1chap1 Mater

Design For Manufacturing and Assembly
Effect of Materials and Manufacturing processes on Design
1.EFFECT OF MATERIAL AND

MANUFACTURING PROCESSES IN DESIGN
1. Introduction:
Design is the process of translating a new idea or a market need into the
detailed information from which a product can be manufactured. Each of
its stages requires decisions about the materials from which the product is
to be made and the process for making it. The number of materials
available to the engineer is vast: between 40000 and 80000. At the
beginning the design is fluid and the options are wide; all materials must
be considered. As the design becomes more focused and takes shape, the
selection criteria sharpen and the shortlist of materials, which can satisfy
them, narrows.
Then more accurate data are required and a different way of analyzing the
choice must be used. In final stages of design, precise data are needed
and the search finally comes to only one. The procedure must recognize
the initial choice, the narrow this to a small subset, and provide the
precision and detail on which final design calculations can be based.
The choice of material cannot be made independently of the choice of

process by which the material to be formed, joined, finished, and
otherwise treated. Cost enters, both in the choice of material is processed.
Good design alone will not sell a product. Industrial design is one that, if
neglected, can also loss the manufacturer his market.
So, Engineering materials are evolving faster, so there are wide options,
which pave way for new innovations. It is important in the early part of
design to examine the full materials, which fulfill the requirements, and
subsequently deciding upon the manufacturing processes. For this, the
knowledge of the Effect of material properties and manufacturing
processes is required.
1.1. Major Phases of Design:

Introduction:
Engineering design work is usually performed on three different levels:
1. Development of existing products or designs, i.e.,

redesign, by introducing minor modifications in size, shape
or materials to improve performance.
2. Adaptation of an existing product or design to operate
in new environment or to perform a different function.
1
3. Creation of totally new design that has no precedent.
This work is more demanding in experience and creativity of the
designer.
1.1.1. Major Phases of Design:

Engineering design is usually an iterative process, which involves a series
of decision-making steps where each decision establishes the framework
for the next one. There is no single, universally recognized sequence of
steps that leads to a workable design as these depends on nature of the
problem being solved as well as the size and structure of the organization.
However, a design usually passes through most of the phases, which are
shown in the Fig 1.
1. Identification of the problem and evaluating the need in order to

define the objective of the design represent the first phase of the
design in most cases.
2. Functional requirements and operational limitations are directly
related to the required characteristics of the product and are
specified as a result of the active phase I.
3. System definition, concept formulation, and preliminary layout are
usually completed, in this order, before evaluating the operating
loads and determining the form of the different components or
structural members.
4. Consulting design codes and collecting information on material
properties will allow the designer to perform preliminary material
selection, preliminary design calculations, and rough estimation of
manufacturing requirements.
5. The evaluation phase involves a comparison of the expected
performance of the design with the performance requirements
established in phase 2.Evaluation of the different solution and
selection of the optimum alternative can be performed using
decision-making techniques, modeling techniques, experimental
work and /or prototypes.
6. In some cases, it is not possible to arrive at a design that fulfills all
the requirements and compiles with all the limitations established in
phase2. This means that these requirements and compiles with all
the limitations established in phase 2.
7. Having arrived at final design, the project then enters the detailed
design stage where it is converted in to a detailed and finished form
for suitable for use in manufacturing. The preliminary design layout,
any available detail drawings, models and prototypes, and access to
the developer of the preliminary design usually form the basis of
the detailed design.
8. The next step in the detailed design phase is detailing, which
involves the creation of detail drawings for every part .All the
information that is necessary to unambiguously define the part
should be recorded in detailed drawing. The material of the part
should also be selected and specified by reference to standard
codes.
2
Major phases of design

Constraints
Safety, LOP Fig 1
1. Identification of the problem
Unavailabl Yes Files

e R&D
informatio Patents
n
No
2. Functional requirements
3. Concept formulation
and preliminary layout.
4. Preliminary material and process

Material selection.
properties, Design
Codes
No
Information
sufficient to Modeling and simulation
reach feasible Prototype
solution? Expt.Work.
Yes
5. Evaluate solution with

functional requirements. Sales
Marketing
Prospective customers
No
Acceptable Revise Functional
Design? requirements.
Yes
Detail Design
Detailing Specifications for standard

Materials and processes items.
specified.
Yes
Design
Changes
necessary
No
Marketing
4. Bill of Materials Purchase and
Accounting.
3
Manufacturing
Customer
9. An important part of the detailed design phase is the preparation of

the bill of materials, sometimes called parts list .The bill of materials
is a hierarchical listing of everything that goes into the final product
including fasteners and purchased parts. Close interaction between
design, manufacturing, and materials engineers is important at this
stage.
10.The relationship between the designer and the product does not
usually end at the manufacturing or even delivery stages. The
manufacturing engineer may ask the detailed designer for a change
in some parts to make fabrication easier or cheaper. Finally when
the product gets in to use, the reaction of the consumer and the
performance of the product in service are of concern to the
designer as the feedback represents an important source of
information for the future design modifications.
1.2. Effect of Material Properties on

Design:
Introduction:
Materials are the food of design. A successful product is one that performs
well, is good value for money and gives pleasure to the user. A successful
design should take in to account the function, material properties and
manufacturing processes, as shown in the following fig., in the context of
selection of material, there are many classes of materials metals,
polymers, and ceramics but in the end, what we seek is a profile of
properties.
Function
And
Consumer
Requirement
Component
Design
Manufacturing
Process Material
Properties
Fig 2 Factors that should be considered in component design.
4
This figure shows that there are other secondary relationships between
material properties and manufacturing processes, and between function
and material properties.
The relationship between design and material properties is complex

because the behavior of the material in the finished product is quite
different from that of stock material used in making it. This point is
illustrated in the following Fig.3
Properties of
Stock
materials.
Behavior of
material in
the
Component
Component
Geometry and Effect of
External forces fabrication
method
Fig 3, Factors that should be considered in anticipating the behavior of

material in the component.
This figure shows the direct influence of the stock material properties
production method, and component geometry and external forces on the
behavior of materials in the finished component. It also shows the
secondary relationships exist between geometry and production method,
and between stock materials and component geometry.
1.2.1 Effect of Component Geometry:

In most cases, engineering components and machine elements have to
incorporate design features, which introduce changes in cross-section.
These changes cause localized stress concentrations, which are higher
than those, based upon the nominal cross-section of the part.
1.2.2 Stress Concentration Factor:

