12.4 

Facilitating assembly 
through reduction in 
parts count (Redrawn 
after Iredale, R. 
"Automatic Assembly — 
Components and 
Products, " Metal- 
working Production, 8 
April 1964. Used by 

Old (13 parts) 

New (2 parts) 

assembly operation becomes simple enough to be performed by a machine rather than 
by a human being. The most important concerns to address involve the orientation, 
handling, and feeding of parts to the assembly machine. The efficiency of performing 
these tasks has a considerable effect on the efficiency and output of the assembly sys- 
tem and, of course, on the assembly cost. This approach is referred to as design for 
ease of automation. Here are the guidelines for product design for automatic assembly: 

1. Reduce the number of different components in an assembly by using the three 
questions listed previously in the design guidelines for manual assembly. An ap- 
propriate approach is to use value analysis in identifying the required functions 
performed by each part and finding out the simplest and easiest way to achieve 
those functions. An example is shown in Figure 12.4, where two products are con- 
trasted, one designed to facilitate assembly through a reduction in the parts count 
and the other designed without ease of assembly being taken into consideration. 

With the new developments in casting and plastics injection-molding tech- 
nologies, complex components can replace entire subassemblies. Nevertheless, the 
designer has to be very careful when combining parts so as not to adversely affect 
the manufacturing cost. In fact, in order to reduce the parts count in assemblies, 
subcontractors and suppliers of electronics manufacturers have been continually 
asked to fabricate extremely complex parts. In short, the rule of reducing the num- 
ber of parts should not be applied blindly because, in many cases, more efficient 
manufacturing can be achieved by breaking a single component into two or more 
parts, as shown in Figure 12.5, which indicates two methods for manufacturing a 
2-foot axle shaft and flange. 

FIGURE 12.5 

Two methods for manufacturing an axle shaft and flange (Redrawn after Lane, J. D., ed. 
"Automated Assembly," 2nd ed.. Society of Manufacturing Engineers, 1986. Used by permission) 


Old (fewer parts) 




New (many parts) 


12 Design for Assembly 

FIGURE 12.6 

Facilitating assembly 
through simplification 
of design (Redrawn 
after Iredale, 1964. 
Used by permission) 

Plate (steel, 
2 required) \ 

Fan (nylon) 


(2 required) 

Plate (steel, 
2 required) 

Bearing (nylor^ 
2 required) 



Shaft (steel) 



Old (8 parts) 

New (3 parts) 

One -piece fan 
and shaft 


Use self-aligning and self-locating features in parts to facilitate the process of their 
assembly. Considerable improvement can be achieved by using chamfers, guide- 
pins, dimples, molded-in locators, and certain types of screws (e.g., cone and oval 
screws). Figure 12.6 is an example of how to facilitate assembly through simple 
design modifications, while Figure 12.7 shows the types of screws that are suitable 
for assembly operations. 

Avoid, whenever possible, fastening by screws because that process is both ex- 
pensive and time-consuming. It is, therefore, recommended to design parts that 
will snap together or be joined together by a press fit. Tighter tolerances are then 
required, and problems may also be encountered in disassembly for maintenance, 

FIGURE 12.7 

Types of screws 
suitable for assembly 
operations (Redrawn 
after Tipping, W. V. 
"Component and 
Product Design for 
Mechanized Assembly," 
Conference on 
Assembly, Fastening, 
and Joining Techniques 
and Equipment, 
Production Engineering 
Research Association 
of Great Britian, 1965. 
Used by permission) 




thread point 




(after thread 

form as 

rolling, the 


point approximates 
to a chamfer) 





12.4 Product Design for Automatic Assembly 441 

repair, or recycling. If screws must be used, then unify their types and head 

4. Make use of the largest and most rigid part of the assembly as a base or fixture 
where other parts are stack-assembled vertically in order to take advantage of 
gravity. This will eliminate the need for employing an assembly fixture, thus sav- 
ing time and cost. Also, remember that the best assembly operation is one that is 
performed in a sandwichlike or layered fashion. If this is difficult or impossible to 
do, the alternative is to divide the assembly into a number of smaller subassem- 
blies, apply the rule stated herein to each separately, and then plug all the sub- 
assemblies together. 

5. Actively seek the use of standard components and/or materials. There should be a 
commitment, at all levels, to the goal of using a high percentage of standard parts 
in any new design. A very useful concept to be adopted in order to achieve this 
goal is group technology. Standardization should begin with fasteners, washers, 
springs, and other individual components. This translates into standardization of 
assembly motions and procedures. The next step is to use standard modules that 
are assembled separately and then plugged together as a final product. Each mod- 
ule can include a number of individual components that are self-contained in a 
subassembly having a specific performance in response to one or more inputs. 
This approach can lead to a considerable reduction in assembly cost, as well as in 
manufacturing and inventory costs. 

6. Avoid the possibility of parts tangling, nesting, or shingling during feeding. A few 
changes in the geometric features may eliminate these problems without affecting 
the proper functioning of the component. Figure 12.8 shows some parts that tend 
to nest during feeding and the design modifications that eliminate this problem. 

