Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
of Shell Structures
by
M.FARSHAD
EMPA,
Switzerland
"
~.
ISBN 978-90-481-4200-2
Preface xi
Appendixes 375
Subject Index
Preface
Shell Structures present immense structural and architectural potential in various fields of
civil, mechanical, architectural, aeronautical, and mru.ine engineering. Examples of shell
structures in civil and architectural engineering are: varieties of concrete shell roofs, liquid
retaining structures and water tanks, concrete silos, cooling towers, containment shells of
nuclear power plants, and concrete arch dams. In mechanical engineering, shell forms are
used in piping systems, curved panels, and in pressure vessel technology. Aircrafts,
spacecrafts, missiles, ships, and submarines are examples of shells used in aeronautical and
marine engineering. Shells are found in various biological forms such as the eye and the
skull, plants, and animal shapes. Thus, another application of shell engineering would be the
field of Biomechanics.
Shell structures developed since ancient times and now ru.·e being increasingly used in various
industries. Shells are used in the covering of large spans, liquid retaining installations, silos,
and containment structures. They are also used in the construction of light-weight vehicles,
pressure vessels, and space structures. Advent of such materials as ferro-cement, fiber-
reinforced concrete, composite materials, and reinforced polymers have all enhanced the
domain of shell technology. With the development of new prefabrication schemes as well as
the need for recycling of materials, the potential of shell applications has further increased. In
addition to mechanical advantages, such as durability, high strength and stability, shell
structures enjoy the unique position of having extremely high aesthetic value in various
architectural designs.
In spite of all these features and potential applications, many engineers and architects are
unacquainted with sheHs as weH as the aspects of sheH behavior and design. The purpose of
this book is to familiarize the engineering and architectural students, as weH as practicing
engineers and architects, with the behavior and design aspects of shell structures. The goal of
this book is to present three aspects: the physical behavior, the structural analysis, and the
xi
xii
design of shells in a simple, integrated, and yet concise fashion. Thus, the book contains three
major aspects of shell engineering. These are: (1) physical understanding of shell behavior,
(2) use of applied shell theories, (3) development of design methodologies together with shell
design examples.
To ac hieve these goals, simplified shell theories have been discussed in this book and have
been immediately applied to actual problems. In this sense, the book bridges the gap between
the elaborate theoretical treatments of shells, on the one hand, and, the practical aspects of the
analysis and design of shells, on the other hand. Being aware of a wide variety of existing
numerical routines for shells analysis, we have, nevertheless, made use of simple analytical
schemes of shell analysis so that the designer can understand the analysis procedure and to
perform parametric studies. The theoretical tools required for rational analysis of shells are
kept at a modest level so that engineering and architectural students, as weH as practicing
engineers and architects, can grasp the fundamentals of sheH behavior and, at the same time,
understand the related theory and be able to apply it to actual design problems. To achieve a
physical understanding of complex shell behavior, quantitative presentations are
supplemented by qualitative discussions so that the reader can grasp a "physical feeling" of
shell behavior. To make the book useful as a reference manual, a number of analysis and
detailed design examples are also worked out in various chapters.
The actual design of sheHs, involves the use of appropriate codes of practice. Thus, while
making use of some existing codes on shells, in order to provide a text that could be used in
various countries, we have attempted to present the designs apart from the existing codes. In
some cases, the common guidelines provided by several standards, including ACI, BS, DIN,
and IS, have been used.
This book can be used as a text book, and I or a reference book in undergraduate as weH as
graduate university courses in the fields of civil, mechanical, architectural, aeronautical, and
materials engineering. It can also be used as a reference and design-analysis manual for the
practicing engineers and architects. To make the book useful to design engineers and
architects, the text is supplemented by a number of appendices containing tables of shell
analysis and design charts and tables. Metric system is used throughout this book.
The material of this book have been developed through many years of teaching at the
Universities of Shiraz and Tehran, University ofToronto, and the Swiss Federal Institute of
Technology (ETHZ) as weH as through research and practical design experience by the
author. Thus, in the development of this text, various viewpoints and experiences have been
extremely constructive.
The author would like to thank the Swiss Federal Laboratories for Materials Testing and
Research (EMPA, Dübendorf) and, in particular, Professor F. Eggimann and Professor U.
Meier and Hr. H. Fritz for supporting this project Special thanks are due to Hr. P. Flüeler
who has given great encouragement and support in bringing this book to its present
publication. The author would also like to thank professor Gladwell for his useful comments
on the manuscript. This book is dedicated to my family (Gowhar, Anahita, and Mazda) who
have shown great patience during the long period of manuscript preparation.
M. Farshad
Switzerland
Introduction to SheIls 1
Chapter 1
Introduction to Shells
1.1 - Introduction
Generally speaking, shells are spatially curved surface structures which support external
applied loads. Shells are found in a variety of natural structures such as eggs, plants, leaves,
skeletal bones, and geological fonns. Shell structures have also been built by man since the
most ancient times. Many shell domes buHt of masonry and stone in ancient times, are still
in existence in some parts of the world.
In this introductory chapter, first some references will be made to a number of important
and weIl-known fields of applications of sheIl structures. Then, engineering classifications
of the surfaces, as the main geometrical feature of sheIls, will be presented. In the
forthcoming chapters, these classifications will be extensively used. They also prove to be
very useful in understanding shell properties and behavior as weIl as being helpful in the
design of varieties of sheIl structures.
2 Design and Analysis of Shell Structures
Shell structures can be efficiently and economically used in various fields of engineering
and architecture. A great variety of shell roofs have been designed and constructed in many
parts of the world. Large spans have been easily covered by reinforced concrete shells.
Water retaining shells have been constructed of steel, concrete and even precast and for
prestressed concrete. Tall silos (up to 60 meters high) and other containment structures of
high efficiency have been built using shell forms. Structurally efficient doubly curved high
arch dams Cup to 300 meters high) have been constructed to resist high intensity
hydrodynamic as wen as earthquake loadings. The containment shells of nuc1ear power
plants, requiring high degree of safety, have always been made of reinforced concrete
shens. TaU chimneys and also huge cooling towers (as high as 200 meters) have been built
of steel or reinforced concrete shens. The set of figures (1-1) show some of these
applications of shell structures .
Recently, with the advent of various fiber-reinforced and laminated composite materials, the
domain of application and range of structural efficiency of shell forms has immensely
increased. Ground, as weIl as space vehic1es, having shell forms, have been designed and
successfully built of high strength temperature resistant composite materials. The skin of
aircraft structures and also ship hulls are composed of shell forms, built of stiffened shells
and I or composite material bodies.
These applications are several important cases of shell technology as applied to various
fields of engineering and architecture. One mayaIso conceive of other applications in
which the structurally efficient, economically promising, and aesthetically appealing shell
forms could be conveniently designed and constructed.
The geometry of a shell is defined by the prescription of its middle surface and its
thickness at all points. Suppose that we have a general surface in space. At any point, A, on
this surface, a plane tangent to the shell can be imagined, figure (1-2). Anormal to the
tangent plane, at this point, would be considered to be the normal to the surface at that
point. Obviously, an infinite number of planes intersecting the surface could be passed
through point A. Some of these intersecting planes contain the normal to the surface at A.
Such planes are normal to the tangent plane at that point and thus could also be considered
to be normal to the shell middle surface.
The plane curves formed by the intersecting planes containing norm als to the surface are
called normal seetions ofthe surface at a point Obviously, every one ofthese curves would
have a local curvature and a corresponding radius of curvature (quantitatively, an inverse of
curvature) at that point. Out of these infinitely many plane curves, formed by intersection,
there will be one curve having a maximum value of curvature (K 1) and another having a
minimum value of curvature (K2).
Plane cont21nlng
normal
/
,
Clane
1/
General
::: 2'"1e Curve
Normalsectlons
The two plane curves, formed by the normal plane sections, are called the principal
sections, and their curvatures, denoted by Kl and K2, are called the principal curvatures of
the surface at A. It can be proved, by differential geometry, that these two intersecting
principal sections are always orthogonal to each other.
Another delineation of surfaces can be may on the bases of their geometrical developablity.
According to this distinction, shell surfaces are either developable or nondevelopable.
Developable surfaces are, by definition, the ones which can be "developed" into a plane
form without cutting andlor stretching their middle surface. A nondevelopable surface, on
the other hand, is a surface which has to be cut and / or stretched in order to be developed
into a planar form.
Introduction to Shells 5
Surfaces with double curvature are usually nondevelopable, whereas surfaces with single
curvature are always developable. Surfaces with positive and negative Gaussian curvature
(i.e., synclastic and anticlastic surfaces) are developable. while those with zero Gaussian
curvature are nondevelopable, figure(1-4).
A third type of classification of surfaces, which is very useful in shell analysis and design,
is the categorization of various surfaces into surfaces of revolution. translational surfaces,
and ruled surfaces.
Surfaces of revolution
Surfaces of revolution are generated by the revolution of a plane curve, called the
meridional curve, about an axis, called the axis of revolution. The axis of revolution, does
not always have to intersect the meridional curve. In the special case of conical surfaces, the
meridional curve consists of a line segment. Examples of surfaces of revolution are shown
in figures (1-5).
axis of revolution
axis of revolution
Surfaces of translation
Surfaces of translation are defined as the surfaces generated by sliding a plane curve along
another plane curve, while keeping the orientation of the sliding curve constant. The second
curve on which the original plane curve slides, is called the generator of the surface. In the
special case in which the generator is a straight line, the resulting translational surface is
called a cylindrical suiface. Examples of translational surfaces are shown in figure (1-6).
,
cj!-
hyperbola
Q}~~:L;:~l (3)
torold
hyperbollc paraboloid
(4)
~~'.
(5~~
Ruled surfaces
Ruled surfaces are obtained by sliding a straight line, two ends of which remain on two
generating curves, in such a fashion that it remains parallel to a prescribed direction or
plane. The generating straight line is not necessarily at right angles to the planes containing
the director curves. Some examples of mIed surfaces are shown in figure (1-7). From a
practical viewpoint, molding of in-situ cast concrete shells, having mIed surface forms, can
be more easily and economically made by the rectilinear forming elements, thus reducing
the forming expenses.
Introduction to Shells 7
hyperbolic paraboloid
Based on our foregoing discussion, we can now summarize and broadly c1assify the shell
surfaces which are commonly used in engineering practice. It may be noted that the same
type of shell may very well appear in more than one category.
2) Doubly curved shells having positive Gaussian curvature (sync1astic shel1s); non-
developable shells
Shells of revolution: Spherical domes. ellipsoids of revolution; paraboloids of
revolution
3) Doubly curved shells with negative Gaussian curvature (antic1astic shel1s ), non-
developable
Shells of revolution: Hyperboloids of revolution of one sheet
Shells of revolution and ruled surfaces: hyperbolic paraboloids; conoids;
hyperboloids of revolution of one sheet
4) Combined shells, partly sync1astic and partly antic1astic shells; shells composed of
simpler shell forms, figure(I-8).
In the course of the chapters that follow, various unique structural features of shells will be
qualitatively and quantitatively introduced and used in design. In this introductory chapter,
only a reference is made to so me of the characteristics of shell structures. The salient
features of shells, as compared with other structural forms such as beams, frames, and
plates can be outlined as follows:
In the forthcoming chapters these aspects of shells behavior will be described in detail.