A geometrical or theoretical stress concentration factor Kt, is usually used
to relate the maximum stress, Smax, at the discontinuity to nominal
stress, Sav, according to the relationship:
5
Kt = Smax/ Sav
In making a design, Kt is usually determined from the geometry of the
part. Under static loading Kt gives an upper limit to the stress
concentration value and applies only to brittle and notch –sensitive
materials. With more ductile materials, local yielding in the very small
area of maximum stress causes a considerable relief in the stress
concentration. So, for ductile materials under static loading, it is not
usually necessary to consider the stress concentration factor.
Guidelines for design:
Stress concentration can be a source of failure in many cases,

especially when designing with the high-strength materials and under
fatigue loading. In such cases, the following guidelines should be observed
if the stress concentrations are to be kept minimum.
1. Abrupt changes in cross-section should be avoided. If they are
necessary, generous fillet radii or stress-relieving grooves should be
provided.
2. Slots and grooves should be provided with the generous run-out
radii in all corners.
3. Stress-relieving grooves or undercuts should be provided at the
ends of threads and spines.
4. Sharp internal corners and external edges should be avoided.
5. Oil holes and similar features should be chamfered and the bore
should be smooth.
6. Weakening features like the bolt and oil holes, identification marks,
and the part numbers should not be located in highly stressed
areas.
7. Weakening features should be staggered to avoid the addition of
their stress concentration factors.
6
Fig 4 Stress concentration factor on Design.
1.2.3 Designing for Static Strength:

Designs bases on static strength usually aims at avoiding yielding of the
component in the case of soft, ductile materials and at avoiding fracture in
the case of strong, low-toughness materials.
Designing for Simple Axial Loading:
Components and structures made from ductile materials are usually

designed so that no yield will take place under the expected static loading
conditions. When a component is subjected to uniaxial stress, yielding will
take place when the local stress reaches the yield strength of the material.
The critical cross-sectional area, A,
Of such a component can be estimated as :
A= KtnL/YS
Where Kt = Stress concentration factor,
L = applied Load,
N = factor of safety,
YS= yield strength of the material
Designing for Torsional Loading:
The critical cross-sectional area of a circular shaft subjected to torsional

loading can be determined from the relationship:
7
2Ip/d = Kt nT/ ‫ح‬
where d = shaft diameter at the critical cross-section,
‫ =ح‬Maximum shear strength of the material
T = transmitted Torque,
Ip = polar moment of inertia of the cross-section
= π d 4/ 32 for a solid circular shaft
= π(d4o – d4i)/ 32 for a hollow shaft of inner dia di and outer dia do
Design for Bending:
When a relatively long beam is subjected to ending, the bending moment,

the maximum allowable stress, and the dimensions of the cross-section
are related by the equation:
Z = (nM)/YS
where M = bending moment.

Z = section modulus = I/c,
I = moment of inertia of the cross-section with respect to the
neutral axis normal to the direction of the load.
c = distance from the center of gravity of the cross-section to the
outermost fiber.
1.2.4 Designing for Stiffness:

In addition to being strong enough to resist the expected service loads,
there may also be the added requirement of stiffness to ensure that
deflections do not exceed certain limits.
When an initially straight beam is loaded, it becomes curved as a result of

its deflection. As the deflection at a given point increases, the radius of
curvature at this point decreases. The radius of curvature, r, at any point
on the curve is given by the relationship:
r = EI /M
The equation shows us that the stiffness of a beam under bending is

proportional to the elastic constant of the material, E, and the moment of
inertia of the cross-section, I. Therefore, selecting materials with higher
elastic constant and efficient disposition of material in the cross-section
are essential in designing beams for stiffness.
Torsional Rigidity of Shafts:
The torsional rigidity of a component is usually measured by the angle of

twist, ø, per unit length, where
Ø = T/ G Ip
Where G = modulus of elasticity in shear
= E/2(1+v)
Where v = Poisson’s ratio.
The usual practice is to limit the angular deflection in shafts to about 1
degree in a length of 20 times the diameter.
8
1.2.5 Designing With High-Strength, Low Toughness

Materials:
High-strength is being increasingly used in designing critical components
to save weight or to meet difficult service conditions. These materials tend
to be less tolerant of defects than the traditional lower-strength, tougher
materials. While a crack-like defect can safely exist in a part of lower-
strength ductile material, it can cause a catastrophic failure if the same
part is made of a high-strength, low toughness material.
In designing with the high-strength, low toughness materials, the

interaction between fracture toughness of the material, the allowable
crack size, and the design stress should be considered. In the case of high-
strength, low-toughness material, as the design stress increases (or as the
size of the flaw increases) the stress concentration at the edge of the
crack, the stress intensity KI increases until it reaches KIC and fracture
occurs.
KI = KIC = YFs(πa)1/2
where Fs = fracture stress (controlled by the applied load and shape of the
part)
a = quality control parameter (controlled by the manufacturing
method)
Y = dimensionless shape factor. (Estimated experimentally,
analytically or numerically)
1.2.6 Designing against Fatigue:

In majority of cases the reported fatigue strengths or endurance limits of
materials are based on tests of carefully prepared small samples under
laboratory conditions. Such values cannot be directly used for design
purposes because the behavior of the component or structure under
fatigue loading does depend not only on the fatigue or endurance limit of
the material used in making it, but also on several other factors including:
 Size and shape of the component or structure
 Type of loading and state of stress.
 Stress concentration
 Surface finish
 Operating temperature
 Service environment
 Method of Fabrication.
The influence of the above factors on the fatigue behavior of the