7. Avoid flexible, fragile, and abrasive parts and ensure that the parts have sufficient 
strength and rigidity to withstand the forces exerted on them during feeding and 

8. Avoid reorienting assemblies because each reorientation may require a separate 
station or a machine, both of which cause an appreciable increase in cost. 

9. Design parts to ease automation by presenting or admitting the parts to the as- 
sembly machine in the right orientation after the minimum possible time in the 
feeder. The process in the feeder consists of rejecting parts resting in any position 
but the one desired. Consequently, reducing the number of possible orientations of 
a part actually increases the odds of that part's going out of the feeder on its first 
try. Figure 12.9 shows the effect of the possible number of orientations on the ef- 
ficiency of feeding. According to W. V. Tipping, two types of parts can easily be 
oriented: parts that are symmetrical in shape (e.g., a sphere or cube) and parts with 
clear asymmetry (preferably with marked polar properties either in shape or 

Symmetrical parts are easily oriented and handled. Therefore, try to make parts 
symmetrical by adding nonfunctional design features like a hole or a projection. 


12 Design for Assembly 

FIGURE 12.8 

Parts that tend to nest 
during feeding and 
design modifications 
that eliminate the 
problem (Redrawn after 
Lane 1986. Used by 

Open-ended spring 
that will tangle 

Closed-ended spring 

that will tangle only 

under pressure 

"Nesting" of 

Increase wire size 
or decrease pitch 

Open up pitch to avoid 
locking angles 

Increase angle 


Add flames or ribs 

Decrease angle 

12.4 Product Design for Automatic Assembly 


FIGURE 12.9 

Effect of possible 
number of orientations 
on efficiency of feeding 
(Redrawn after Lane, 
1986. Used by 

Number of 

Types of Parts 




flat washer 

Tapered washer 

Parts that naturally fall in one of two 
possible positions 

Parts having four natural positions 


Number of 


(out of the feeder) 





Required Rate 

of Feeding 

(into the feeder) 




Figure 12.10 shows some small changes in the design of parts that result in full sym- 
metry. Generally, it is easy to achieve symmetry with sheet metal and injection- 
molded parts because the manufacturing cost of adding a feature is relatively low. 

If it is too difficult or too expensive to achieve symmetry, nonfunctional fea- 
tures must then be added to make identification and grasping easier. This approach 
is also employed for parts for which orientation is based on hard-to-detect features 
like internal holes. In addition, components having similar shape and dimensions 
are difficult to identify and orient, and changes in dimensions or additions of de- 
sign features must be made. Recent research work has come up with a concept, 
called feedability, that involves quantitative estimation of the odds of feeding a 
part having certain geometric characteristics to the assembly station in a specific 
orientation. Figure 12.11 shows some design changes that exaggerate asymmetry 
or indicate hidden features, while Figure 12.12 shows the effect of changing geo- 
metric features on the calculated values of feedability. 

10. Try to design parts with a low center of gravity (i.e., it should not be far above the 
base). This gives the part a natural tendency to be fed in one particular orientation. 

FIGURE 12.10 

Examples of design 
changes that give full 
symmetry (Redrawn 
after Lane, 1986. Used 
by permission) 

C= 1 

Difficult to orient 

Easy to orient 




Usual design of dowel pin 

Redesigned dowel pins 

Before (2 natural orientation) 

After (1 orientation required) 

FIGURE 12.11 

Examples of design 
changes that 
exaggerate asymmetry 
or indicate hidden 
features (Redrawn after 
Iredale, 1964. Used by 

Difficult to orient with 
respect to small holes 

Flats on the sides make it much 
easier to orient with respect to 
the small holes 

No feature sufficiently significant 
for orientation 

When correctly oriented will 
hang from rail 

Triangular shape of part makes 
automatic hole orientation 

Nonfunctional shoulder permits 
proper orientation to be 
established in a vibratory feeder 
and maintained in transport rails 


12.5 Product Design for Robotic Assembly 


FIGURE 12.12 

Effect of changing 
geometric features on 
calculated values of 
feedability (Redrawn 
after Kim et al., "A 
Shape Metric For 
Design-for-Assembly, " 
Proceedings of the 
Conference on Robotics 
and Automation, 1992. 
Used by permission) 


o o 

f = 0.232 f = 1 

Symmetric part has a higher feedability 

f = 0.2 f = 0.25 

Asymmetric part has higher feedability than part with ambiguous symmetry 


f = 0.25 f = 0.5 

Gripping feature can increase the feedability 


/\ o 

f = 0.25 f = 0.30 f = 0.47 f = 0.38 

Avoiding toppling does not necessarily increase feedability 

Also, when such a part is transferred on a conveyor belt, it will not tip or be mis- 
oriented due to the force of inertia. 



The product design rules for robotic assembly are basically the same as those for man- 
ual and /or automatic assembly. There are, however, two very important and crucial 
considerations that have to be taken into account when designing components for ro- 
botic assembly. They can be summed up as follows: 

1. Design a component so that it can be grasped, oriented, and inserted by that robot's 
end effector. Failure to do so will result in the need for an additional robot and, con- 
sequently, higher assembly cost. 