Introduction to Shells 9
1.1 - M. Farshad, Structural Forms, (in Farsi), Vol. 1,1986, Vol. II, 1987, University of
Shiraz Publications, Shiraz, 1973
1.2 - M. Farshad, Principles of Structures, (in Farsi), Dehkhoda Publishers, Teheran, 1983
1.6 - D. P. Billington, Thin Shell Concrete Structures, McGraw HiIl Book Co.N.Y., 1965,
revised edition 1982
1.7 - G.S. Ramaswamy, Design and construction of Concrete Shell Roofs, Mcgraw-Hill
Book Co.,N.Y.,1968
1.8 - V.S. Kelkar and R.T. Sewell, Fundamentals of the Analysis and Design of Shell
Structures, Prentice-Hall,INC.,N.J., 1987
Preliminaries of Shell Analysis and Design 11
Chapter2
2.1 - Introduction
The behavior of shell structures is, in various aspects, different from that of so-called
"framed structures". This feature originates mainly from the geometrical features of shells
which make the intern al force system in shells differ from those in other types of structural
forms. The internal force distribution in shells is, in general, three dimensional, i.e., spatial.
Moreover, shell structures carry the applied forces mostly by the so-called membrane
forces, whereas other structural forms carry the applied loads by bending mechanisms.
These unique features of shells are also reflected in thcir design as weil as in their method
of construction.
In this chapter, we will first define the internal stress and force system in thin shells. Then,
to achieve an overall understanding, a qualitative discussion of structural behavior of shells
will be presented. Later on in the chapter, theories of shell analysis will be placed in
perspective and the bases of shell analysis methodologies will be presented. Finally, a
number of general shells design considerations will be outlined.
12 Design and Analysis of Shell Structures
The surface passing through the mid thickness of the shell at each point is, by definition,
called the middle-surface of the shell, figure (2-1). If the thickness of the shell is very small
compared with the radii of curvature of the sheH mid-surface, then the sheH is considered as
a thin shell. The thickness to radii ratio, or sometimes the thickness to span ratio, of about
1/200, occurring in reinforced concrete shells, puts the actual shells weH in the range of
being "thin shell" structures. For metallic and composite sheHs, this ratio is in practice
much sm aller, ofthe order of 1/300.
Consider a shell with a general geometry. An infinitesimal element of this shell can be cut
out by intersecting it with two pairs of principal plane seetions which are located at are
lengths ds x and ds y apart. Two intersecting planes, are from each pair the nOlmal to the
shell at the common corner point. The resulting plane curves of interseetion are principal
seetions and are thus perpendicular to each other, figure(2-2). The principal plane curves
have principal radii of curvature which, in this figure, are designated by rx and ry.
fZ
.-1..
x".
••
The internal stresses acting on an infinitesimal element of the shell may be properly
integmted across the shell thickness to give the so-called resultant internal forces. From a
statical point of view, the resultant force system at any section consists of a force and a
moment vector. These internal force vectors can be resolved into components which
altogether constitute the internal force system in the shell. Figure (2-3), shows the
components of resultant internal forces in a shell element.
Referring to figure (2-2), we can readily derive the relations between stress components and
components of internal force resultant in a shell element. The desired relations (2-2) have
been obtained using the requirement of statical equivalency between the internal stresses
and their resultant forces. In these relations, t is the thickness of the shell.
The internal forces at each point of the shell may be placed in one of two groups of force
fields: membrane forces; bending forces. Figure (2-4a) shows the membrane force system
in an infinitesimal shell element. Figures (2-4b) and (2-4c) show the bending force system
in the same shell element. The overall force field in a point of a shell consists of the forces
shown in figures (2-4a), (2-4b), and (2-4c).
14 Design and Analysis of Shell Structures
N =
x r+ t
2
O'x(1 + ~) dz
r
y
N =
y
t~ a (1 +~) dz
y r
X
N =
xy r~ 2
t
2
0'
xy
(1 +~)
r
y
dz N
yx
=
t~ 0'
yx
{l + ~)dz
r
x
(2-1)
t t
+ -- + --
2 2
a (1 + _z_l dz a (1 + _z-ldz
Qx xz r ~= yz r
y x
t t
2 2
+~ t
+ --
2 2
M za x (1 + _Z_) dz M = ZO' (1 + _z-)dz
x r y y r
y x
t
t
2
2
v-J
+ --
t
2
t
2
ZO'
xy
(1 +~)
r
y
dz
f-t
M =
yx
t
?
za
yx
(1 + ...!..)dz
r
x
Preliminaries of Shell Analysis and Design 15
The membrane jorces, as the name implies, are the resultant internal forces which lie
"inside" the mid-surface of the shell. The membrane force field causes the stretching or
contraction of the shell, as a membrane, without producing any bending and I or local
curvature changes. The membrane force field consists of two membrane normal resultant
forces and a membrane shear force.
The second group of internal forces are called the bending jorces, since they cause bending
and twisting of the shell cross-sections. The ben ding force field consists of bending
moments, twisting couples, and transverse shear forces.
~
NM.,
~
~ M.
Mt;. ~
M••
Ng
(a) (b) (c)
Shell structures support applied external forces efficiently by virtue of their geometrical
forms. Shells, having their spatial curvature, are much stronger and stiffer than other
structural forms. For this reason shells are sometimes referred to as form resistant
structures. The strength to weight ratio of a shell structure is usually much smaller than that
of other structural systems having the same span and overall dimensions. A simple
experiment, demonstrated in figure (2-5), shows the structural efficiency of a shell and also
a folded plate as compared with a flat plate of the same matetial and dimensions.
As we have seen before, the load-carrying mechanism, i.e., the intern al forces at any point
of a shell, consist of ten component internal force resultants (Nx, Ny, Nxy, Nyx, Mx, My,
Mxy, Myx, Qx, Qy). These components, can be separated into two groups, entitled
membrane and bending internal force field, as follows :
In this terminology, Mx and My stand for bending moments while M xy and M yx represent
the twisting couples. Qx and Qy represent the out-of-plane shear forces.
For a material body in spatial equilibrium there are six governing equilibrium equations.
Since there are more than six force resultants, we conclude that a shell is, in general, an
internally statically indeterminate structure.
The internal force redundancy, although it is an indication of additional load carrying
mechanisms, is not always required for shell equilibrium. Let us imagine a shell subjected
to applied loading in which only the membrane force field has been produced and the
bending field is absent. By writing the moment equation of equilibrium about the normal to
the she~l elem~nt (z axis) we can conclude that Nl'-Y =Nyx. Therefore, the membrane force
field wIll conslst of the forces Nx, Ny, and Nxy = Nyx.
In a shell in which only the membrane field exists, three of the six equilibrium moment
equations (Mx =0, My = 0, Mz = 0) are identically satisfied. We are then left with three
remaining force eqmlibrium equations and three internal membrane forces to be
determined. Since the number of equilibrium equations and the number of unknown forces
are equal, the membrane shell is statically determinate and its internal force system can be
determined by the use of the equilibrium equations alone, without the need of any auxiliary
relations.
The membrane force field is, of course, associated with the membrane normal and shear
forces, which are assumed to be uniformly distributed through the thickness of the shell. A
shell in which only the membrane force field exists is said to have a membrane behavior.
The resultant theory which is called the membrane theory of shells.
A shell will have a pure membrane behavior provided certain boundary requirements,
loading conditions, and geometrical configurations are satisfied. In order that a membrane
theory be totally applicable, the forces and the displacements at the shell boundaries must
be force-compatible and deformation- compatible with the true membrane behavior of the
shell.
There may be some conditions in which the pure membrane action of a given shell could be
disturbed and thus the premises of a membrane theory would be violated. The most
prominent of these conditions are the following:
(a) Deformation constraints and some boundary conditions which are incompatible with the
requirements of a pure membrane field.
(b) Application of concentrated forces, and change in the shell geometry and I or sud den
change of curvature.
In figure (2-6a), the shell support and its boundary conditions are such that, assuming a
pure membrane field in the shell, all equations of equilibrium are satisfied. On the other
hand, the membrane forces alone can not satisfy the equilibrium at the boundary of figure
Preliminaries of Shell Analysis and Design 17
(2-6b). Also, the fixed boundary condition of figure (2-6c) is incompatible with the
requirements of a pure membrane field.
~
~~
(a) membrane compatible (b) membrane incompatible (c) localized edge effects
For a pure membrane field to exist in a shell, displacement requirements dictated by the
membrane behavior at the shell boundary must also be fulfilled. Figure (2-7) shows two
examples in wh ich the requirements of a pure membrane behavior are not completely
fulfilled. In this case, pure membrane action requires that the domes subjected to applied
loading or temperature variations have free boundary displacements, whereas the actual
support displacement conditions impose some constraints to such freedom of membrane
action and hence disturb the pure membrane field in the shell.
The loading and the shell geometrical conditions must also confOIm to a pure membrane
field. For example, kinks and other discontinuities in shell geometry and also concentrated
loadings as demonstrated in figure (2-8), would disturb the membrane mode of shell
behavior.
18 Design and Analysis of Shell Structures
In the foregoing situations, exemplified by figures (2-6), (2-7), and(2-8), the membrane
behavior of the shens is, to some extent, disturbed. In other words, the membrane field of
forces and deformations will not, by themselves, be sufficient to satisfy all equilibrium
andlor displacement requirements in the regions of equilibrium unconformity, geometrical
incompatibility, loading discontinuity and geometrical nonuniformity. Therefore, the so-
called membrane theory would not hold throughout such shells. In these circumstances,
some (or all) ofthe bending force field components are produced and, by being activated in
those regions, compensate for the shortcomings of the membrane field in the disturbed
zone.
If it is developed in some region of a shell, the bending field usually has a local range of
influence. Laboratory and field experiments, as wen as elaborate theoretical calculations,
show that the bending field produced in any one of the above-mentioned situations would
mostly remain confined to the region in which the membrane conditions are violated. This
feature is shown in figures (2-5) and (2-7) by a bending field that "decays out" from the
source ofmembrane nonconformity.
The bending forces, being confined to a small region, leave the rest of the shell virtually free
of bending actions. Therefore, in most cases, the major part of a shell structure behaves as a
true membrane. This very interesting and unique character of shells is the result of the
inherent curvature in the spatial shell form. It is this salient feature of shells that is
responsible for the most profound and efficient structural performance of shells observed
in nature, as wen as in the shells designed and constructed in engineering practice.
To summarize, shell structures carry the applied extern al forces mostly by the mechanism
of membrane action. In some regions of the shell a bending force field may develop to
satisfy specific equilibrium or deformation requirements. The range of influence of the
bending field is local and is confined to the vicinity of loading and geometrical
discontinuities andlor the deformation incompatibilities. The rest of the shell is virtually
free from bending actions and can be analyzed and designed as a membrane. Depending on
the nature of the applied forces, this membrane shell may be in tension or compression or
partly both. The extent of the domain of influence of bending depends on the particular
shell geometry and its edge and loading conditions.
PreIiminaries of Shell Analysis and Design 19
Many shell theories have been established to analyze the structural shell behavior. The
factors influencing the assumptions and domains of applications of the individual shell
theories have been the material type and behavior; the shell geometry; the loading
conditions; the deformation ranges; the particular shell behavior desired, and the
computational means. Accordingly, there are linear and nonlinear theories, membrane and
bending theories of shells. The nonlinearity can be material and / or geometrical. Shell
analysis computer programs are based on these theories and thus enjoy the capabilities of
their theoretical foundation or suffer from the shortcomings of their theoretical modeling.