component can be accounted for by modifying the endurance limit of the
material using a number of factors. Each of these factors is less than unity
and each one is intended to account for a single effect.
9
Se = ka kb kc kd ke kf kg kh S′ e
Where, Se = endurance limit of the material in the component.
S′ e = endurance limit of the material as determined by laboratory
fatigue test.
ka = surface finish factor.
Surface finish factor varies between unity and 0.2 depending upon
surface finish and strength of the material.
kb = size factor.
Size factor is 1.0 for component diameter less than 10mm; 0.9 for
component diameter in the range of 10 to 50 mm.
kc = reliability factor.
Reliability factor is 0.900 for 90% reliability
0.814 For 99% reliability
0.752 For 99.9% reliability
kd = operating temperature factor.
Operating temperature value is 1.0 in the range of -45° to 450°C
Its value is 1- 5800(T-450) for T between 450° - 550°C
Its value is 1- 3200(T- 840) for T between 840°- 1020°C
ke = loading factor.
Loading factor is equal to 1 for applications involving bending.
It is equal to 0.9 for axial loading.
It is equal to 0.58 for torsional loading.
kf =stress concentration factor.
kg = service environment factor.
Service Environment factor varies from 0.72 to 0.19
kh = manufacturing process factor.
Manufacturing factor is generally taken as 0.3-0.5.
The above equation can be used to predict the behavior of the component
or a structure under fatigue conditions provided that the values of the
different modifying factors are known.
Cumulative Fatigue Damage:
Engineering components and structures are often subjected to different

fatigue stresses in service. Estimation of the fatigue life under variable
loading conditions is normally based on the concept of cumulative fatigue
damage, which assumes that successive stress cycles cause a progressive
deterioration in the component.
The Palmgren -Miner rule, also called Miner's rule proposes that if a cyclic
stressing occurs at a series of stress levels S1, S2, S3…..Si each of which
would correspond to a failure life of N1, N2, N3,….Ni if applied singly, then
the fraction of total life used a each stress level is the actual number of
cycles applied at this level n1, n2, n3, .ni divided by the corresponding life.
The part is expected to fail when the cumulative damage satisfies the
relationship:
n1 n 2 n3 ni
+ + + ......... + =C
N1 N 2 N3 Ni
10
The constant C can be determined experimentally and is usually found to

be in the range of 0.7-2.2. The Palmgren - Miner rule does not take in to
account the sequence of loading nor the effect of mean stress and it
should be taken as rough guide to design.
1.2.7 Designing under High-Temperature

Conditions:
Service temperature has a considerable influence on the strength of
materials and consequently, on the working stress used in design.
Depending on the temperature range, the design can be based on:
1. Short-time properties of the material, i.e., ultimate tensile

strength, yield strength for moderate temperatures.
2. Both the short time and creep properties for intermediate
temperature range.
3. Creep properties of the materials for high temperatures.
In addition to creep, the other factors, which must be taken in to

consideration when designing for elevated temperatures, include:
1. Metallurgical and micro structural changes, which occur in
the material owing to long-time exposure to elevated
temperature.
2. Influence of method of fabrication, especially welding, on
creep behavior.
3. Oxidation and hot corrosion, which may take, place during
service and shutdown periods.
Design guidelines:
For design purposes, creep properties are usually presented on plots,

which yield reasonable straight lines. Common methods of presentation
include log-log plots of stress vs. steady state creep rates and stress vs.
time to produce different amounts of total strain as shown in the Fig.5. A
change in the microstructure of the material is usually accompanied by a
change in creep properties, and consequently a change in the slope of the
line.
Stress (log scale)
Increasing
temperature
11
Creep rate (%/1000h) (log scale)
Fig5, Variation of stress with steady-state creep rate at various

temperatures.
Rupture
Strength
Stress (log scale)
Increasing total
strain
Time (h) (Log scale)

Fig.6, Variation of stress with time to produced different amounts of total
strain at a given temperature.
Larsen- Miller Parameter:
In many cases, creep data are incomplete and have to supplemented or

extended by interpolation or, more hazardously, extrapolation. This is
particularly true of long-time creep and stress-rupture data where the
100,000 hour (11.4 years) creep resistance of newly developed materials
is required. Reliable extrapolation of creep and stress-rupture curves to
longer times can be made only when no structural changes occur in the
region of extrapolation. Such changes can affect the creep resistance,
which would result in considerable errors in the extrapolated values.
The basic idea of these parameters is that they permit the prediction of
long-time creep behavior from the results of shorter time tests at higher
temperatures at the same stress. A widely used parameter for correlating
the stress rupture data is the Larson-Miller parameter (LMP), where LMP is
described as,
LMP = T(C + log tr)
12
Where T= the test temperature in kelvin (°C+273) or degrees Rankine (°F+
460)
tr= time to rupture in hours (the log is to the base 10)
C= the Larson- Miller constant which generally falls between 17
and 23, but is often taken to be 20.
100
Stress (Mpa)
10
20 30 40
T(C+logt)
Fig.7 Larsen-Miller Plots.
Life under Variable Loading
The stress-rupture life of a part or a structure, which is subjected to a

variable loading, can be roughly estimated if the expected life at each
stress level is known. Under such conditions, the life fraction rule assumes
that rupture occurs when:
t1 t2 t3
+ + + .... + = 1.
tr1 tr 2 tr 3
Where t1, t2, t3, are the times spent by the part under stress levels
1, 2, 3… respectively.
tr1, tr2, tr3…. are the rupture lives of the part under stress levels 1,
2, 3...
respectively.
Life under Combined Fatigue And Creep Loading:
Similar reasoning can also be applied to predict the life of a part or a

structure when subjected to combined creep and fatigue loading.
Cumulative fatigue damage laws,e.g. Palmgren-Miner Law, can be
13
combined with the life fraction rule, given in the equation, to give a rough
estimate of expected life under combined creep-fatigue loading. Thus:
t1 t2 t3 n1 n2 n3
+ + +... + + + +..... =1
tr 1 tr 2 tr 3 N1 N2 N3
Where n1, n2, n3... are the number of cycles at stress levels 1, 2, 3…
respectively.
N1, N2, N3… are the fatigue lives at stress levels 1, 2, 3… respectively.
1.3 Effect of Manufacturing Process on

Design
Introduction
It is now widely recognized that design, materials selection, and
manufacturing are intimately related activities, which cannot be
performed in isolation of each other. Creative designs may never develop
into marketable products unless they can be manufactured economically
at the required level of performance. In many cases, design modifications
are made to achieve production economy or to suit existing production
facilities and environment. Modifications of design may also be made in
order to improve quality and performance, in which case the cost of
production may increase.
1.3.1 Design Considerations for Cast Components