2. Design parts so that they can be presented to the robot's arm in an orientation ap- 
propriate for grasping. Also, eliminate the need for reorienting assemblies (or sub- 
assemblies) during the assembly operation. Ignoring this rule will cause an increase 

446 12 Design for Assembly 

in assembly time by consuming the robot's time for no valid reason. It also will 
cause an increase in the assembly cost per unit. 



At this point, let us review some of the methods currently used in industry, in America 
and abroad, for evaluating and improving product DFA. Because so many methods, 
systems, and software packages have recently been developed, the survey here will be 
limited to the most commonly known and used methods, for which substantial infor- 
mation and details have been published. There is no bias here for or against any 
method that has or has not been covered. 

As you will soon see, most of the methods are based on measuring the ease or dif- 
ficulty with which parts can be handled and assembled together into a given product. 
This does not mean that the components are physically brought together but rather that 
an analytical procedure is followed where the problems associated with the compo- 
nents' design are detected and quantitatively assessed. The right answer or optimal de- 
sign comes from you, the engineer, when you use a particular DFA method as a tool in 
evaluating and comparing alternative design solutions. Following is a survey of each 

The Booth royd-Dew hurst DFA Method 

The Boothroyd-Dewhurst DFA method was developed in the late 1970s by Professor 
Geoffrey Boothroyd, a pioneer in the area of DFA, at the University of Massachusetts 
Amherst in cooperation with Salford University of England. First, the appropriate as- 
sembly method is selected by means of charts. Then, the analytical procedure corre- 
sponding to the assembly method selected is used (i.e., there is a separate, though 
similar, procedure for each of the assembly methods). Figure 12.13 is a diagram of the 
stages of the Boothroyd-Dewhurst DFA method. 

As an example, let us now examine the analytical procedure for manual assembly 
as the DFA analysis procedures for the other assembly methods are not much different. 
Note that the analysis cannot be employed to create a design from nothing but rather 
is used to evaluate and refine an existing design. In other words, the starting point is 
an assembly drawing of the product (either a prototype or an actual product). The first 
step in the analysis is to determine the assembly sequence (i.e., the part that is to be 
placed first and the parts that are to follow it in the order to be used for attaching them 
together). Boothroyd and Dewhurst proposed the worksheet shown in Figure 12.14 for 
effective bookkeeping of the assembly time and cost. 

When more than one part is to be used in an operation, the assembly time for that 
operation is obtained by multiplying the assembly time for one part by the number of 
parts (see Figure 12.14). Required but nonassembly operations must also be included 
in the sequence. Each time the unfinished assembly is reoriented during the assembly 
process, the reorientation operation is entered into the worksheet and a time is alio- 

12.6 Methods for Evaluating and Improving Product DFA 


FIGURE 12.13 

Stages of the 
DFA method (Redrawn 
after Miles, B.L. 
"Design for Assembly — 
A Key Element within 
Design for 
Manufacture, " 
Proceedings of the 
Institution of 
Mechanical Engineers, 
1989. Used by 

Select the 






for manual 


for high-speed 



for robotic 




Improve the 
design and 

cated for it. The assembly time for each component part is then obtained by adding the 
handling time of that part to its insertion time. These two times are extracted from 
charts that include assembly data. The data were compiled by Boothroyd, Dewhurst, 
and their coworkers based on practical observation over long periods of time and on 
research. In order to use the handling-time chart, a two-digit handling code must first 
be determined for each part based on its size, weight, and geometric attributes. A two- 
digit insertion code (and thus time) must also be obtained for each part based on ac- 
cessibility, vision restriction, and resistance to insertion. Once the components and the 
assembly time for each are known, it is easy to estimate the total assembly time and 
assembly cost for the existing design. 

The next step is aimed at reducing the parts count by totally eliminating some 
parts or combining them with neighboring parts. This is achieved by answering the 
previously listed three questions about the movement of the part relative to adjacent 
parts, its materials, and the need to have it separate for assembly and/or disassembly. 
Candidates for elimination can be identified and subtracted from the total number of 
parts to obtain the number of "theoretically needed" parts. Assuming that an ideal as- 
sembly operation of a component takes 3 seconds (1.5 seconds for handling and 1.5 
seconds for inserting), the total ideal assembly time is given by the following equation: 

total ideal assembly time = 3N^ 


where N M is the theoretical minimum number of parts. 

Boothroyd and Dewhurst used a design efficiency index to evaluate the improve- 
ment in design in a quantitative manner. This index can be given by the following 

design efficiency = 


calculated total assembly time 



12 Design for Assembly 

FIGURE 12.14 

The Boothroyd- 
Dewhurst bookkeeping 










Name of 


© l. 


tr — 


J= "° =». 

(o "E S> 
© ca -2 
E " B 

r -^ o 

2 § c 

j3 ffl O 

E © a 
3 Q- O 

z ° 


1 ©l 
C "O o 

7? © 

O (0 c 

I- s 

£ r 

X3 © 
C Q. 

n >- 

.C 0) 
_ Q. 

a © 

c E 

© 5= 

1 ©1 

(0 o " 
E ° E 

h=; t: T3 

V © © 

O <" c 

g = © 

H S 

.2 c 

■C « 
© Q. 