Any shell theory,is, as any other theory in continuum mechanics, founded on three set of
relations. These relations are the equilibrium equations, kinematical relations, and
constitutive relations. To be complete, these three sets of field equations must be
accompanied by the appropriate boundary conditions of the particular shell problem.
The membrane theory of shells, based on the assumption of membrane behavior of shells,
is a simple but useful analytical tool for shell analysis and design. Earlier, it was pointed
out in some detail that a shell structure depicts an overall membrane behavior and that the
bending actions may, in most cases, be neglected, or at most be taken into account as
"corrections" to the membrane field. The membrane theory can certainly be used in the
initial designs and analyses. It could also be used as a tool to understand the structural
behavior of a shell without elaborate formulations or numerical computations. Calculations
based on membrane theory, can also serve as guidelines by which the order-of magnitude
of the expected results and I or the computer outputs can be interpreted and checked.
The so-called classical theories of shells are based on the following assumptions:
1) The shell is assumed to be thin, Le., its thickness is small compared with its
representative minimum radius of curvature, or lateral dimensions.
2) Plane sections originally normal to the shell mid-surface remain plane and perpendicular
to the deformed mid-surface. The lauer assumption is equivalent to ignoring the shear
deformations.
3) The stress component normal to the shell mid-surface is very small compared with other
stress components, and can be neglected.
4) The displacements and strains are so small that their higher powers can be neglected.
These assumptions are valid for most engineering shell structures. Therefore, the classical
theories based on these assumptions can be used for analysis and design of a variety of
shell structures, including reinforced concrete and metallic shells. For composite shells,
refinements in these assumptions are sometimes necessary.
20 Design and Analysis of Shell Structures
To perform a shell structural analysis, one of the two general well known methods of
structural analysis may be used. These are: the force method (or compatibility method);
the displacement method (or the stiffness method). Many Finite Element shell computer
programs are based on the stiffness method. However, for manual calculations, the
compatibility method offers certain advantages over the stiffness method.
The analysis of a given shell, according to the force method exemplified by figure (2-9a),
includes the following stages:
(1) First, a membrane analysis of the given shell subjected to applied distributed external
loading, figure(2-9b), is carried out. In this stage, the boundary conditions are assumed to
be compatible with the requirements of the membrane action of the shell. The internal
forces as well as the edge dis placements and rotations are to be determined from this
membrane analysis. At this stage, the shell is statically determinate.
(2) The unknown corrective redundant bending forces are applied to the shell from which
the distributed loading is now removed, figure(2-9c). The shell is analyzed with the help of
an appropriate bending theory. The internal forces as weH as the edge displacements and
rotations are obtained. This stage of analysis yields the corrections, i.e., the redundant
forces, due to the bending field which exist at the shell boundaries. The internal forces and
edge deformations of the shell are obviously expressed in terms of unknown bending
forces.
(3) The results of analyses performed in stages(l) and (2) are combined to satisfy the
compatibility requirements at the shell boundaries. The compatibility requirements,
expressed in terms of known membrane displacements and unknown edge forces, yield a
set of simultaneous algebraic equations from which the redundant edge forces can be
determined.
(4) Having performed the membrane analysis and having obtained the corrective boundary
bending effects, one can now superimpose these two fields to determine the complete
force and deformation field in the shell. This completes the force method analysis of the
shell.
The force method of analysis can also be applied to the analysis of a shell structure
composed of several shell segments. It is also applicable in the analysis of the shells having
flexural edge members. It can be automatized as computer software.
(1) Choice of a shell geometry which meets the overall architectural requirements.
(2) An trial choice of the dimensions of the shell and its supporting members, based on
previous experience and / or observation of similar shell structures.
(3) A preliminary design and analysis of the shell. For this purpose we can use the
membrane theory as well as other simplified shell analysis schemes, such as, the beam
theory of cylindrical shells.
In this stage we obtain the initial dimensions of the shell system including the shell
thickness, the pattern of shell reinforcements, and the tentative values of required materials
and reinforcements. These results are useful for a more comprehensive analysis and also
for the initial material and cost estimates.
The preliminary analysis and design of a shell also provides an insight into the general
behavior of that shell. Therefore, having performed an initial design-analysis, the designer
would have a free hand in changing its design or refining it without having to go into
extensive calculations and / or unwanted expenses.
(4) A more comprehensive analysis and detailed design of the shell using bending
theory and / or the available computer programs. In many cases, this stage is only required
for a final check of the design. A logically based preliminary design usually remains valid,
except for some local modifications or perturbations.
Determination of the form, dimensions, and reinforcements of some shells can also be
obtained through experiments on physical models of the shell. This is particularly so for
the complicated shell geometries and / or cases in which there are no reliable analytical
tools. Many arch dames and shell roofs have been designed by the help of experiments
performed on scaled models of the actual structure. The experimental and theoretical tools
can also complement one another to achieve a sound engineering design of the shell
structures.
(5) The shell designer should always bear in mind that the shell body, in a shell structure, is
only an element of the whole structural system. The shell elements, in a shell structure, are
usually accompanied by the strengthening members such as stiffeners, edge and / or ridge
beams, and end diaphragms. The structural unity of these various elements, specially under
dynamic Ioadings, is of prime importance. Therefore, design - construction precautions
must be taken to assure the illtegrity of the structure as a whoie.
An integrated structural shell design, as any other sound structural design, is one which
takes into consideration a compiete pattern of force flow, figure (2-10). Also, a good design
should aiways involve considerations reiated to construction, shell manufacturing
processes, and shell aesthetics.
22 Design and Analysis of Shell Structures
(a)
-- ~ = - ~-:..-~--=- ...~
(c)
Metallic, composite and concrete shells are truly thin-walled structures. For example, the
thickness to span ratio of reinforced concrete roofs usually falls below 1/200. This ratio is
still sm aller for metallic and composite shells. On the other hand, there exist situations in
which, due to applied loading and / or thermal effects compressive internal stresses may be
produced in the shell. These effects could make the shell elastically unstable and thus cause
buckling failure of the shell structure.
The buckling strength of shells can be several order of magnitudes lower than the material
strength of the shell. Furthermore, thin shells have been shown to be among the so-called
imperfection-sensitive structures. This means that the buckling load of a shell is highly
dependent on imperfections in shell geometry and loading. As a consequence, the true
buckling load of an actual shell could be several times sm aller than the buckling strength of
a corresponding ideal perfect shell.
Premature failure of shells, due to the buckling, makes the stability design of shells
important. In fact, dimensioning of shell thickness is usually based on buckling
considerations rather than material strength criteria. For in-situ reinforced concrete shells,
constructional processes also playamajor role in dimensioning the shell thicknesses.
Preliminaries of Shell Analysis and Design 23
Chapter 13 of this book, will treat the buckling instability of shells. That chapter will
present some simplified and practical formulas for the design of shells. These formulas
may be used in the actual design of shell structures.
The practical design of shell structures are based on Codes of Practice. Codes of practice,
offer general design guidelines specifying limiting design parameters, such as the minimum
reinforcement, maximum stress and strain, and minimum thickness.
There are many codes and standards: American Concrete Institute (ACI) codes on shells;
German Norm (DIN), the British Standards (BS); and the Indian Standards (IS). These
codes help the designer to carry-out a practical design of shells. These standards can be
used as guidelines and as official criteria in the design process. In this text, we will use
some of these standards in working out the detailed design examples.
24 Design and Analysis of Shell Structures
2.1 - M. Farshad, Shell Structures, (in Farsi) Shiraz University Press, Vol. I, 1986, Vol. 11,
1987, Shiraz
2.5 - E. H. Baker, L. Kovalevsky, and F.L. Rish, Structural Analysis 0/ Shells, McGraw-
Hili Book Co.,N.Y.,1972
Membrane Behavior of Cylindrical Shells 25
Chapter3
3.1 - Introduction
Cylindrical shell fOnDS are used in water and gas retaining structures, circular silos, pipes,
pressure vessels, and cylindrical vaulted shell roofs of various kinds. This chapter studies
the membrane behavior of cylindrical shells. First, we derive the governing membrane
equations. Then, we apply these equations to the analysis of various types of cylindrical
shells, including vessels, pipes, and vaults. The results of the present chapter will also be
used in more comprehensive analysis and design of cylindrical shells treated in future
chapters. Some basic design considerations related to such shells will also be presented.
26 Design and Analysis of Shell Structures
A cylindrical surface is generated by sliding aplane generating curve (the plane of which
remains parallel to a fixed plane) along a straight line called the directrix of the
cylinder.The generating curve of a cylindrical shell may be a c10sed profile (pipes) or open
(vaults). In the particular case of a circular (i.e., constant curvature) profile, the resulting
shell is called a circular cylindrical shell.
The mid-surface of a cylindrical shell is a surface with single curvature. The curvature of a
generating curve may vary along the profile curve. Any point on the mid-surface of a
cylindrical shell can be identified by two coordinate parameters. These two parameters are
the longitudinal distance from some arbitrarily chosen origin, x, and the angle <p between the
normal to the shell at that point and a reference normal at some chosen origin, figure (3-1).
Consider an element of a cylindrical shell, such as the one shown in figure(3-1). Let (x, <p)
designate the coordinates of a corner point of this element. The free body diagram of this
element is shown in figure (3-2). In this figure, the r direction identifies the normal to the
shell surface at the point (x,<p). The symbols Nx, Ncp. Ncpx, and Nxcp denote the unknown
membrane forces in the shell; these have units of force per length. The known quantities
Px, P" and Pr, having units of force per area, represent the intensity of the applied
distributed forces in the x, <p, and r directions, respectively. All these parameters are
functions of the position coordinates x and <p.
Membrane Behavior oE Cylindrical Shells 27
7fx = 0
~
aNx
()x dx. o
Equilibrium along the tangent to the shell, the <I> direction, yields:
H<j> = 0
3N4> dN 4>
~ d<j> • dx + ax
X
dx. rd.p + P<D • dx. rd~ = 0
Finally, by writing the equilibrium relation in the r-direction, i.e., along the normal to the
shell, we obtain:
Lf = 0
r
o
Dividing both sides of these equations by the arbitrarily small, but nonzero, length
parameters ds and r d<l>, we finally obtain the following membrane equations of equilibrium
for cylindrical shells.
28 Design and Analysis of Shell Structures
N
q, = Pr r (5-1)
3Nxep 1 dN~
3):(= - P<j> ---- (5-2)
r "'1>
dN
x
3x (5-3)
These equations can be integrated in a sequential manner to yield the membrane force field
functions Nq,. Nq,x, and Nx. The general integral expressions are,
N = rP (5-4a)
4> r
aN
N
xep
=- J (P4> +.l..r _.~
34>
)dx + fl (ep) (5-4b)
N
x
= - J(P x + 1 dNXt!l
--~-)dx + f2(ep)
r oGl (5-4c)
The functions f 1 (<I» and f2 (<I», resulting from integration, depend on the boundary
conditions of the particular shell. They can be uniquely obtained for each individual shell
problem. In the following seetions, we will apply these expressions to some shell analysis
problems.