Casting covers a wide range of processes which can be used to shape
almost any metallic and some plastics in a variety of shapes, sizes,
accuracy, and surface finish. In some cases, casting represents the
obvious and only way of manufacturing, as in the case of components
made of the different types of cast iron or cast alloys. In many other
applications, however a decision has to be made whether it as
advantageous to cast a product or to use another method of manufacture.
In such cases, the following factors should be considered:
1. Casting is particularly suited for parts which contain internal cavities

that are inaccessible, too complex, or too large to be easily produced
by machining.
2. It is advantageous to cast complex parts when required in large
numbers, especially if they are to be made of aluminum or zinc alloys.
3. Casting techniques can be used to produce a part, which is one of a
kind in a variety of materials, especially when it is not feasible to make
it by machining.
4. Precious metals are usually shaped by casting, since there is little or no
loss of materials.
5. Parts produced by casting have isotropic properties, which could be
important requirements in some applications.
14
6. Casting is not competitive when the parts can be produced by
punching from sheet or by deep drawing.
7. Extrusion can be preferable to casting in some cases, especially in the
case of lower- melting nonferrous alloys.
8. Castings are not usually a viable solution when the material is not
easily melted, as in the case of metals with very high melting points
such as tungsten.
A general rule of solidification is that the shape of the casting should allow
the solidification front to move uniformly from one end toward the feeding
end, i.e. directional solidification. This can most easily be achieved when
the casting has virtually uniform thickness in all sections. In most cases
this is not possible. However, when section thickness must change, such
change should be gradual, in order to give rise to stress concentration and
possible hot tears in the casting. Figure 8.gives some guidelines to avoid
these defects.
Another problem, which arises in solidification, is caused by sharp corners;

these also give rise to stress concentration and should be replaced by
larger radii. When two sections cross or join, the solidification process is
interrupted and a hot spot results. Hot spots retard solidification and
usually cause porosity and shrinkage cavities.
Effect of material properties

The type and composition of the material play an important part in
determining the shape, minimum section thickness, and strength of the
casting. Materials, which have large solidification shrinkage and contain
low – melting phases are susceptible to hot tears. Another material
variable is cast ability, which can be related to the minimum section
thickness, which can be achieved. It should be noted that the shape and
size of the casting as well as the casting process and foundry practice
could affect the minimum section thickness.
Correct Designs Incorrect designs
Solidifications of intersecting sections results in hot

spots and shrinkage activities
15
Stagger section Use a core or Use External Use a riser

internal chill Chills
Fig 8
1.3.2 Design Considerations for Molded Plastic

Components
Compression, transfer, and injection molding processes are the commonly
used methods of molding plastic components. These processes involve the
introduction of fluid or a semi fluid material into a mould cavity and
permitting it to solidify into the desired shape.
Guidelines for design
Experience shows that the mechanical, electrical, and chemical properties

of molded components are influenced by the flow of the molten plastic as
it fills the mold cavity. Streamlined flow will avoid gas pockets in heavy –
sectioned areas.
An important common feature in molding processes is draft, which is

required for easy ejection of molded parts from the mold cavity. A taper of
1 to 4 degree is usually used for polymers, but tapers of less than 1
degree can be used for deep articles. Another common feature is the
uniform thickness. Non-uniformity of thickness in a molded piece tends to
produce non-uniform cooling and unbalanced shrinkage leading to internal
stresses and warpage.
If thickness variations are necessary, generous fillets should be used to

allow a gradual change in thickness. The effect of junctions and corners
can also be reduced by using a radius instead, as shown in Fig 9.The
nominal wall thickness must obviously such that the part is sufficient
strong to carry the expected service loads. However, it is better to adjust
the shape of the part to cope with the applied load than to increase the
wall thickness. This is because thick sections retard the molding cycle and
require more materials.
The presence of holes disturbs the flow of the material during molding and
a weld line occurs in the side of the hole away from the direction of flow.
This results in a potentially weak point and some from of strengthening,
such as bosses may be necessary as in Fig 10.Through holes are preferred
to blind holes from a manufacturing standpoint. This is because core prints
can often be supported in both halves of the mold in the case of through
holes, but can only be supported from one end in the case of blind holes.
Accuracy of molded parts.
Dimensional tolerances in molded plastic parts are affected by the type

and constitution of the material, shrinkage of the material, heat and
16
pressure variables in the molding process, and the toolmaker’s tolerances
on the mold manufacture. Shrinkage has two components:
Mold shrinkage, which occurs upon solidification; and
After shrinkage, this occurs in some materials after 24 hours.
For example, a thermosetting plastic like melamine has mold shrinkage of
about 0.7 to 0.9 %, and an after shrinkage of 0.6 to 0.8%. Thus a total
shrinkage of about 1.3 to 1.7 % should be considered. On the other hand,
a thermoplastic like polyethylene may shrink as much as 5% and nylon as
much as 4%. In addition, the value of tolerance depends on the size of the
part. Larger dimensions are normally accompanied by larger tolerances.
For example, dimensions less than 25mm (1 in) can be held within ± 50
µ m. Larger dimensions are usually given tolerances of ± 10 to 20 µ m/cm.
The value of tolerances also depends on the direction in relation to the
parting plane.
Poor Design Better Design
(a)
Fig 9 some design features of plastic parts. (a) Using radii instead of sharp
corners.
(b)
Fig 10 some design features of plastic parts (b) Use of bosses to

strengthen areas round holes and slots.
1.3.3 Design Considerations for Forged Components
17
Forging processes represent an important means of producing relatively
complex parts for high-performance applications. In many cases forging
represents a serious competitor to casting especially for solid parts that
have no internal cavities. Forged parts have wrought structures, which are
usually stronger, more ductile, contain less segregation, and are likely to
have less internal defects than cast parts. This is because the extensive
hot working, which is usually involved in forging, closes existing porosity,
refines the grains, and homogenizes the structure.
On the other hand, cast parts are more isotropic than forged parts, which
usually have directional properties. This directionality is due to the fibre
structure, which results from grain flow and elongation of second phases
in the direction of deformation. Forged components are generally stronger
and more ductile in the direction of fibres than across the fibres.
Guidelines for Design
Rapid changes in thickness should be avoided because these could result

in laps and cracks in the forged metal as it flows in the die cavity. To
prevent these defects, generous radii must be provided at the locations of
large changes in thickness. Another similarity with casting is that vertical
surfaces of a forging must be tapered to permit removal from the die
cavity.
A draft of 5 to 10 degrees is usually provided. It is better to locate the

parting line near the middle of the part in order to avoid deep impression
in either of the two halves of the die and allows easier filling of the die
cavity. A design would be more economically produced by forging if
dimensions across the parting line are given appropriate mismatch
allowance, and parallel dimension are given a reasonable die closure
allowance. Specifying close tolerances to these dimensions could require
extensive machining which would be expensive.
Machined Forged
Fig 11 Schematic comparison of the grain flow in forged and machined