<" ,_ 
= © 
_ Q. 

a © 

c E 
© a 



c + 
° * 

© x 








3 (0 

5 E S 

o „.£ a 

*= Or ^ 

© .£ c o 


oi S: c 

© c 














T M 

C M 

N M 

In column 9, if part is not essential, put 0; if required, put 1 . 

12.6 Methods for Evaluating and Improving Product DFA 


The mechanism for improving the design, according to this method, involves a review 
of the worksheet in order to pinpoint components that can be eliminated and that have 
relatively high handling and insertion times. The number of components or parts must 
then be reduced by eliminating some or most of the components so identified. This 
process is repeated until an optimal design (i.e., one having a design efficiency much 
higher than that of the initial design) is obtained. 

Because it is rather time-consuming to perform the Boothroyd-Dewhurst proce- 
dure manually, a software package for DFA analysis based on their structured analysis 
has been developed. The latest commercially available version is very user friendly 
and runs in a Windows environment. Again, note that the system does not make any 
decisions for the designer; it is the designer who, with rational thinking and good en- 
gineering sense, ultimately decides what is right and appropriate. 

One final note here: Although this DFA analysis would certainly decrease the parts 
count, it can often result in the manufacture and use of complex components. Bearing 
in mind that the assembly cost is only about 5 percent of the total production cost, the 
finalized "optimal design" may be easy to assemble but expensive (or difficult) to man- 
ufacture. In fact, the absence of a manufacturing-knowledge-based supporting system 
was the main shortcoming of the initial DFA techniques. Realizing that fact, 
Boothroyd and Dewhurst supplemented their DFA software with what they called de- 
sign for manufacture software. This software is actually a product cost estimator for a 
few selected manufacturing processes and is used to estimate the manufacturing cost 
of the different alternative designs. The optimal design can then be selected based on 
both the assembly and the manufacturing costs. 

The Hitachi Assembly Evaluation Method 

Another method with a proven record of success is the Hitachi assembly evaluation 
method (AEM). It was employed to refine the designs of tape recorder mechanisms in 
order to develop an automatic assembly system for producing those subassemblies. 
That pioneering and original work by S. Hashizure (a research engineer at Hitachi) and 
his coworkers was awarded the Okochi Memorial Prize in 1980. Although this method 
does not explicitly distinguish between manual and automatic assembly, this difference 
is accounted for implicitly within the structured analysis. Also, the method was sub- 
jected to refinement in 1986 with improvements to its methodology, and a computer- 
based version is now available. 

The Hitachi AEM approach is based on assessing the assemblability of a design 
by virtue of the following two indices: 

1. An assemblability evaluation score (E) is used to assess design quality or difficulty 
of assembly operations. The procedure to compute £ is based on considering the 
simple downward motion for inserting a part as the "ideal reference." For more 
complicated operations, penalty scores that depend upon the complexity and nature 
of each operation are assigned. The Hitachi method uses symbols to represent op- 
erations, and there are about 20 of them covering operations like the straight down- 
ward movement for part insertion and the operation of soldering, as shown in 
Figure 12.15. 


12 Design for Assembly 

FIGURE 12.15 

Examples of Hitachi 
method symbols and 
penalty points (Redrawn 
after Miyakawa, S., and 
T. Ohashi, "The Hitachi 
Evaluation Method 
(AEM), " Proceedings of 
the international 
Conference on Product 
Design for Assembly, 
April 1986. Courtesy of 
Institute for 
Competitive Design) 

Elemental operation 

AEM symbol 

Penalty score 

r 1 h 








After completing a worksheet in the same order as the anticipated assembly se- 
quence, the penalty score for each part is manipulated to give the assemblability 
evaluation score for that part. The E values for all parts are then combined to pro- 
duce an assemblability evaluation score for the whole assembly. Because a penalty 
score of zero corresponds to an E value of 100 percent, the higher the E score for 
an assembly, the lower the assembly time and cost. Accordingly, if each part of an 
assembly is to be added by a simple downward motion, the E score for each part 
and, therefore, for the whole assembly will be 100 percent. The E score is employed 
in simplifying the various operations and not explicitly in reducing the parts count. 

2. An estimated assembly cost ratio (K) is an indication of the assembly cost im- 
provements. As the name suggests, K is the ratio between the assembly cost of the 
new (modified) design divided by the assembly cost of the initial and /or standard 
design. It is clear that when K is 0.7, there is a 30-percent saving in the assembly 
cost as a result of modifying the design. The method of estimating the time (and 
cost) of an operation involves breaking it into its elemental components and allo- 
cating time for each elemental motion based on compiled practical observations. 
Any saving in the assembly cost can be achieved by reducing the parts count in a 
product and/or simplifying the assembly operations. 