Pep = PSincp
Hence, using the expressions (3-4), and using the condition of Nx = 0 at the ends we find
the following membrane field:
N.(> =- Pa Cost>
N
c;,x =- 2P x Sin:;. (3-5)
N
x
=- ~ {~2
4a
- 4x 2 )Cos4>
Figure (3-3) shows the variations of internal membrane forces in an arbitrary section,
located at a distance x.
Membrane Behavior of Cylindrical Shells 29
Referring to these graphs, and expressions, we can make the following observations:
(a) At the longitudinal edges of the shell, i.e., at <11= ± 1t /2, we have: Ncp=O. This means that
this shell has no need of longitudinal normal support.
(b) At the longitudinal edges (<11= ±1t /2), the membrane shear force in the shell is Nxcp =
2px. This means that there is a need for a longitudinal edge member to absorb this force
and to transfer it to the end diaphragms. This edge member is shown in figure (3-4).
The longitudinal edge member required by the membrane theory is called an edge beam.
Figure (3-3) shows the mechanism of shear load transfer from the shell to the edge beam.
The shear force in the shell, upon its transfer to the edge beam, becomes an axial tension in
the beam. The magnitude ofaxial tension force in the edge beam at any section is obtained
by integration of edge shears along the edge. 1ts expression is:
H =J~.\!, N
xlji
dx =- 2PJX xdx
R.
= !4 p(.\!,2 - 4x 2 ) (3-6)
"2 2"
At the ends of the shell,Le., at x=+ I /2, we have Nx=O, but NX<jl= -pI sin cp. This means that
part of the applied load is transferred to the end diaphragm by the shear force mechanism.
30 Design and Analysis of Shell Structures
Px = 0 , P~ = 0 P = y(H - x) (3-7)
r
N = 0 (3-8a)
x
xcp = 0
N (3-8b)
,·t[- ., . F,~\-
tI
+
L@ __
I - _ l _ _'_x-.-, I
L[W--
a~ N.
L 1 I ~
The pressure head, measured at the level of cylinder axis, is assumed to be Po. Thus the
components of applied loading are
(3-lOa)
Membrane Behavior of Cylindrical Shells 31
'(2
X2
Cos~ - ax df 1 (CP)
d<p + f 2 (CP)
The unknown functions f 1(</» and f 2 (</» are to be detennined from the boundary conditions:
(N )
x x=0
= 0 (N)
x x="
n =0
The constant C, representing the non-existing resultant end torsional couple may be set
equal to zero. Therefore, the solution to this problem is
N = P ä - ya 2 Cos<p (3-11a)
<p 0
Nx = - ~ x (~ - x)Cos<P
(3-11c)
If this cylinder were viewed as a horizontal beam, then the derived expressions for Nx and
NXq> could be interpreted as the longitudinal bending stress and shear stress, respectively.
Figure (3-5) shows that Nx is linear in the sectional height.
In order to obtain the membrane displacement field in a cylindrical shell, with already
known internal forces, we use the constitutive and the shell kinematic relations. We assume
that the shell is made of a linearly elastic and isotropie material which obeys Hooke's law;
the two elastic constants are Young's modulus E and Poisson's ratio v.
32 Design and Analysis of Shell Structures
For a linearly elastie and isotropie cylindrieal shell of thiekness t the constitutive relations
are:
EX ~t lN x - vN;j»
E\jJ it-C
1
N<j> - vN) (3-12)
20+'J) N
YXtj! = Et x,p
To derive the kinematie relations we eonsider a displaeed configuration of the shell element,
as shown in figure (3-6b). If we denote the axial strain by ex, the transverse strain by E</>,
and the shear strain by Y</>x, then we can write the linear strain-displacement relations as:
dU
EX dX
1 dV
E\jJ = r(d\jJ + w) (3-13)
_ dV + 1 dU
Yx\jJ - dX r ä(j)
dU = ..!-_ (N - \JN )
3x rt x ~
dV 1 dU 2 (1 + v) '1 (3-14)
dX + r a~ = ~-t- "x'~
.:::!.... + ~ 3v = l... (Ci - \l}; )
r r dtP Et 4> x
Assuming that the membrane forces Nx, N</>, Nx</> are already detennined, we Can integrate
the relations (3-14) in a sequential manner, to obtain
Membrane Behavior of Cylindrical Shells 33
The functions f3 ($) and f4 ($) in the integration process can be determined for each shell
from specific boundary conditions.
Return to the circu1ar cy1indrica1 shell of figure (3-4). We wou1d 1ike to determine the
displacement fie1d in the shell under a dead weight 10ading of intensity p. As specific
boundary conditions, we assurne that at x=+ 1/2 we have v=Ü and w=Ü.
If we substitute the membrane force expressions (3-5) into (3-15) we obtain the following:
(3-16)
Note that the constraint on the radial displacement, w, at x=+l /2 vio1ates the requirements
of the membrane action of the shell at that section. As a resu1t, some ben ding will develop
around that region, and can be predicted by the bending theory of cylindrical shells.
Solution
Evaluating the expressions (3-16), for the shell displacements, and substituting the assumed
nurnerical values in the resulting expressions, we obtain
34 Design and Analysis of Shell Structures
Uo = Ik (0) = 0
vo = 0
=0.06 cm = O.6mm
As we observe, the resulting displacement is very small.It would be interesting to compare
this value with a relatively high lateral deflection of a beam having the same span and
loading.
Whether the cylindrical shell has mostly the beam action or the arch action depends on the
shell geometry and the edge conditions. For example, longitudinal supports transfer the
applied forces in the longitudinal direction.
neutral plane
Long cylindrical shells resting on end supports act like simply supported beams. A long
shell acts like a beam, having as its section the shell profile, and as its span the span of the
shell. The extern al bending moment at each section is supported by the resultant of
compressive longitudinal membrane forces acting as the compression flange. and the
tension in the edge beam acting as the tensionflange, figure (3-7).
Membrane Behavior of Cylindrical Shells 35
As we have seen in our membrane analysis of long vaults, the applied forces are transferred
longitudinally to the end supports. The transfer of internaiload to the end diaphragms or
end arches takes place by the internal shear force mechanism, as shown in figure (3-8).
I
i
I
.I
~
Figure (3-8) Shear force mechanism of applied fOI·ces to the end
arches in a cy lindrical vault
In an overall beam action of the long shell, part of the applied load could be transferred by
internal shear forces to the longitudinal beams, as demonstrated in figure (3-9). However, if
there are no edge beams, then because the free edge situation is incompatible with the
requirements of the membrane theory, some bending field of forces will develop along the
longitudinal edges
the shell
The Ion ger the shell is, the stronger is the beam action of the shell. On the other hand, in a
short shell, the so-called arch action is more effective and the applied load is mostly
transferred by the trans verse arches.
Figure (3-10) shows the longitudinal internal stresses in two end suppOited cylinders, one
long and the other short. As we see, the behavior of the longer shell is c10ser to the action
of a simple beam, whereas that of the shorter shell is different.
36 Design and Analysis of Shell Structures
p p
tens\J~Jll
Il@
tension tension
So far, in our qualitative presentation of cylindrical shells behavior, we have discussed the
membrane action of the shell. As we pointed out earlier, in chapter 2, some bending fields
can develop in shells. For example, end supports (diaphragms or arches) give rise to
internal bending forces. In these cases the ben ding force field is needed to satisfy
compatibility. The region of influence of the bending field depend on the end conditions,
the loading, the thickness, and the length of the cylindrical shell.
Figure (3-11) shows how the behavior of long and short shells are influenced by the
bending field. The effect of the ben ding field, caused by the end diaphragms, is local and
confmed to the ends of a long shell. In shorter shells, the influence of bending field is more
pronounced. For very short shells, the bending field penetrates throughout the shell and
affects the "whoie system".
Figures (3-12) summarize the cylindrical shell behavioral patterns. These figures
demonstrate c1early the beam action, figure(3-12a) and (3-12b); the arch action, figure (3-
12c); the bending fields developed by the stiffening longitudinal beam, figure (3-12a), and
end arches, figure (3-12c).
Membrane Behavior of Cylindrical Shells 37
The qualitative understanding of the cylindrical shell behavior, arrived at in this seetion, is
extremely useful in cylindrical shell design and analysis. The quantitative analysis of
cylindrical shells for bending effects will be carried out in the next chapter.
38 Design and Analysis of Shell Structures
Problems
P 3.1 - Consider a cylindrical vessel with elliptical profile as shown in figure (P 3.1). The
vessel is in a horizontal position and is simply supported at its two ends by means of end
rings or end diaphragms. Determine the internal forces in this shell for Px =0 and arbitrary
distributions of pr and Pr. Show that the resulting force field has the following form:
1 ClNcp
Nx ,.
'f'
=- x(P,.
'f'
+ -_)
a Clcp
N = - ...!. 4X2)~ (P ClNp
x 8a
(R,2 _
dcp cP + a1 Clcp )
Figure (P 3-1)
P 3.2 - Consider a cylindrical vault having a parabolic profile as shown in figure (P 3-2).
The equation of the profile curve is
r;;2 = 2r E;.
o
Figure (P 3-2)
Membrane Behavior of Cylindrical Shells 39
wherein, ro is the radius of curvature of the parabola at the origin. At any other point, the
radius of curvature can be expressed as
r = [1 + (d~/ds)2]3/2
(d2~)/(dsL)
and since
so rO
tanljJ = ds r = ---,,-
d~ Sin 3 Q
P = 0
x
r Px = 0
I PA = - P SintCbsljJ
<;J
1 Pr = - P Sin 2 1jJ
(3) Lateral wind load of intensity p:
Show that the internal force field for these loading systems are as folIows:
Px
Nx - 2r (9. - x)Sin~.j;
o
40 Design and Analysis of Shell Structures
r
o
Nq, p SinljJ
N = 0
x
Interpret these results and note that for snow loading, parabolic arches are funicular and
hence the load is carried mainly in the transverse direction. By plotting the membrane force
field for dead loading, reproduce the following figure:
T -
"
N,p., • =l
,
,-1 .....
"
Note that the effect of shell on the end diaphragm is in apparent contradiction to intuition.
Verify this result and carry out a physical interpretation with the help of the following
diagrams.
Membrane Behavior of Cylindrical Shells 41
3.4 - A vertical chimney of circular cross section and radius a is subjected to a wind load
defined by Px = 0, P~ = 0, Pr = q Cos 8. Here 8 is the horizontal angle and q is the
intensity. of wind load. Find the membrane stress resultants in this chimney shell. Compare
the results with those obtained by classical cantilever beam theory.
3.5 - Consider a simply supported shell vault with a catenary cross section. The equation of
the middle surface in the cross-sectional coordinate system y-z is given by,
z = - a ( 1 - 85 h y / a)
Obtain the membrane stress resultants in the shell under uniformly distributed dead loading
of intensity p.