components.
18
1.3.4 Design Considerations For Powder Metallurgy

Parts
Powder metallurgy (P/M) techniques can be used to produce a large
number of small parts to the final shape in few steps, with little or no
machining, and at high rates. Many metallic alloys, ceramic materials, and
particulate reinforced composites can be processed by P/M techniques.
Generally, parts produced by the traditional P/M techniques contain 4 to
10 vol % porosity. The amount of porosity depends on part shape, type
and size of powder, lubrication used, pressing pressure, sintering
temperature and time, and finishing treatments.
The distribution and volume fraction of porosity greatly affect the

mechanical, chemical, and physical properties of parts prepared by P/M
techniques. An added advantage of P/M is versatility. Materials that can be
combined in no other way can be produced by P/M. Aluminum - graphite
bearings, copper - graphite electrical brushes, cobalt - tungsten carbide
cutting tools (cermets), and porous bearings and filters are such.
The Powder Metallurgy Parts Association and Metal Powder Industries

Federation have made certain rules. They are:
1. The shape of the part must permit ejection from the die, Fig 12
2. Parts with straight walls are preferred. No draft is required for ejection
from lubricated dies.
3. Parts with undercuts or holes at right angles to the direction of pressing
cannot be made, Fig 13.
4. Straight serrations can be made easily, but diamond knurls cannot, Fig
14.
5. Since pressure is not transmitted uniformly through a deep bed of
powder, the length/diameter ratio of a mechanical pressed part should not
exceed about 2.5: 1.
Fig 12 Reverse taper should be avoided, use parallel sides and machine
the
Required taper after sintering.
19
Fig 13 undercuts and holes at right angles to pressing direction should be

avoided; if necessary such features are introduced by machining after
sintering.
Fig 14 Diamond knurls should be replaced by straight serrations.
1.3.5 Design of Sheet - Metal Parts
Parts made from sheet metal cover a wide variety of shapes, sizes, and
materials. Many examples are found in the automotive, aircraft, and
consumer industries. Generally, sheet-metal parts are produced by
shearing, bending, and/or drawing. The grain size of the sheet material is
important and should be closely controlled. Steel of 0.035 - 0.040 mm
(0.001 - 0.0016) grain size is generally acceptable for deep- drawing
applications. When formability is the main requirement in a sheet
material, drawing - quality low carbon steels represents the most
economic alternative.
The most important factor, which should be considered when designing

parts that are to be made by bending, is bend ability. This is related to the
ductility of the material and is expressed in terms of the smallest bend
radius that does not crack the material. Bend ability of a sheet is usually
expressed as 2T, 3T, 4T, etc. A 2T material has greater bend ability than a
3T material.
Another factor which should be considered when designing for bending is

spring back, which is caused by the elastic recovery of the material when
the bending forces are removed. One way of compensating for spring back
is to over bend the sheet. Another method is bottoming which eliminates
the elastic recovery by subjecting the bend area to high-localized stresses.
20
1.3.6 Designs Involving Joining Processes
The major function of a joint is to transmit stress from one part to another
and in such case the strength of the joint should be sufficient to carry the
expected service loads. In some applications, tightness of the joint is also
necessary to prevent leakage. Because joints represent areas of
discontinuities in the assembly, they should be located in low-stress
regions especially in dynamically loads structures.
Welding
Welding has replaced riveting in many applications including steel

structures, boilers, tanks, and motorcar chassis. This is because riveting is
less versatile and always requires lap joint. Also, the holes and rivets
subtract from strength, and a riveted joint can only be about 85%as
strong, whereas a welded joint can be as strong as the parent metal.
Welded joints are easier to inspect and can be made gas and liquid-tight
without the caulking which has to be done in riveted joints. On the
negative side, however, structures produced by welding are monolithic
and behave as one piece. This could adversely affect the fracture behavior
of the structure. For example, a crack in one piece of a multipiece riveted
structure may not be serious, as it will seldom progress beyond the piece
without detection. However, in the case of a welded structure, a crack that
starts in a single plate or weld may progress for a large distance and
cause complete failure.
Another factor, which should be considered when designing a welded

structure, is the effect of size on the energy-absorption ability to steel. A
Charpy impact specimen could show a much lower brittle-ductile transition
temperature than a large welded structure made of the same material.
Guidelines for design of weldments
1.Welded structures and joints should be designed to have sufficient

flexibility. Structures that are too rigid do not allow shrinkage of the weld
metal, have restricted ability to redistribute stress, and are subjected to
distortions and failure.
2. Accessibility of the joint for welding, welding position and component

match-up are important elements of the design.
3. Thin sections are easier to weld than thick ones.
4. Welded section should be about the same thickness to avoid excessive

heat distortion.
5. It is better to locate welded joints symmetrically around the axis of an

assembly in order to reduce distortion.
6. Whenever possible the meet of several welds should be avoided.
21
7. Use weld fixtures and clamps to avoid distortion.
Adhesive Bonding
Adhesives represent an attractive method of joining and their use is

increasing in many applications. Some of main advantages in using
Adhesives are as follows:
1. Thin sheets and parts of dissimilar thickness can be easily bonded.

2. Adhesive bonding is the most logical method of joining polymer-
Matrix composites.
3. Adhesives are electrical insulators and can prevent galvanic
Action in joints between dissimilar metals.
4. Flexible adhesives spread bonding stresses over wide areas and
Accommodate differential thermal operation.
5. Flexible adhesives can absorb shocks and vibrations, which
Increases fatigue life.
6. The preparation of bonded joins requires no fastener holds, which
Gives better structural integrity and allows thinner gage materials to
be used.
The main limitations of adhesives are as follows:
1. Bonded joints are weaker under cleavage and peel loading than
under tension or shear.
2. Most adhesives cannot be used at service temperatures above 300
degree C(600 degree F).
3. Solvents can attack adhesive-bonded joints.
4. Some adhesives are attacked by ultraviolet light, water, and ozone.
5. The designer should also be aware of the adhesive's impact resistance
and creep, or cold flow, strength.
Design of adhesive joints
The strength of the adhesive joint depends on the geometry, the direction
of loading in relation to the adhesive material, surface preparation, and
application and curing technique. As the bonded area limits the strength
of an adhesive joint, lab and double-strap joints are generally prepared to
butt joints. If the geometry constrains do not allow for such joints, a scarf
or double -scarf joint should be made.
When a lab joint is used to bond thin sections, tensile shear causes
deflection, and this results in stress concentration at the end of the lab.
Tapering the ends of the joints, gives more uniform loading throughout the
joint. Since adhesive joints are weaker under peeling forces, joint design
should avoid this type of loading.
1.3.7 Designs Involving Heat Treatment:

Heat treatment represents an important step in the sequence of
processes that are usually performed in the manufacture of metallic parts.
Almost all ferrous and many nonferrous alloys can be heat treated to
22
achieve certain desired properties. Heat treatment can be used to make
the material hard and brittle, as in the case of annealing.
Generally, hardening of steels involves heating to the austenitic

temperature range, usually 750 to 900 °C (1400 to 1650° F), and then
quenching to form the hard martensitic phase. The nonuniform
temperature distribution that occurs during quenching and the volume
change that accompanies the martenstic transformation can combine to
cause distortions, internal stresses, and even cracks in the heat treated
part. Internal stresses can warp or dimensional changes when the
quenched part is subsequently machined or can combine with externally
applied stresses to cause failure. Corrosion problems can also be
aggravated owing to the presence of internal stresses. These difficulties
can be reduced or eliminated by selecting steels with hardenability as
they require a less cooling rate to achieve a given hardness value.
Manganese, chromium, molybdenum are commonly added to steels to
increase their hardenability.
1.3.8 Designs Involving Machining Processes

The following discussion illustrates some component shapes and features
which can cause difficulties in machining, take an undue length of time to
machine, call for precision and skill that may not be available, or which
may even be impossible to machine by standard machine tools and
cutting tools.
1. The workpiece must have a reference surface, which is suitable for
holding it on the machine tool or in a fixture. This could be a flat base
or a cylindrical surface.
2. Whenever possible, the design should allow all the machining
operations to be completed without resetting or reclamping.
3. Whenever possible, the radii between the different machined surfaces
should be equal to the nose radius of the cutting tool.
4. If the part is to be machined by traditional cutting methods, deflection
under cutting forces should be taken into account. For the same cutting
force, the deflection is higher for thinner parts and for lower elastic
moduli. Under these conditions, some means of support is necessary to
ensure the accuracy of the machined part.
5. Features at an angle to the main machining direction should be
avoided as they may require special attachments or tooling. Fig 15
6. To reduce the cost of machining, machined areas should be minimum
as shown Fig 16
7. Cutting tools often require run-out space, as they cannot be retracted
immediately. This is particularly important in the case of grinding
where the edges of the grinding wheel wear out faster than the center.
Fig 17 gives some examples to illustrate this point.
23
Fig 15 (a) Poor design as drill enters and exists at an angle to the surface.
(b) Better design, but drilling the holes need a special attachment.
(c) Best design.
Poor design Better Design
(a) (b) (c)
Added materials to
reduce machine
area
Relief’s to
reduce
machined
areas
Fig 16 Some design details which can be introduced to reduce machining.
Fig 17 Some design details which can be introduced to give run-out for
grinding wheels.
1.4.The Materials Selection Process:
24
One of the most important requisites for the development of a satisfactory
product at a competitive cost is making sound economic choices of
engineering designs, materials, and manufacturing processes. The large
number of materials and the many manufacturing process available to the
engineer, coupled with the complex relationships between the different
selection parameters, often make the selection process a difficult task. A
rigorous and through approach to materials selection is, however, often
not followed in industry and much selection is based on past experience.
It is often said, “When in doubt make it stout out of the stuff you know
about.” While it is unwise to totally ignore past experience, the frequent
introduction of new materials and manufacturing process, in addition to
the increasing pressure to produce more economic and competitive
products, make it necessary for the engineer to be always on the lookout
for possible improvement. The reasons for reviewing the types of material
and processes used in making an existing product are:
1. Taking advantage of new materials or processes.

2. Improving service performance, including longer life and
higher reliability.
3. Meeting new legal requirements.
4. Accounting for changed operating conditions.
5. Reducing cost and making the product more competitive.
Selecting the optimum combination of material and process can be

performed at one certain stage in the history of a project; it should
gradually evolve during the different stages of product development.
These are:
1. Analysis of the performance requirements.

2. Development of alternative solutions to the problem.
3. Evaluation of the different solutions.
4. Decision on the optimum solution
1.4.1 Analysis of the Material Performance

Requirements:
Functional Requirements:
Functional requirements are directly related to the required characteristics

of the part or the product. For example, if the part carries a uniaxial
tensile load, the yield strength of the material can be directly related to
the load-carrying capacity of the product. For the evaluation process of
the characteristics of material properties like thermal shock resistance,
wear resistance, reliability etc., and simulation service tests are employed.
Processability Requirements:
The processability of the material is a measure of its ability to be worked

and shaped in to a finished part. With the reference to a specific
manufacturing method, processability can be defined as a castability,
25
weldability, machinability etc.,Ductility and hardenability can be relevant
to processability if the material is to be deformed or hardened by heat
treatment respectively. The closeness of the stock form to the required
product form can be taken as a measure of processability in some cases.
The material properties are closely related to functional requirements.
Cost:
Cost is usually the controlling factor in evaluating materials, because in

many applications there is a cost limit for a material intended to meet the
application requirements. When the cost limit is exceeded, the design may
have to be changed to allow the use of a less expensive material. The cost
of the processing often exceeds the cost of the stock material.
Reliability Requirements:
The reliability of the material can be defined as the probability that it will
perform the intended function for the expected life without failure.
Material reliability is difficult to measure, because it is not only dependent
upon the material’s inherent properties, but also greatly affected by its
production and processing history.
Though there are difficulties in evaluating reliability, it is often an

important selection factor that must be taken in to account. Failure
analysis techniques are usually used to predict the different ways in which
a product can fail, and can be considered as a systematic approach to
reliability evaluation.
Resistance to Service Conditions:
The environment in which the product or part will operate plays an

important role in determining the material performance requirements.
Corrosive environments, as well as high or low temperatures, can
adversely affect the performance of most materials in service. Whenever
there is more than one material involved in an application, compatibility
becomes a selection consideration. For example, In thermal environment,
the coefficient of thermal expansion of all the materials involved may have
to be similar in order to avoid thermal stresses. In applications where
relative movements exist between different parts, wear resistance of the
materials involved should be considered.
1.4.2 Cost per Unit Property Method:

In simplest cases of optimizing the selection of materials, one property
stands out as the most critical service requirement. In such simple cases
the cost per unit property can use as a criterion for selecting the optimum
material. Consider the case of a bar of given length (L) to support a tensile
force (F). The cross-sectional area (A) of the bar is given by:
A=F/S ()
Where S is the working stress of the material, which is related to its yield
strength by an appropriate factor of safety.
The cost of the bar is given by:
26
C′ = Cρ AL = (Cρ FL)/S
Where C = cost of the material per unit mass, and
ρ = Density of the material.
In comparing different candidate materials, only the quantity (Cρ )/S,
which is the cost of unit strength, needs to be compared, as F and L are
constant for all material. The material with the lowest cost per unit
strength is the optimum material.
When one material is considered as a substitute for an existing material,
the two materials a and b can be compared on the basis of relative cost
per unit strength (RC′ ):
RC′ = (C′ )a
(C′ )b
which is equal to Caρ aSb
Cbρ bSa
RC′ less than unity indicates that the material a is preferable to material
b.
Equations similar to () an () can be used to compare the materials on cost
basis.
1.4.3 Weighted Properties Method:

The weighted properties method can be used in optimizing materials
selection when several properties should be taken into consideration. In
this method each material requirement, or property, is assigned a certain
weight, depending on its importance. A weighted property value is
obtained by multiplying the numerical value of the property by the
weighting factor (α ). The individual weighted property values of each
material are then summed to give a comparative materials performance
index (γ ). The material with the highest performance index (γ ) is
considered as the optimum for the application.
When evaluating a list of candidate materials, one property is considered

at a time. The best value in this list is rated as 100 and the others are
scaled proportionally.
B= scaled property = Numerical value of property x 100
Maximum value in the list
For properties like cost, corrosion or wear loss, weight gain in

oxidation, etc., a lower value is more desirable. In such cases, the lowest
value is rated as 100 and B is calculated as:
B= scaled property = Minimum value in the list x 100

Numerical value of property
For material properties that can be represented by numerical values,

applying the above procedure is simple. However, with properties like
corrosion and wear resistance, machinability and weldability, etc., are
rarely given and materials are usually rated as very good, good, fair, poor
etc. In such cases, the rating can be converted to numerical values using
an arbitrary scale. For example, a corrosion resistance rating of excellent,
27
very good, good, fair and poor can be given numerical values of 5,4,3,2
and 1 respectively. Then,
n
Material performance index,γ = ∑ Biα i
i=1
Where i is summed over all the n relevant properties.
In the cases where numerous material properties are specified, the digital
logic approach is used as a systematic tool to determine α . In this
procedure evaluations are arranged such that only two properties are
considered at a time. Every possible combination of the properties or
performance goals is compared and no shades of choice are required, only
a yes or no decision for each evaluation. To determine the relative
importance of each properties or goal a table is constructed, the
properties or goals are listed in the left hand column, and comparisons are
made in the columns to the right, as shown in the table.
Table 5.1 Determination of the relative importance of performance goals

using the digital logic method
Goals Number of possible decisions Positive Relative Emphasis

[N=n(n-1)/2] decisions Coefficient (α )
1 2 3 4 5 6 7 8 9 1
0
1 1 1 0 1 3 α 1=0.3
2 0 1 0 1 2 α 2=0.2
3 0 0 1 0 1 α 3=0.1
4 1 1 0 0 2 α 4=0.2
5 0 0 1 1 2 α 5=0.2
Total number of positive =10 ∑α =1.0
decisions
In comparing two properties or performance goals, the more important is

given numerical one (1) and the less important is given zero(0).The total
number of possible decisions N=n(n-1)/2 , where n is the number of the
properties or goals under consideration. A relative emphasis coefficient or
weighting factor,α for each goal is obtained by dividing the number of
positive divisions for each goal (m) into the total number of possible
decisions (N). In this case ∑α =1.
However, if there are large numbers of properties to consider the

importance of cost may be emphasized by considering it separately as a
modifier to the material performance index (γ ). In the cases where the
material is used for space filling, cost can be introduced on a per unit
volume basis. A figure of merit (M) for the material can then be defined as:
M=γ /(Cρ )
Where C= total cost of the material per unit weight (stock, processing,
finishing, etc.)
28
ρ = Density of the material
The weighted properties method can be used when a material is

considered as a substitute for an existing one. This is done by computing
the relative figure of merit (RM), which is defined as,
RM = Mn/Mc
Where Mn and Mc are the figures of merit of the new and existing
materials respectively. If the RM is greater than unity, the new material is
more suitable than the existing material.
The steps involved in the weighted properties method can be written in
the form of a simple computer program to select materials from the data
bank. An interactive program can also include the digital logic method to
help in determining weighting factors.
1.4.4 Limits On Properties Method:

In the limits on properties method, the performance requirements are
divided into three categories:
1. Lower limit properties

2. Upper limit properties
3. Target value properties
The limits on properties method are usually suitable for optimizing
material and process selection when the number of possible alternatives is
relatively large. This is because the limits, which are specified for the
different properties, can be used for eliminating unsuitable materials from
data bank. The remaining materials are those whose properties are above
the lower limits, below the upper, and within the limits of target values of
the respective specified requirements. After the screening stage, the limits
on properties method can be used to optimize the selection from among
the remaining materials.
As in the case of the weighted properties method, each of the
requirements or properties is assigned a weighted factor, α , which can be
determined using the digital logic method, as discussed earlier. A merit
parameter, m,is then calculated for each material according to the
relationship:
 Y   X   X 
m =  ∑α  +  ∑ α
i
 +  ∑ αj
− 1  k
X   Y   Y
i j k
 i 1 j 
k
where l,u, and t stand for lower limit, upper limit, and target value
properties respectively.
nl,nu,and nt are the numbers of the lower limit, upper limit, and
target value properties respectively.
α i, α j, α k are the weighting factors of the lower limit, upper limit,
and target value properties respectively.
Xi,Xj and Xk are the candidate material lower limit, upper limit, and
target value properties respectively.
29
Yi,Yj,and Yk are the specified lower limit, upper limit, and target
value properties respectively.
According to the equation the lower the value of the merit parameter m,
the better the material.
As in the weighted properties method, the cost can be considered in two

ways:
1. Cost is treated as an upper limit property and given the
appropriate weight.
2.Cost is included as a modifier to the merit parameter as follows:
m´ = (CX/CY)m
Where CY and CX are the specified cost upper limit and candidate material
cost,
In this case the material with the lowest cost-modified merit parameter, m
´, is the optimum.
1.5. Case Study for Material Selection:

1.5.1 Materials for springs:
Springs come in many shapes as shown in the Fig 18, and have
many purposes: one thinks of axial springs, leaf springs, helical springs,
spiral springs, torsion bars. Regardless of their shape or use, the best
material for a spring of minimum volume is that with the greatest value of
σ 2f / E , and for minimum weight it is that with the greatest value σ 2f / Eρ .
30
Fig 18 springs store energy. The best material for any spring, regardless of
its shape or the way in which it is loaded, is that with the highest value of
σ 2f / E Or if weight is important, σ 2f / E .
The primary function of the spring is that of storing elastic energy and
when required releasing it again.The elastic energy stored per unit volume
of material stressed uniformly to a stress σ is
1 σ2
Wv =
2 E
Where E is young’s modulus. It is Wv that to be maximize. The spring will
be damaged if the stress σ exceeds the yield stress or the failure stress
σ f; the constraint is σ <=σ f. So the maximum energy density is
2
1σ f
Wv =
2 E
Torsion bars and leaf springs are less efficient than axial springs because
much of the material is not fully loaded: the material at the neutral axis,
for instance, is not loaded at all. For torsion bars
2
1σ f
Wv =
3 E
And for leaf springs,
2
1σ f
Wv =
4 E
But this has no influence on the choice of the material. The best material
for springs is that the biggest value of
σ2f
M1 =
E
If weight rather than the volume, matters, we must divide this by
the density ρ (giving energy stored per unit weight) and seek materials
with the high value of
σ2 f
M2 =
ρE
31
Fig 19 Materials for small springs. High strength (‘spring’) steel is good.
Glass, CFRP and GFRP all under right circumstances, make good springs.
Elastomers are excellent. Ceramics are eliminated by their low tensile
strength.
1.5.2 The Selection

The choice of materials for springs of minimum volume is shown in
the Fig 19 family lines of slope ½ link materials with equal values of
σ2f
M1 =
E
Those with the highest values of M1 lie towards the bottom right.
The heavy line is one of the families; it is positioned so that a subset of
materials is left exposed. The best choices are a high0strength steel
(spring steel) lying near the top end of the line, and at the other end,
32
rubber. But certain other materials are suggested too: GFRP (truck
springs), titanium alloys, glass and
Nylon.
1.6.Problem :
1. Suggest a suitable operation sequence for the stub carrier shown in
Fig.20 and redraw the component incorporating features to facilitate
manufacture. The carrier is to be produced from a steel casting and the
symbol indicates a ground surface for the 30 mm diameter f8 limits.
33
2. The proposed machining procedure for the plate Fig 21

(1) Bore and face, reverse, face other side - turret.
(2) Drill and ream four 25 mm H8holes - drill, drill jig.
Suggest a design modification which will permit of an alternative
procedure to achieve a substantial reduction in machining time. State the
procedure for producing the modified design.
3. A Cast iron bearing bracket is shown in Fig 22. Indicate the preferred
parting line and any necessary sand cores. Offer a design modification
that will reduce or eliminate the need for sand cores.
34
35

1chap1 Mater

Diunggah oleh

Informasi Dokumen

Deskripsi Asli:

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

1chap1 Mater

Diunggah oleh

Hak Cipta:

Format Tersedia

Design For Manufacturing and Assembly

Effect of Materials and Manufacturing processes on Design

1.EFFECT OF MATERIAL AND

The choice of material cannot be made independently of the choice of

1.1. Major Phases of Design:

1. Development of existing products or designs, i.e.,

1.1.1. Major Phases of Design:

1. Identification of the problem and evaluating the need in order to

Major phases of design

Unavailabl Yes Files

4. Preliminary material and process

5. Evaluate solution with

Detailing Specifications for standard

9. An important part of the detailed design phase is the preparation of

1.2. Effect of Material Properties on

Fig 2 Factors that should be considered in component design.

The relationship between design and material properties is complex

Fig 3, Factors that should be considered in anticipating the behavior of

1.2.1 Effect of Component Geometry:

1.2.2 Stress Concentration Factor:

Guidelines for design:

Stress concentration can be a source of failure in many cases,

Fig 4 Stress concentration factor on Design.

1.2.3 Designing for Static Strength:

Designing for Simple Axial Loading:

Components and structures made from ductile materials are usually

Designing for Torsional Loading:

The critical cross-sectional area of a circular shaft subjected to torsional

Design for Bending:

When a relatively long beam is subjected to ending, the bending moment,

where M = bending moment.

1.2.4 Designing for Stiffness:

When an initially straight beam is loaded, it becomes curved as a result of

The equation shows us that the stiffness of a beam under bending is

Torsional Rigidity of Shafts:

The torsional rigidity of a component is usually measured by the angle of

1.2.5 Designing With High-Strength, Low Toughness

Guidelines for design:

In designing with the high-strength, low toughness materials, the

1.2.6 Designing against Fatigue:

The influence of the above factors on the fatigue behavior of the

Cumulative Fatigue Damage:

Engineering components and structures are often subjected to different

The constant C can be determined experimentally and is usually found to

1.2.7 Designing under High-Temperature

1. Short-time properties of the material, i.e., ultimate tensile

In addition to creep, the other factors, which must be taken in to

For design purposes, creep properties are usually presented on plots,

Creep rate (%/1000h) (log scale)

Fig5, Variation of stress with steady-state creep rate at various

Time (h) (Log scale)

Larsen- Miller Parameter:

In many cases, creep data are incomplete and have to supplemented or

LMP = T(C + log tr)

Fig.7 Larsen-Miller Plots.

Life under Variable Loading

The stress-rupture life of a part or a structure, which is subjected to a

Life under Combined Fatigue And Creep Loading:

Similar reasoning can also be applied to predict the life of a part or a

1.3 Effect of Manufacturing Process on

1.3.1 Design Considerations for Cast Components

1. Casting is particularly suited for parts which contain internal cavities