The Lucas DFA Method 

The Lucas DFA method was developed in the 1980s as a result of collaborative work 
between the Lucas Corporation and the University of Hull (both in England). The mo- 
tivation for developing this method, as stated by its creators, B. L. Miles and K. G. 
Swift, was to have the best features of the commercially available DFA software pack- 
ages within a simple system and to aim its application at an early stage of the design 
process. Unlike the previous two methods, the Lucas DFA evaluation is not based on 
monetary costs, but on three indices that give a relative measure of assembling diffi- 
culty. The goal of reducing the parts count and the analysis of the insertion operations 
based on an encoded classification system, however, are shared with the previous two 

12.6 Methods for Evaluating and Improving Product DFA 


methods. Also, an easy-to-use computer version of this method is now commercially 

Figure 12. 16 shows an assembly sequence flowchart (ASF) of the Lucas DFA pro- 
cedure. As can be seen, the analysis is carried out in three sequential stages: the func- 
tional, feeding (or handling), and fitting analyses. It can also be seen that the existence 
of a well-defined product design specification (PDS) is a must for carrying out the first 
stage of the DFA analysis. 

Functional analysis. In the functional analysis, components are divided into two 
main groups. The first group includes components that perform a primary function 
and, therefore, exist for fundamental reasons. These components are considered to 
be essential, or "A," parts. The second group involves nonessential, or "B," compo- 
nents that perform only secondary functions like fastening and locating. The design 

FIGURE 12.16 

The Lucas DFA 
assembly sequence 
flowchart (Redrawn 
after Miles, 1989. 
Used by permission) 

Product specification 


Product design 



Functional analysis 

< J 




Feeding analysis 



Fitting analysis 












452 12 Design for Assembly 

efficiency is the product of dividing the number of essential parts by the total num- 
ber of parts and can be given by the following equation: 

design efficiency = x 100 (12.3) 

According to the flowchart (see Figure 12.16), if the design efficiency is low, it should 
be improved through design modifications aimed at eliminating most of the nonessen- 
tial parts. A clear advantage of the Lucas DFA method is that performing the functional 
analysis separately, before the other two analyses, acts as an initial "screening mecha- 
nism" that returns back poor designs before further effort is encountered in the detailed 
analysis. For this initial stage, the target objective is to achieve a design efficiency of 
60 percent. 

Feeding analysis. The feeding analysis is concerned with the problems associated 
with handling components (and subassemblies) until they are admitted to the assembly 
system. By answering a group of questions about the size, weight, handling difficul- 
ties, and orientation of a part, its feeding /handling index can be calculated. Next, the 
feeding/handling ratio can be calculated by using the following equation: 

„ ,. ,, ,,. . feeding /handling index ,,. ., 

feeding/handling ratio = — (12.4) 

number or essential components 

An ideal value for this ratio and one that is often taken as a target goal is 2.5. 

Fitting analysis. The fitting analysis is divided into a number of subsystems includ- 
ing gripping, insertion, and fixing analyses. An index is given to each part based on its 
fixturing requirements, resistance to insertion, and whether or not there will be re- 
stricted vision during assembly. High individual values and/or a high total value of 
these indices means costly fitting operations, in which case the product should be re- 
designed with the goal of eliminating or at least reducing these operations. The fitting 
index is manipulated to yield the fitting ratio as given by the following equation: 

fitting index ,,-_, 

fitting ratio = - (12.5) 

number of essential components 

Again, for the design to be acceptable, the value of the fitting ratio should be 
around 2.5. 

Note that while the feeding /handling and fitting ratios can certainly be used as 
"measures of performance" to indicate the effectiveness of the design quality with re- 
spect to assembly, the absence of a mechanism to evaluate the effect of design changes 
on the manufacturing cost is a clear shortcoming of this method. 

The Fujitsu Productivity Evaluation System 

Some technical information about the Fujitsu productivity evaluation system (PES) 
was published in the Fujitsu Scientific Technical Journal in August of 1993. Unlike 
other DFA techniques, this method was developed not as a refinement procedure after 
the completion of the detailed design, but rather as a tool to aid in obtaining a detailed 

12.6 Methods for Evaluating and Improving Product DFA 


FIGURE 12.17 

The Fujitsu PES 
(Redrawn after 
Miyazawa, A. 
"Productivity Evaluation 
System, " Fujitsu 
Science Technology 
Journal, December 
1993. Used by 

Assembly sequence 
specification subsystem 

evaluation subsystem 

evaluation subsystem 

Design ideas and know- 
how reference subsystem 

design that is easy to manufacture and assemble and also is cost effective. This method 
is, however, limited to bench-type manual assembly of relatively small parts, exclud- 
ing, for example, products like automobiles and refrigerators. As can be seen in Fig- 
ure 12.17, the Fujitsu PES (which is actually a software package) consists of four 
subsystems. It is based upon making full use of an expert system involving practical 
manufacturing and design data and rules of thumb that are gathered from the finest in- 
dustry experts. The software addresses a problem by carrying out a rough evaluation 
that can be followed by detailed evaluations made concurrently with the product de- 
velopment process. The system is capable of performing absolute evaluation of as- 
sembly time and cost, as well as comparative evaluation as a percentage of that of a 
reference design. Figure 12.18 indicates the procedure for applying the productivity 
evaluation system throughout the product development cycle. Let us now discuss the 
function of each of the subsystems. 

Assembly sequence specification subsystem. The function and the operation of the 
assembly sequence specification subsystem are shown in Figure 12.19. The designer 
selects parts similar to those envisioned to be used in the product, according to the 
conceptual design, and forms a library of parts and then specifies their assembly se- 
quence. The system promptly retrieves previously stored values for assemblability and 
manufacturability that can be used by the evaluation subsystem to obtain assembly 
time and cost. 