42 Design and Analysis of Shell Structures
3.2 - A. M. Haas, Design o/Thin Concrete Shells, John Wiley & Sons, New York, 1962
3.4 - A. Pflüger, Elementary Statics 0/ Shells, 2nd Edition, McGraw-Hill Book Co., N.Y.,
1961
3.5 - M. Salvadori and R. Heller, Structure in Architecture, Prentice-Hall, INe., N.J., 1963
Bending Analysis of Circular Cylindrical Shells 43
Chapter 4
4.1 - Introduction
A complete analysis of various cylindrical shell forms, for silos, press ure vessels,
containment shells, containers, and shell roofs, would require an appropriate bending
theory. A general bending theory would embody the membrane theory of shells. It would
also predict the ben ding action of the shell. This chapter introduces a general bending
theory of circular cylindrical shells. The theoretical basis of the present chapter forms the
foundation of approximate bending theories for cylindrical shells. The special theories for
cylinders, such as axisymmetric cylindrical shell theory and the bending theories of
cylindrical vaulted roofs, developed in future chapters, can be considered as the offspring of
the general theory presented in this chapter. The theoretical developments in this chapter
will lead to a set of useful relations for treating a variety of practical shell analyses and
design problems.
44 Design and Analysis of Shell Structures
a () () .
~=
M ••
The equations of equilibrium of internal and external forces, along the x, <1>, and r axes, are
as follows:
u'X + NA,
'+'x
+ aP
X
= 0
(4-1a)
N + N'
<jl x<jl
- Q + aP
<jl <jl
=0
(4-1b)
Q<jl + ~ + N.p - aPr = 0 (4-1c)
Bending Analysis of Circular Cylindrical Shells 45
Compared with the membrane equations for cylindrical shells, these equations contain extra
terms emanating from the bending effects. When the bending field comes into play they are
needed for providing the equilibrium and / or compatibility to the shell elements.
M' + M
x lIx
- aO
-X
=0
(4-2b)
aN -aN +M =0
x<j> ~x q,x
(4-2c)
Note that, due to the presence of the twisting couple M<jlx. the membrane shear forces NX<jl
and N<jlx need no longer be equal.
If we eliminate the two shear force functions ox and 0<j> from the two sets of equations (4-
1) and (4-2), we obtain the following equilibrium equations:
(4-3a)
N~ + Nq,x + aPx = 0
aN
~
+ aN'
x(j,
- t-l
~
- M'
x~
+ a2p
<P
=0 (4-3b)
M + M"
ep xep + M"+x + ll"x + aN
ep - a 2p
r
= 0 (4-3c)
o
(4-3d)
There are ten unknown internal force components (N x • N<jl, NX<jl. N<jlx. Mx, M<!>. M<jlx. Mx<!>. Ox,
O<!», on the one hand, and six equilibrium equations, on the other. Therefore, the shell is
statically indeterminate and hence requires extra relations for its analysis.
To derive the kinematical relations for circular cylinders, we use the basic kinematic
assumptions outlined in section 2.5. If we denote the components of displacement field of
the shell middle-plane by u,v,w, then we can express the displacement of an arbitrary point
located at a distance z from the middle-surface in terms of these components. An outward
radial displacement, w, is considered positive.
Figures (4-2a) and (4-2b) show the undeformed and the deformed longitudinal and
transverse sections of the shell element. In these figures, points Ao and A represent typical
material points originally located on the shell middle-surface, and a distance z from the
middle-surface, respectively. The components of displacement of the material point Aare
denoted by UA, v A, W A.
46 Design and Analysis of Shell Structures
~I-u" -l I undeformed
a I--u-l
t (a) longitudinal section (b) transverse section
2
Using the basic kinematic assumptions of section 2.5, we can express the displacement
components of an arbitrary material point located at a distance z in tenns of the middle-
surface displacement components as folIows:
u A= U - -a
w'
:z (4-4)
a + z w
l.=-a- v --a z (4-5)
w*
A
=w (4-6)
Having found the displacement relationships, we can now fonnulate the strain-displacement
relations. Tbe components of strain in cylindrical coordinates can be expressed as
(4-7a)
v* + w*
A A
a + z (4-7b)
Using relations (4-4), (4-5), and (4-6) we ean write these as finally:
w" (4-8a)
E = ~- z~
x a
z Iv W
E p =~-
a
---+--
aa+z a+z
(4-8b)
Yx4J - u
- +a -+a-2zv ,
a + z
\V'
-(
a
Z
a
+ Z
a + z
) (4-8e)
These relations eonstitute the kinematie relations for eireular eylindrieal shells.
Throughout this book, we shall assume that the shells under eonsideration are eomposed of
isotropie materials having linearly elastie behavior. Therefore, assuming a generalized plane
state of stress, we may write the stress-strain relations, at any point of the shell as folIows:
(4-9a)
I; = -E- (E + VE,j)
K l-v 2 X 'T'
\j = _E_ (E + VE ) (4-9b)
,~ 1 - V2 4> x
G" _ E
KejI - 2 (1 + v) Yxq, (4-ge)
We can express these eonstitutive relations in terms of internal force resultants and the
displaeement eomponents of the middle surfaee. In order to do so, we use the set of
relations (2-1), (4-8), and (4-9). The eombined eonstitutive and kinematic relations are:
K
N
4>
=.-2.
a
(v + w + V u') + - 3 (Iv + w)
a (4-lOa)
N
x
= ~(u' + ,v + VVl) - -a~- w"
:1 (4-lOb)
(4-lOe)
(4-lOe)
K
MijJx = az (1 - v) (w'· + ~ - ~')
48 Design and Analysis of Shell Structures
N
q,x
= J2...
a
1 - v (u· +
2
V ') + K 1 - v (.
aT -2- u - w ' ·) (4-lOg)
Kxq, = aJ 1 -v
-2- (u + v') + aK1-,)
3- - 2 - (v' - w ' ·) (4-lOh)
K = [t 3
(4-11)
12(1 - v 2 )
From a physical point of view, the parameter K signifies the bending stiffness and D the
membrane stiffness of the shell. Therefore, all terms containing K represent contributions
of the bending field to the total field. So, if we assume no ben ding stiffness for the shell,
i.e., if we disregard the terms containing K, in the above relations, we shall obtain the
constitutive relations for a membrane elastic cylindrical shell.
The shell constitutive relations (4-10) can be cast in a more appealing and yet compact
form. In looking back to relations (4-8) we recognize the middle-surface strain field, as
defined by Ex, Ecp, and 'Yxcp to be related to middle-surface displacement components as
follows:
v+w (4-12)
E:<p = - a -
Also, the terms containing thickness variable, z, are recognized to be contributions from the
sectional curvature change and twist. Hence, using relations (4-8) we can write the change
in the shelliocal curvature as weIl as the local twist in terms of parameters 'l'x. 'l'cp. and 'l'xcp
as:
w" (4-13a)
= a2
Now, if we insert the expressions (4-12) and (4-13) in relations (4-10), we obtain the
following constitutive relations:
(4-14a)
(4-14b)
N
x
= D (EX + vE..,)
'I'
- JS....~,
a x
(4-14c)
N = D(l - v) - K(l - v) (1jJ + "Y'x<P )
~x 2 Yx<p + 2a x<p 2a
(4-14e)
(4-14f)
(4-14g)
M = K(l - v) 1jJ
cjJx xq,
- (4-14h)
yx~
Mx</> = K(l - v) (iJix<P - 2a )
These relations, are a set of fairly complete but useful forms of shell constitutive relations
expressed in terms of internal force resultants. They also embody the kinematic relations.
In these expressions, the terms containing K signify the contributions arising from the
bending action of the shell.
50 Design and Anal ysis of Shell Structures
The eoIlection of equilibrium, kinematie, and eonstitutive equations, (relations (4-3) and (4-
10», eonstitute the field equations of bending theory of cireular eylindrical shells. In
passing, we note that by using the expressions for Nxcp , Ncpx, and Mcpx we can satisfy the last
equation of (4-3) identieally. Hence, the total of 11 equations eorrespond weIl with the total
of 11 unknowns (internal forees and displacements).
By eombining the relations (4-3) and (4-10), we can obtain the displacement equations
governing the bending behavior of circular eylindrieal shells. So, by substituting (4-10) into
(4-3), and making some elementary manipulations, we obtain
11 1 - v .• 1 + \J ,'. , k [1 - v .. '"
u +-2-u+-2-v "".1W + --2-u-w (4-15a)
a2
"'J+ -0---
P
+I
--2v- w x - 0
1 + \J .
-2- u '+v + 1 ; \ \." _ ,.:, + k [ ; (l - \1)\1"
(4-15b)
P~a~
3 - \. ".1
- 2 - \.' - --- 0
:J
The parameter k in these equations represents the ratio of bending to membrane stiffness. It
has the following definition:
(4-16)
The displacement equations (4-15) governing the ben ding behavior of eircular cylindrieal
shells are quite general. These equations ean be solved by numerical sehern es or, in some
special cases, by analytieal means. However, eertain simplifieations ean be made 10 make
them more useful for praetical analysis and design. If we assurne that the shell is relatively
thin, then we may only retain those terms containing the parameter k which are of higher
gradients, i.e., higher order of differentiation with respeet to the x and <l> variables. If we do
so, we obtain the more simplified equations which read:
Bending Analysis of Circular Cylindrical Shells 51
P a2 (4-17a)
u" +l--2
v" l+v "
- u +-2- v + \JW' +~
0
0
P a2 (4-17b)
1 ... v " - v v " +W
+v +1- 2 + 4>
0
- 2 - -u - ])
P a2 (4-17c)
\u· + v+ W+ r
k (w'''' + 2w"" + w) - - 0 - = 0
These equations can be simplified further. We shall discuss special theories based on these
equations in the forthcoming sections and chapters.
The equilibrium equations for the axisymmetric theory of cylinders are thus:
fi' + aP = 0
x x
(4-18a)
:-J + 0' - aP = 0
'.j> -x r (4-18b)
1'1' - aO = 0 (4-18c)
x -x
52 Design and Analysis of Shell Structures
Tbe last two equations, representing the axisymmetric bending (now decoupled from the
axial deformation) may be combined to yield the single equation:
aN
I, + M"x - a 2 Pr = 0
1 1 d"!-1 x 1 (4-19)
~=> Pr = -. M"
a":< + -N.
a . , = dx 2 + a N'"
'P
(4-20a)
N<jJ = a1 D (w + \IU')
(4-20b)
N
x
= 1..-
a
D(u' + \'w)
(4·20c)
1 (4-20d)
K(w + vw")
a2
(4-20e)
M
x
= 1_
a2
K(w" - u')
(4-20t)
M = M = 0
4Jx x4J
In order to derive a single differential equation on the radial displacement, we decouple the
axial effects from radial bending by assuming that Px = O. By assuming a constant
thickness shell, we obtain
(4-21)
N
<p
= - a1 Dw(l' v )
2
(4-22)
M
Ij>
= vr1
x
M
a
x = 12 Kw"
Bending Analysis of Circular Cylindrical Shells 53
(4-23)
(4-24)
This fourth order differential equation resembles the equation of a beam resting on an
elastic foundation and laterally loaded by the distributed force Pr. From a physical
viewpoint, the equivalent elastic foundation effect comes from the inclined hoop force, Ncp.
acting on a longitudinal" strip" of the shell in its radial beam-type behavior.
In this chapter, as in other chapters, we have treated static problems for shells. To deal with
dynamic problems we must include the inertia effects in the equilibrium equations. The
theory governing the axisymmetric vibrations of circular cylindrical shells is quite simple.
In fact, all we have to do is to add a lateral inertia term to the right hand side of the equation
(4-24).
In this expression, the function f(x) represents a particular solution to (4-24). To obtain the
general solution to nonhomogeneous equation this particular solution is to be
superimposed on the general solution to the corresponding homogeneous equation. Being a
fourth order homogeneous differential equation, it would contain four integration constants
Cl. C2. C3. <4.
The general solution to the homogeneous equation (i.e., the differential equation without
the right hand side ) gives the correction to the particular, i.e., membrane solution so that the
prescribed boundary conditions can be satisfied. The integration constants Cl. C2. C3. C4
are determined by the four boundary conditions (two at each edge) prescribed for each
case.
Consider the circular cylinder of figure (4-3) subjected to a distributed edge moment Mo
and a distributed edge shear force Qo.
Mo
f----,--,.
Since the shell of figure (4-3) is acted upon by the edge forces alone, the particular solution
to this problem is identically zero, i.e., f (x) = O. The edge effects should decay out as we
move further away from the loaded edge. So that Cl =C2 =O. The general solution to this
problem is therefore
(4-26)
Two remaining constants of integration can be determined from the prescribed force
boundary conditions:
(4-27)
By substituting the expression (4-26) into these end conditions and solving two resulting
simultaneous equations for C3 and C4, we obtain
-Mo (4-28)
C4 = 2ß2K
This gives the general solution, (4-26), for this problem as follows:
(4-29)
Bending Analysis of Circular Cylindrical Shells 55
This is the expression for the radial displacement of a circular cylinder under edge loading.
The resulting hoop force and bending moment at any seetion can be determined from the
relations (4-22). As we expect, all these field functions decay character with distance from
the loaded edge.
Q.
Before entering into the analysis of this problem, we offer a physical observation on the
expected behavior of this shell. If the ends of this cylinder were free from all restrietions,
the shell under interna! pressure would act as a membrane. The membrane hoop stress
obtained from the membrane theory of cylindrica! shells would be
(4-30)
(4-31)
However, the imposed edge constraints inhibit the free membrane expansion of the shell.
To compensate for the incompatibility of membrane action, a bending field develops at the
edges and influences the shell behavior.
We assume that the cylinder is long enough so that the influence of the constraint at one
end would not be feit at the other. Therefore, the edge effect solution, (4-26), can be applied
to each end region. If we did not assume this, then for Sh0l1 cylinders we would have to
obtain all four constants Cl, C2, C3, C4 simultaneously.
56 Design and Analysis of Shell Structures
The end conditions, being the same for both ends in this problem, are
o o (4-32)
Substituting the sum of the fundamental edge effect solution (4-29) and the membrane
solution (4-31) into these relations, we obtain
MO -- 2S 2 1(D -- 262
P (4-33)
,. 0
Having found the quantities Mo and Qo, we can easily write down the expressions for the
radial displacement, the bending moment, and the shear force at each section of the shell.
p = Y (H - xl (4-34)
r
d~w ~ Y (H - xl (4-35)
dx 4 + 4ß w = y
From a physical point of view, this solution is the membrane displacement fjeld of the same
container having unconstrained edges. The general solution of the governing differential
equation, with edge constraints, is
Px -Px
W = e~ (C 1cosßx + C2Sinßx) + e ~ (C 3 Cosßx
(4-37)
+ C4 Sinßx) + Y (! IEt
- x) a 2
If we assume the container to be high enough so that the effect of the constraint at the base
would not be feIt at the top, then we may set Cl = C2 = 0; then
(4-38)
The remaining integration constants can be obtained through the use of the boundary
conditions at x =0, which are the same as (4-32); we obtain
Y 2 1
C = - aCH - - )
4 Et :3
ya" ( -ßx [ 1
w = Et 1H - x - e IICosßx + (H - 13) Sin6x ]} (4-39)
Having determined the radial displacement function, w (x), we can also write down the
expressions for the hoop stress and the bending moment, at each section of the cylinder, as
wall as the value of constraining forces at the base. These expressions are
-Px 1 -;lX ]
;~, ya [ H - x - He I-' Cos6x + (-6 - H)e ~ Sinsx
<,:
(4-40a)
Yat [1 _P -IOx ]
M ( - - H) e ,0 x Cosßx + He ~ Sinßx
x 112(1 _ v") ß
Solution:
First, we calculate the value of the parameter 13:
e= [ 3~;t~ V
2
)J!.o = 1.59 l/m
Q:2.7 Sm
r-:~~-:--~
I Q:2.75 m
l
~c::_ m
t : 2S c m I H:3.65
~J
Figure (4-6) The cylindrical water tank of ex am pie 4.1
Referring to relations (4-40), we find the hoop force and the bending moment
N
~
= :750 [3.65 - x - 3. 65e-1.59xCos1. 59x - 3 .02e -1. 59xSinl . 59x i.
Mx =-198.1 [_3.02e-l.S0xCOS1.59x+3.65e-1.59XSinl.59X]
l'-1 \
+5~8 .:2 6 k,g -mim
x x= 0
Bending Analysis of Circular Cylindrical Shells 59
Because of our sign convention, the positive value for the bending moment at the base
shows that the outer longitudinal fibers are in compression while the inner ones are in
tension.
+ 5 98.
Figure (4-7) Variations of ben ding moment and hoop force along
the height of the water container of example 4.1
Figure (4-7) shows the hoop stress and the bending moment variations with height. We
observe that the maximum hoop stress occurs somewhere along the cylinder wall, and not at
the base of the shell, as the membrane theory would have had predicted. In fact, the hoop
force at the base, being proportional to w, is equal to zero.
60 Design and Analysis of Shell Structures
Problems
P 4.1 - Consider a long pipe, of radius a, with a radialline load P (per unit width) applied
around the circumference at the mid-Iength of the pipe, figure (P 4-1). Show that the
longitudinal bending moment under the load is,
What would the expression for the bending moment under the load P be if the pipe had a
relatively short length of 2L while being free at the both ends ?
<
- - - - - - -I--- - - - - - f -
Figure (P4-1)
P 4.2 - Consider a long pipe with stiffening rings spaced at 2L centers and subjected to a
uniform external pressure p, figure (P4-2). Outline the general approach needed to analyze
such a structure.(Hint: there are two redundant forces at the pipe ring junction which, due 10
symmetry, are the same on the two sides of the ring-pipe intersection. These redundant
forces can be determined by writing the appropriate compatibility relations).
p p p
4.2 - S. Timoshenko and S. Woinowsky-Krieger, Theory of Plates and Shells, 2nd edition,
McGraw-Hill Book Co., N.Y., 1959
4.3 - G. P. Manning, Reinforced Concrete Design, Third edition, Longmans, Green and
Co. Ltd.,England, 1967
4.4 - Krishna and Jain, Reinforced Concrete, Vol.l,2, McGraw-Hill, New Delhi
4.5 - J. Faber and F. Mead, Reinforced Concrete, Third edition, E. and F.M. Spon Ltd.,
London, 1967
Design of Concrete Cylindrical Shell Roofs 63
Chapter5
5.1 - Introduction
Thin concrete cylindrical shells can cover the roofs of various buildings efficiently and
aesthetically. Large roof spans of bus, railroad, and air terminals, sport stadia, and aircraft
hangars have been effectively covered with reinforced concrete shells, many of which have
been cylindrical. Cylindrical shell forms can be easily shored and easily reinforced.
Cylindrical concrete shell roofs can also be constructed from the precast shell elements.
In previous chapters, we have discussed the membrane and the bending theories of
cylindrical shells. Those theories, and their simplified versions, can be used in the design of
concrete cylindrical shells. In the present chapter, we will present fundamentals of thin
concrete shell roof design. we also present a number of design recommendations and
practical analysis schemes. These design-analysis methodologies will be accompanied by a
set of detailed design and analysis examples.
64 Design and Analysis of Shell Structures
The first step in the structural design of a cylindrical shell roof is the design of the
geometrical configuration of the shell and the structural elements associated with the shell
body. Figures (5-1) show a number of choices which can be made for such a purpose.
These simple shell modules include a single shell simply supported on end arches (or
diaphragms), figure(5-la); a group of simply supported shells, figure (5-1 b); a continuous
cylindrical shell with intermediate stiffening arches and supports, figure (5-lc).
Cylindrical shell roofs are generally composed of the main shell together with some
stiffening and / or supporting elements. These elements could form a collection consisting
of edge beams, stiffening arches, and end members. Figure (5-2) shows a typical
continuous cylindrical shell roof having these elements.
The end edge members themselves can have various forms. They can be in the form of a
solid diaphragm, figure (5-3a); a tied arch (with or without verticals), figure (5-3b); a simple
arch, figure (5-3c); an elevated grid, figure (5-3d); or an arched truss, figure (5-3e). Figures
(5-3) show varieties of end-stiffening and supporting members associated with a shell roof
structure.
I~
(a) solid diaphragm (b) tied arch with verticals (c) simple arch
HJD1
-.~.>
- - ..
-. -- --=-- --=
A shell roof could be placed on vertical supports or on its foundation in a variety of ways.
Figures (5-4) show some possible vertical support arrangements for eylindrieal shell roof
structures.
(4)1 (3)W
(1) (2)
(6)
We propose some suitable relative dimensions for cylindrical shell roofs in figure (5-5)
and in its accompanying table. Figure (5-5a) relates to a group of cylindrical shells with an
edge beam located at the longitudinal edge of the last shelllocated in the shell group.
The shell dimensions given in figure (5-5b) are for a particular type of cylindrical roof
called the Northlight shells. This shell type is used to provide natural light through the
roof. These shell types are particularly suitable for factory and manufacturing installations.
The information presented in figure-table (5-5) serves as a useful guide in proportioning
the cylindrical shell roofs.
Large shell spans, up to 120 meters can be constructed using plain or prestressed
reinforced concrete material. These large spans are used in terminals, airplane hangars,
sport stadia, and unobstructed storage places. In such cases, a number of transverse arches
are normally constructed with the shell. These arches stiffen the shell and carry its load to
the foundations. Shelliengths up to 30 meters can be made of non-prestressed reinforced
concrete. For longer shell lengths, the shell should be prestressed. The longitudinal
expansion joints are usually provided every 45 meters of the shelllength.
I I
i I: ! r t
(m) 1(_)
3
;"')
I
I (mml i~~;~~·(~i
:0 60 ;70
-"
30 I :5 :1 70 300
';0 I :) :5 I 80 350
I
3 :5 :8
I 100 I I
390
'~~.'~;j"
\
,.. support
b;O.6d-\.Od
choice of the shell thickness. As a general guideline, a thickness to chord width ratio of
11300 to 11200 could be considered asound choice for concrete shell thickness.
The thickness of concrete shells is usually increased near the edge members, and gradually
reduced to a constant design value. This simplifies the placing of reinforcement: It also
prevents stress concentrations, and facilitates flow of forces to the edges and supporting
members.
Concrete materials have high compressive strength but low tensile resistance. Therefore,
thin concrete shell roofs must be reinforced to acquire sufficient tensile strength.
Reinforcing of the concrete shells is usually provided by means of steel reinforcing bars.
The existing codes of practice for shells normally require that the reinforcing bars be
placed along the general directions of principal tensile stresses. However, deviations from
these directions are also conditionally allowed by these codes. Figure (5-7) shows the
stress trajectories (i.e., the principal stress directions) in a simply supported single
cylindrical shell subjected to its own dead weight. These types of results will be obtained in
the later sections of this chapter.
tensile trajectories
compression trajectories
As we know from the elementary courses on solid mechanics, the stress trajectories
consist of two families of curves which are mutually orthogonal. In cylindlical shell roofs,
one family of these curves represents the directions of principal tensile stresses while the
other family of trajectories gives the directions of principal compressive stresses. In this
sense, the behavior of the shell can be conceived to consist of a mesh of eables and arehes.