Assemblability evaluation subsystem. This tool is employed to estimate the assembly 
time and evaluate the ease of assembly. It is based upon a library of subassemblies (or 
"mechanisms") and their number of essential parts that are stored by functional mod- 
ule. The printing module, for example, includes dot printing (10 essential parts), ther- 
mal printing (8 essential parts), and laser printing (15 essential parts). As soon as the 
designer specifies the subassembly, a detailed drawing together with all the informa- 
tion appears on the screen of the monitor. The analysis addresses the handling and in- 
sertion of parts, specifies the target number of essential parts, and identifies high-cost 
processes and parts. Figure 12.20 shows the operation of the assemblability evaluation 
subsystem, as well as the types of input and output data. In fact, the system breaks 
down the assembly time of each part into handling time, insertion time, and so on, and 
displays it as a bar chart, as shown in Figure 12.21a. The system also shows the as- 
semblability evaluation score for the whole product as well as for assembly and ad- 
justment operations, as shown in Figure 12.21b. 





12.6 Methods for Evaluating and Improving Product DFA 


FIGURE 12.19 

The assembly 
sequence specification 
subsystem (Redrawn 
after Miyazawa, 1993. 
Used by permission) 

Parts name and assembly sequence 

Production tree 

Product t- — 

Part A 

Unit 1 





Unit 2 






FIGURE 12.20 

The assemblability 
evaluation subsystem 
(Redrawn after 
Miyazawa, 1993. Used 
by permission) 

Input data 


Assembly operation by parts 
Size Direction 

Connecting method 
Adjustment method 
Adjustment points 
Adjustment accuracy ... 
Jigs and Tools 

Output data 

Target number of parts 
Assembly time 
Adjustment time 
Assemblability evaluation 

Diagnostic Notes 

Manufacturability evaluation subsystem. As previously mentioned, the objective of 
reducing the parts count in a product can be achieved at the expense of the ease and 
cost of manufacturing some of the parts of that product. For this reason, the manufac- 
turability evaluation subsystem was developed. As can be seen in Figure 12.22, it is 
used as a tool by the designer to estimate the manufacturing cost and evaluate manu- 
facturability in a quantitative manner (on a score scale of 100). This can be done at two 
levels as desired by the designer: a rough evaluation and a detailed one. 

Note that although the Fujitsu PES seems to address the requirement essential for 
carrying out comprehensive design for manufacturability and design for assembly 
analyses, it is based on retrieving previously compiled data. Because these data are gath- 
ered during the production, the PES software may only be successfully applied to prod- 
ucts identical or similar to those falling within the range of products of that company. 
The system certainly has a proven record of success in the design of Fujitsu products, 
but its success when applied to other types of products has not yet been demonstrated. 
Figure 12.23 is an example of a product that was redesigned using the Fujitsu PES. 

Other Methods 

In addition to the DFA methods previously discussed, there are some software pack- 
ages that are commercially available and/or being applied in-housc in large corpora- 
tions. These include packages by AT&T, Sony, and Sapphire. Unfortunately, technical 
details are not yet available, so coverage of these methods was not possible in this text. 


12 Design for Assembly 

FIGURE 12.21 

evaluation results for a 
product: (a) assembly 
time for each part; (b) 
evaluation for the whole 
product (Redrawn after 
Miyazawa. 1993. Used 
by permission) 

Evaluation score (Max 100) 

Part name 

Assembly time (s) 



Part A 


Part B 




Unit 1 


Part E 


Unit 2 







■ H 

♦ „■ t U L \ f 





Total number of parts 



Target ! Secondary ! Fastening 

Total assembly time (h) 

1 2 

Assembly time j Adjustment time 

Fastener^.- — r+9€L^ Parts 

Fastener/ /JtB \j . \ j nc| 




Check Accuracy 



12.6 Methods for Evaluating and Improving Product DFA 


FIGURE 12.22 

The manufacturability 
evaluation subsystem 
(Redrawn after 
Miyazawa, 1993. Used 
by permission) 

Rough evaluation 

Input data 

Shape of part 


Detailed evaluation 

Input data 

Sheet metal 

Number of holes 





Number of holes 





Output data 

Parts evaluation 
Material cost 
Manufacturing type 
Treatment (plating, casting) 
Manufacturing cost 
Total evaluation 
Manufacturability evaluation 


Diagnostic notes 

FIGURE 12.23 

Example of a product 
redesigned using the 
Fujitsu PES: (a) before 
redesign (Redrawn after 
Miyazawa, 1993. Used 
by permission) 

Keyboard cover 




23 i 

r~\ _ . . Screw 

Lower cover 



FIGURE 12.23 

Example of a product 
redesigned using the 
Fujitsu PES: (b) after 
redesign; (c) product 
cost; (d) assemblability 
estimation score 
(Redrawn after 
Miyazawa, 1993. Used 
by permission) 

Connect board 

Lower cover 

s \ Connect board 




Assembly costs 
75 I 100 (%) 



Chapter 12 Design Project 


Review Questions 


1. What options are available when selecting an 13. 
assembly method? 14_ 

2. What are the major characteristics of each of 

the available assembly methods? 15_ 

3. Discuss some of the factors that affect the se- 
lection of an appropriate assembly method. 16 # 

4. Define the term design for assembly. 

5. What are the benefits of applying the concept of 17. 
design for assembly? 

6. What has always been the traditional approach 18. 
for DFA? 

7. What three questions form the criteria for elim- 19. 
inating a part from an assembly or combining it 20. 
with its neighboring part? 

8. List the guidelines for product design for man- 21. 
ual assembly. 

9. What is the ideal insertion motion? Why? 22. 

10. Why should you try to avoid reorienting parts 
during assembly? 23. 

11. What effect does the concept of standardization 
of parts have on the assembly process? 

12. Does nesting (or tangling) of parts while in the 
bulk have any effect on the assembly operation? 

Why should parts be symmetrical? 

What is your advice if you cannot get the parts 
to be symmetrical? 

List the guidelines for product design for auto- 
matic assembly. 

How can the use of self-aligning and self- 
locating features facilitate automatic assembly? 

Can screws be considered as essential parts in a 
product? Why? 

What is the ideal fixturing method in automatic 

Discuss the concept of feedability. 

Why should parts with a low center of gravity 
be favored in automatic assembly? 

List two rules for product design for robotic 

What are the methods for performing DFA 

List some of the advantages, characteristic fea- 
tures, and limitations of each of the DFA analy- 
sis methods. 

Design Project 


Choose a fairly simple product (e.g., a shower handle or coffeemaker), disassemble it, 
and make an assembly drawing or an exploded view of its parts. Next, study the func- 
tion and material of each part, as well as the assembly sequence. Then, use the three 
questions (elimination criteria) of Boothroyd and Dewhurst to identify parts that are 
candidates for elimination or combining with other parts. Finally, modify your design 
in order to reduce the parts count and provide an assembly drawing of the new design, 
as well as a workshop drawing for each part. 

Chapter 13 



esign and 



The increasing problems of landfill usage, the rising cost of energy and raw ma- 
terial, the greenhouse effect, and the decay of the ozone layer are among the 
major environmental concerns that prompted the second environmental revolu- 
tion that is now taking place in the United States. Unlike the first environmen- 
tal revolution in the 1970s, which was aimed at cleaning up hazardous waste 
from contaminated sites and natural resources, the second revolution is ad- 
dressing waste reduction at the source. This goal can be achieved through the 
design of products that promote recycling as well as through the design of man- 
ufacturing processes that minimize waste, by-products, and emissions, and, 
therefore, utilize resources more efficiently. The magnitude of our current envi- 
ronmental problems is vast indeed. Consider, for example, the alarming trend 
of solid-waste generation. According to the Environmental Protection Agency 
(EPA), Washington, D.C., the United States generates 160 million tons of solid 
waste every year, and most of it goes to landfills that are nearly full. Further- 
more, the EPA predicts that slightly less than half of the existing landfills will 
close before the turn of the century. The key to the solution of these problems 
lies in the policy of adopting environmentally friendly products and production 
operations — what environmentalists refer to as the concept of the eco-factory. 
Currently, Europe seems to have a lead over the United States in solid- 
waste management and control as a result of government mandates that make 
both producer and consumer responsible for disposing of a product after its 


Introduction 461 

service life is over. Germany, for example, used to send 800,000 metric tons 
of appliances and computers to landfills every year, but as of January 1, 1994, 
producers had to take back and salvage their products and design new ones in 
a way that facilitated recycling. With 3 million metric tons of solid waste gen- 
erated annually in Europe as a result of the disposal of vehicles, the automo- 
tive industry is expected to be targeted next. 

Specialists believe that environmental problems are interrelated and, 
therefore, should be addressed at the same time (see Manufacturing Engi- 
neering, October 1993). In other words, progress has to be made on all fronts 
and not just in one specific area. Accordingly, reduction of wastes and pollution 
at the source should take the form of an overall process with the objective of 
meeting all of the following requirements: 

• Design products for reusability and recycling. 

• Design production processes to eliminate unusable waste, by-products, and 
emissions and to make efficient use of raw materials. Consider the waste, 
not as an unavoidable result of the process, but rather as a factor that ad- 
versely affects the efficiency. 

• Design products to be serviced and maintained easily so as to ensure 
longer service life. This will eliminate one reason for obsolescence of a 
product (i.e., failure) and thus minimize the number of obsolete products 
dumped every year. 

• Establish a material reclamation process based on waste management, re- 
cycling and recovery of materials, and minimal residues. 

You may think that meeting these requirements would be very expensive 
and would, therefore, increase the production cost and make the products less 
competitive. But companies that have successfully adopted such policies claim 
direct and indirect benefits that surpass expenditure. It has been reported that 
over 50 percent of the activities of waste reduction at the source pay back in 
only six months. This actually means that money is saved after the initial pay- 
back period. In addition to meeting the expectations and demands of the pub- 
lic, which has been showing an ever-increasing environmental conscience, 
reducing solid waste at its source can yield the following benefits: 

462 13 Environmentally Conscious Design and Manufacturing 

• Eliminating the cost of the disposal of used products in landfills and junk- 

• Conserving natural resources as a consequence of reusing recovered and 
recycled materials in new products. (This would save the sources of raw 
materials and reduce energy consumption, especially in the aluminum 

• Providing a cash return as a result of selling the recycled material to other 

• Improving yield and quality (as a consequence of reducing waste and scrap) 
and increasing the efficiency of material utilization. 