As we noted earlier, thin shells are usually accompanied by stiffening and / or supporting
members. These elements must be monolithic with the shell so that they can transfer the
internal forces. On the other hand, these members produce a bending field. This bending
field would require extra reinforcement which must be provided in addition to the
membrane field reinforcements.
Design of Concrete Cylindrical Shell Roofs 69
The knowledge of the bending field in the shell, is crucial in the design of the extra
reinforcements. We presented a qualitative discussion of the cylindrical vaults behavior in
chapter 3; that is useful in devising the general pattern of shell reinforcements, but a
detailed design must be based on a quantitative analysis of the shell.
Tensile stresses in concrete shells tend to cause cracks in the concrete. Figure (5-8) shows
the most probable cracking pattern in a single cylindrical shell. This figure shows the
positions as weIl as the directions of the probable cracks.
The transverse cracks appearing in the longitudinal edges of figure (5-8) are due to
longitudinal tensile force. The longitudinal cracks designated by solid and dotted lines can
be caused by transverse bending moment. Finally, the inclined cracks at four corners of the
shell are produced by the diagonal tension. these latter cracks are called shear cracks. In
addition to these global crack field, local cracks may occur at the junctions of the sheIl with
its stiffening members.
The pattern of shell reinforcement must naturally follow the probable cracking. The steel
reinforcement should , in principle, be placed in the directions perpendicular to the possible
cracks. Figure (5-9) shows a reinforcing pattern for the shell of figure (5-8). A review of
figure (5-8) would easily lead to the reinforcement design strategy adopted in figure (5-9).
Based on this knowledge and the eurrent practiee, we offer a praetical eonstruetion tip for
placing of reinforcements. The shear reinforeement whieh is to absorb the diagonal tension
field is normally plaeed in the lower layer. The longitudinal reinforcement oeeupies the
mid-Iayer; the transverse reinforeement is usually placed on the top. These struetural
reinforcements also serve as shrinkage, distribution, and temperature reinforeements.
The total amount of steel reinforeement in a eylindrieal shell roof depends on the individual
designs. However, as an initial emde estimate of steel eonsumption, the following formula
could be used.
q= t(t + B) + 6
20f (5-1)
In this approximate formula, I is the shelliength, B is the ehord width, fis the rise of the
shell, and q (in kilograms per square meters of shell surfaee) is the amount of the
eonsumed reinforeement.
The values of principal stresses at each point of the shell ean be obtained from the
following relation:
N+N /N-N 2
~ = x 2 cP ± I ( x 2 $) 2 + NX $ (5-2)
and the principal directions ean also be determined from the following formula:
2NXp
tan29 (5-3)
Nx - N$
Assuming that the internal forces Nx. N</I, and NX</I are determined through the analysis of
the shell, we can determine the principal stresses and the principal directions at each point
of the shell, and plot the stress trajeetories for each case.
Figure (5-10) shows the plan of stress trajectories for a unifonnly loaded simply supported
cylindrical vault. In this figure, the solid curves represent the directions of compressive
principal stresses while the dashed lines represent the curves that are tangent to the
directions of maximum tensile stresses.
According to the American Concrete Institute (ACI) code of reinforced concrete shell
design, the reinforcement is to be placed in the general direction of principal tensile
stresses. This code allows for slight deviations of reinforcement direction from principal
direction. In this code, a directional error of 5 degrees or less is considered acceptable.
However, due to constructional considerations, it is usually convenient to place the
reinforcing bars in perpendicular directions. The codes have provisions for these cases. In
these cases, the reinforcing network must be designed so that it can absorb the principal
tensile stress at each point. This situation is shown in figure (5-11).
(a)
Figure (5-11) (a) Plincipal tensile force acting in some elemental
direction, (b) contributions of perpendicular steel reinforcement
to absorb the principal tensile stress
Referring to figure (5-llb), we can write the internal equilibrium equation can be written as
(5-4)
T ds
P
= ASI f SI dx Cos 0 + A
S2
f
S2
dy SinD
In this relation, As 1 and AS2 are the cross sectional areas (per unit shell width) of
perpendicularly placed reinforcing bars, fS l and fS 2 are the stresses in reinforcing bars, Tp
is the value of principal tensile force, and Ö is its angle of inc1ination. Using the geometrie al
relations between ds, dx, and dy, we may rewrite the relation (5-4) as,
(5-5)
(5-7)
This relation ean be effectively used in designing reinforeement, i.e., in detennining the area
of reinforcing bars.
72 Design and Analysis of Shell Structures
One of the simple methods for preliminary analysis and design of cylindrical shell roofs is
called the beam-arch method. The "beam-arch" method makes use of combined beam and
arch actions of cylindrical shells. In this method, beam analysis and arch analysis of the
shell are carried out alm ost independently by the well-known simple theories of the
structural analysis. The results of these analyses are, then, combined appropriately to yield
the spatial behavior of the cylindrical shells.
In the following, we shall outline the principles of the "beam-arch" method for cylindrical
vaults having an arbitrary profile. Later on, we shall also apply the method to the analysis of
a vaulted roof problem.
The beam-arch method of vaulted roof analysis is composed of two steps: (1) the beam
analysis, (2) the arch analysis.
In this stage, the shell is conceived as a beam resting on the end supports and behaving
according to the cIassical beam theory. The beam has the profile of the cylindrical shell as
its cross seetion and the length of the shell as its longitudinal span.
To perform the beam analysis of a circular cylindrical shell, we consider the profile of the
shell to be as shown in figure (5-12). In this figure, <l>k is the half central angle, a is its
radius, and t is the shell wall thickness. The origin of cross-sectional coordinate system y-z
coincides with the centroid of the section, designated by the letter G. This centroid is
assumed to be located at a vertical distance ofz from the crown of the shell. For a circular
sector shown in figure (5-12), the following formula for z in terms of cross-sectional
parameters (a and <1>0 can easily be derived:
(5-8)
The longitudinal membrane stress (expressed as force per unit width, N x) can be calculated
by the flexure theory of beams. The appropriate bending stress formula is
M
N
x
= (...Y:l
I
z)t (5-9)
yy
In this relation, M yy is the bending moment of applied loads calculated about the y-y axis,
~y is the second moment of cross sectional area with respect to y-y axis. A general formula
for Iyy for a circular sector is
2t J <P=4>
<p=O
k ad<jl(aCos<P
.
_ aSl.IlQ')
tP k
2
(5-10)
According to classical beam theory, the relations (5-8) to (5-10) are valid in a principal
coordinate system for the cross section. For non-principal coordinates, we must use a more
general beam formula
The transverse shear stress (expressed by the membrane shear force, NX $) can also be
determined by a formula given in the classical beam theory. So, using the formula for the
shear stresses in beams, we find
(5-11)
In this relation, V is the total vertical shear force at any arbitrary section of the shell, the
symbol Q represents the value of the first moment of the part of the areal section lying
beyond the layer at which the magnitude of shear stress is to be ca1culated. A general
formula for Q for a circular sector is
= 2a 2 t(Sinrp - ~ Sinrp )
<Pk k
The longitudinal and shear membrane forces Nx and NX $, determined through beam
analysis, are functions of x and <». This completes the beam analysis.
74 Design and Analysis of Shell Structures
The second stage of beam-arch method, consists of analysis of transverse arches. In this
stage, a transverse arch having an arbitrarily smalliongitudinal width, dx, is taken out from
the shell. Figure (5-13) shows the free-body diagram of such an arch slice cut out from
some arbitrary location along the shell.
The free body diagram of figure (5-13) shows the applied loading as weIl as the effect of
the rest of the system; this includes the longitudinally varying shear force which has
already been determined. The longitudinal edge effects consist of horizontal shear forces
and the bending moments, both distributed on the edges of the arch.
VQ
P
I Tb
In the vertical equilibrium of transverse arches, the applied vertical fm"Ces are supported by
the vertical component of the internal shear stresses. In this sense, one may conceive of this
arch to have been hanging from the rest of the system, i.e., to have been vertically supported
by the adjacent arches through which the applied load is uitimately transfelTed to the end
supports. Figure (5-14) shows an arbitrary transverse arch in the shell and its role in
supporting its share of the load.
(a) profile of the arch (b) lateral view of arch (c) internal stress in beam
Figure (5-14) The side and the end view of an arbitraJY transverse
arch in a cylindlical shell
Design of Concrete Cylindrical Shell Roofs 75
For a uniformly distributed applied loading, the change of internal shear force, in the
longitudinal direction, is
= ~(~
= n yy
dNX <1> t) Q dv Q (5-13)
dx dx 2It' dx =~yy
w
(5-14)
Having defined the transverse arch and its loading, we can proceed with its analysis.
Depending on the number of unknown boundary forces (M and / or H), the arch could be
statically determinate or indeterminate; in any case, it can be analyzed by standard methods
of structural analysis. The transverse arch analysis gives the transverse membrane force,
Nq,. and the transverse bending moment, Mq,. are determined.
By combining the results of the beam and arch analyses we can find the internal force
system in the shell, consisting of Nx. Nxq,. Nq" and M<I1. The strength of the shell roof can
now be verified against the applied loads. The calculated internal fOl·ces can be used to
design the proper shell reinforcement to support the prescribed loading.
The beam-arch method for cylindrical vaults can be displayed in a tabular form. Table (5-
1) shows coefficients of internal force system for a circular cylindrical shell subjected to its
dead weight and snow type loading. This table gives the internal forces for various values
of the half central angle <Pk.
(a) Single shells with no edge beams provided that I / a >5 (1 is the length and a is the radius
of the shell )
(c) Shell units in a shell group with small edge bearn, if 1/ a > 1.67
Modified version of the beam-arch method mayaiso be used for the analysis of folded
plates.; the topie which will be discussed in chapter 11.
76 Design and Analysis of Shell Structures
,jllllllll g 11 1 1 I I I 11
'~~F4
• I.--- , ----I ~:~-~
"--,-_
.e.
tL
t ".
Cil
"t
[21
5
(l)
tlt
[.)
".
(5)
"t
(6)
5
(7)
"+
(81
N.
~,
• N
x N. S
". N
x
5 M.
.,
Je& (1) ( 2) (l) 14) 15) 16) ,7) (8)
.,
0
0.75+,
4.146
-\.853
-1.498
0.260
-I. 335
-1.127
0.000
0.000
1. 211
-0.01509
-0.00742
-0 00277
4 458
-2 013
-I 617
0.311
-1.404
-1. 176
o
o
1.315
000
000
-0 01796
-0 00897
-0.00337
<0 0 O. SO" -0.443 -0.592 1 929 o 00595 -0 482 -0. S92 2 095 0.00719
0.25 .. 1. 279 0.005 1 67S o 00578 1.389 0 055 1 819 o 00705
0 3 616 0.241 o 000 -0 01641 .3 9~8 o 297 o 000 -0 02006
shell thickness: t = 10 cm
half central angle: <Pk =40 0
Solution:
Location of cross sectional centroid:
Sin~'k
Z= r(l - - - ) 10 (1 _ 0 - 643 )
0_698
~
Design of Concrete Cylindrical Shell Roofs 79
2Sin<l>k ]
I yy = r 3 t [ <l>k + Sin<l>k (CostPk - if>k ) +
2 x 0.643 )]
= (10)3 (..:!:Q. .}0.698 + 0.643 (0.766 0.698
100 L
4
= 0.587+0.10 = 0.687m
To perform the numerieal analysis of this symmetrie problem, we eonsider one half of the
cross section. Then, we divide it arbitrarily into 8 segments, figure (5-16), and from now
ealeulate everything for these discrete segments.