• Reducing pollution and toxins. 

• Providing safer workplaces where occupational health hazards are absent. 

Before discussing the guidelines for environmentally conscious design and 
manufacture, let us first examine the sources of solid waste and the various 
methods of solid-waste management. 


In our modern societies, there is an abundance of products, such as appliances, elec- 
tronic equipment, and transportation vehicles, that sooner or later have to be dumped 
in landfills. Because parts of the used products are either reused, recycled, or recov- 
ered, the term solid waste is used to describe the parts that remain in landfills. The ex- 
tremely high and ever-increasing annual disposal rates of solid waste can be attributed 
to two main causes: The first is the huge amount of mass-produced appliances and 
electronic equipment sold to consumers every year, and the second is the high mortal- 
ity of those items due to a relatively short service life. The service life is short not 
only because the products fail but also because they go out of style or become tech- 
nologically obsolete. Currently, the service life ranges from 13 years for major appli- 
ances to less than 4 years for personal-care items like hair dryers (50 million of which 
are disposed of annually in landfills worldwide). The following discussion focuses on 
some major sources of solid waste. 

Automotive Industry 

About 30 million vehicles are scrapped every year worldwide, with the shares of Eu- 
rope and the United States being 14 and 10 million, respectively. In the United 
States, more than 90 percent of the vehicles are sent to scrap dealers and then to 
shredders, where the various metals are easily separated and salvaged. Annually, 

13.1 Solid-Waste Sources 463 

about 1 1 million tons of ferrous metals and 800,000 tons of nonferrous metals are re- 
covered. About 30 percent of each vehicle (by weight) is left unrecovered in a land- 
fill. This solid waste, or "fluff," is comprised mainly of various types of plastics and 
rubber. Plastics are particularly hard to recycle because, although they may look the 
same, they have many different chemical structural formulas (see Chapter 8). Fur- 
thermore, scrap plastic may be coated with paint or other chemically dissimilar ma- 
terial. Unfortunately, this landfilled fluff amounts to 3 million tons every year in the 
United States alone, and it is increasing at a steady rate. This increase can be attrib- 
uted to the current trend of using more plastics in cars to reduce the weight of the 
car, provide resistance to corrosion, improve noise-damping characteristics, and en- 
sure excellent thermal insulation properties. 

Appliances Industry 

Examples of appliances that are disposed of in landfills at the end of their service life 
include refrigerators, stoves, dishwashers, and washing machines. These major appli- 
ances usually have a service life of 10 years or more. Small appliances for such uses 
as personal care, entertainment, and coffeemaking are also included under this cate- 
gory. About 350 million appliances, both small and major, were disposed of in land- 
fills worldwide in 1993. As in the case with automobiles, plastic components are 
rapidly replacing metal components previously produced by stamping, die casting, or 

Business Equipment and Computers 

As a result of adopting the philosophy of design for assembly, several metal compo- 
nents in an electronic unit can be replaced by a single, complex, injection-molded plas- 
tic component. Although this design would certainly facilitate assembly, it would 
create environmental problems at the end of the unit's service life because plastic is 
considered to be a major challenge for the recycling industry. 

Housing and Construction Industry 

Plastics and fiber-reinforced plastic composites are finding widespread application in 
the housing and construction industry. In 1995, it was estimated that about 9 percent 
of all plastic solid waste would come from construction. Examples of plastic parts cur- 
rently used (and, of course, eventually requiring disposal) include pipes (water, 
drainage, and sewer), bathtubs, and floor tiles. 

Consumer Goods 

Consumer goods represent the third largest use of plastics after the packaging and con- 
struction industries. As of 1995, about 10 percent of all plastic solid waste was esti- 
mated as coming from scrapped consumer goods. Examples include disposable diapers 
and napkins and throwaway plasticware (utensils, trays, razors, lighters, pens, watches, 
and cameras). 

464 13 Environmentally Conscious Design and Manufacturing 

Furniture Industry 

Plastic furniture is replacing wood furniture, especially for use on beaches, in gardens, 
and in offices. In addition, synthetic carpets are becoming very popular in homes, of- 
fices, and public places. In 1995, the amount of plastic furniture and synthetic carpets 
disposed of in the United States was estimated to be about 3.1 billion pounds, or 7.2 
percent of all plastic solid waste. 

Packaging Industry 

Packaging (e.g., for cosmetics and food) is currently the biggest market for plastics 
and the largest source of plastic solid waste as well (about 44 percent of plastic waste). 
This is due to the very low level of recycling of plastic packaging and is what makes 
paper sometimes more appealing than plastic in the packaging industry. In some cases, 
the paper recycling level goes as high as 50 percent. Nevertheless, paper amounts to 
38 percent of total landfill volume in the United States, as opposed to plastic, which 
comprises o