Table (5-2)
0 40 O.
7
I-
1 35 i 0.223
2 30 0.358
3 25 0.418
4 20 0.413
'"
5 15 0.355
6 10 O. 259 0:10 m
7
8
5
0
0.136
O. .
~.,.,,100
Beam Analysis:
The longitudinal bending moment, M yy ' and the longitudinal stress, N xx , are ealeulated at the
mid-Iength of the shell, where they are maximum, while the shear force, V, and related shear
stress are ealeulated at the ends of the shell. The related ealculations are shown below. The
numerieal results for discrete segments are given in table (5-3).
80 Design and Analysis of Shell Structures
M
N = (...nI Z) t
x yy
Wf =(2x250xS + 120B+60)~
2
~ =
,.
=(250 x 10 x 2 x 0.698 + 150 x 2 x 10 x 0.623+60) (2~)
N = ~ _ 54190 _
x4> 2Iyy Q - 2 x 0.687 Q - 39418Q
Toble (5-3)
I Q
pOint! 4>. I
'?
N NX $
'" x
from top m m 3 kg/m kg/m
0 40 1.547 0 60972 0
8 0 -0.793 0 -31248 0
Design of Concrete Cylindrical Shell Roofs 81
Arch Analysis
Since the applied uniform loading is constant along the shell the shear force varies linearly
in the longitudinal direction. Therefore, the longitudinal change of shear force, Nxq,. i.e., its
derivative with respect to x, is obtained by dividing the maximum values, calculated at the
end supports, by 11 2. These values as weIl as the vertical and horizontal components of
shear force, for each discrete segment, are given in table (5-4).
Figure (5-17) shows the side view of half of the transverse arch together with its applied
loading, as extracted from table (5-4). As we see, for this half arch, the vertical as weIl as the
horizontal equilibrium equations are satisfied.
Tab/e (5-4)
tI
'174.5
270.1
144.1 61.'
214.5 I7B',
"'.1 13017"- 11'1.1
IIOJ 1067. . IU1.S .
627.,
,
0= 10 m
The longitudinal edges of the shell in figure (5-15) are fixed against rotation and horizontal
translation. Therefore, the transverse arch can be considered a fixed arch. The redundant
forces are taken to be the bending moment, Mo, and horizontal shear force, H, at the crown
of the arch, as shown in the figure (5-18). To determine the redundant forces, we use the
energy theorems of structural analysis, and in particular the method of least work .
m8 = 0.0
x ( 0.2985 +
2 0.3688,) -- -....
4'3] . --
I:J
,._,g-rn
Jo
s M~ds
U=2 l.
2EI
where
M.
l.
= M0 +HZ' +rn.
l.
Z' = a(l - cos~}
dU au
t:.H = '"
oH = 0 , t:.e:: dM = 0
o
Therefore
S
dU - 2 \
ClH-
~
)0 EI (M0 + HZ' + rn.l. }ds 0
dU
dMo
=2 Jo
S ,
~I(Mo + HZ' + rni)ds =0
Since the bending stiffness of the arch cross section,EI, has been assumed constant, these
relations can be rewritten as
MO J(os Z 'ds + H JS
0 Z2'ds Z'rn.ds
l.
o
Numerical calculations related to these integrals can be perfonned in a routine fashion; the
final results are shown in table (5-5).
84 Design and Analysis of Shell Structures
Table (5-5)
. I
seetlon Z' da m. Z'· Z'da Z"da miZ'ds mida
1.
8 0 0.436 0 0 0 0 0 0
MO - -350.15 Kq-.!m
H .. 5498.24 kq/m
Onee the quantities Mo and H have been determined, the intern al fOJ·ces ean be easily
obtained using onIy the equations of statics. These ea1culations and the resulting values of
internal forces are presented in tables (5-6) and (5-7).
Design of Concrete Cylindrical Shell Roofs 85
Table (5-6)
8 0 0 -350.15 0 -350.15
Table (5-7)
1356.8
~
1039.4
rvs~ N... 5+6
Xqfin
102.9 1142.3
1 35 41 729.2 597.3 26.4 623.7
2 30 -238.5 -337.8 -292.5 -119.3 -411.8
3 25 -508.7 -1639.5
I -1485.9 -215 -1700.9
4 20 -652.8 -2973 II -2793.7 -223.3 -3017
5 15 -624.7 -4154.1
I -4012.6 -161.7 -4174.3
6 10 -432.3 -5031.6 -4955.2
.-
-75.1 -5030.3
-
7 5 -124.5 -5498.2 -5477.3 -10.9 -5488.2
8 0 50 -5498.2 -5498.2 0 -5498.2
The results of these ca1culations are summarized in the set of figures (5-19); these show the
variations of internal forces across the shell profile.
86 Design and Analysis of Shell Structures
Finally, we calculate the horizontal thrust, R, along the longitudinal edges of this inner shell;
it is
aN
R; H - roxX~dY; 5498.2 - 6855.0; -1356.8 Kg~
The lateral thrust, R, is negative; this means that the direction of R is outwards and not
inward, as would have been the case for arches and frames. In other words, the inner shell
has a tendency to shrink inwards along its longitudinal edges, so that the restraining effect
of adjacent shells is to constrain this inward movement. This quality differs from our
experience in the analysis of framed structures; and has its own structural implications in
the design of such shells.
I 1
(a) (b)
Nx ,,/m
.•
-, -1 a .. I .. 2 .. 1....1o
The analysis of this problem could have also been carried out with the help of table (5-1).
Design of Concrete Cylindrical Shell Roofs 87
Calculations in the ASCE manual are based on simplified versions of the exact cylindrical
shell equations (Flügge equations). These calculations relate to the so-ealled "long shells"
(for which I / a>1.67) and short shells (for which I / a <1.67).
The ASCE tables are based on the force method of structural analysis which has been
outlined in previous chapters; the forees at the longitudinal edge of the shell are chosen as
the redundant forces.
The ASCE tables use Fourier se ries expansions. All distributed applied loads, as weIl as
distributed edge effects, are expanded into Fourier si ne or eosine series in the longitudinal
variable, x. Figures (5-20) show the first terms of the Fourier expansion of the redundant
edge force distributions. Similar type of distributions are also assumed for the edge beams.
Note that the ASCE tables choose the origin of the transverse angular coordinate, <1>, at the
longitudinal edge of the shell. Therefore, the angle defining the normal to the surfaee at
each point is measured from the normal to the shell at its longitudinal boundary.
In the analysis of cylindrical shells by the ASCE tables the following procedure should be
followed:
(4) Superposition
Having obtained the redundant forces, we superpose the results of calculations performed
in steps (1) and (2) and find the total force field in the shell.
The ASCE tables provide a set of eoefficients for the interna! forces in a variety of circular
cylindrical vaults, including single shells with or without edge beams,and shell groups with
or without edge members. The ASCE tables are included in the appendix.
88 Design and Analysis of Shell Structures
(al (bI
Positive dlrectlons
of N)(~
b J cos T
(cl
(dl
In this section, we present some design examples of reinforced concrete shell roofs. We
assurne that the analysis of the shells under design has been calTied out by the beam-arch
method, the ASCE tables, or some computer program. In practical designs, differences in
solutions obtained by different means of analysis are usually of secondary importance.
Design:
The overall dimensions of the shell are given in figure (5-21). The objective is to design the
steel reinforcement for given shape.
Tables (5-8) and (5-9) summarize the results of analysis of this shell performed by using
the ACI tables. We would have obtained similar results if we had analyzed the shell by
other methods.
90 Design and Analysis of Shell Structures
Table (5-8) Internal forces in the single simply supported shell of tigure (5-
21)
.0 ~
T
". (J<q/ll» ".. (J<q/ll»
111,
N. 1I<q/ll»
11' I 3/8
M4 II<q....vmj
" .....,
1/8 1/' 3/8 1/2 0 1/8 1/' 3/8 1/8 1/2 0 1/8 1/, 3/8 1/2
...
'0 o -2U3 ,",331 -6853 0 0 0 0 0 -13'9 -2493 1-3257 -3525 o -50S -933 -1220 -1320
30 o -2564 ... 738 ,",190 -67QO -2013 -1860 -1423 -770 0 -5513 -10186!-13308 o --457 -
.....
-14405 -1102 -1193
20 o -2115 -3907 -5105 -55216 -84<&' -7804 -5913 -]233 0 o -11591 -2H1S!-27984 -30290 . 0 -267 .... S -698
10 o -942 -1740 -2214 -2461 -1-4798 -13672 -1046" -S663 0 -1353 -2500 1-3267 -3536 o -34 "'2 -81 -OB
0 0 0 0 0 0 0 0 0 0 0 .52641 +972671+127086 +137557 0 0 0 0 0
~
Table (5-9) Principal tensile stresses and directions in shell oftigure (5-21)
N., ..... • ...., ..... • .... , .... , 8 ...., ..... • ..... ..". •
40 0 0 0 -2623 -1349 0 -4946 -2493 0 -6331 -3257 0 -6853 -3525 0
30 -2013 2013 4S -6-412 -1665 26 -10535 -4J89 ,. -13390 -6108 6 -6700 -14405 0
.
20 -8477 B477 45 -15983 "2277 29 -23261 -2064 17 -28432\ -4657 8 -5526 -)0290 0
10 -1479B 14798 4S -14821 +12590 45 . -12526 +-8351 -.'SS I ·2914 42 -2461 -)536 0
- -
0 0 0 0 0 +52641 0 0 +97267 0 0 ! ·'27ca, 0 0 +-137557 0
Figure (5-22) shows variations of internal forces Nx. N<I>. and M<I> at the mid-length and the
variation of NX<jI at the end support. For Nx• the numbers in this tigure should be multiplied
by a factor of 10000 (kg Im). For other force quantities. the multiplication coefticient of
1000 should be applied.
crown "0 , ..
I Cl)
•0 Nt M•
I
,
.
2
3.'
..
u Nx
u.
'lO
c: "- I
(\j ?' i I
iii
,.
I
'5 '0
~ I
,
~ I '\....
........ .
€
~
0
I I
10 12 •
U
Once the internal forces are known, the reinforcement design, for each subdivided region of
the shell, can be carried out according to the standard Working Stress Design (WSD)
method.
Figure (5-23) shows the plan of reinforcement pattern for the shell of figure (5-21). Due to
the double axial symmetry, only a quarter of the shell is shown. Figure (5-24) shows a
detailed section of the reinforcement at the corner region. The reinforcement of the shell in
this region consists of longitudinal, transverse, and diagonal (shear) reinforcements.
• ., .. AI JO cm '11 al 20 cm
end diaphragm
~3SaE _ _Rlm_ _mm~~
~12at20cm~----~~~~
section
,.10.0 m
To design an inner shell of the shell group, we take it out and analyze it using one of the
methods discussed before. Figures (5-26) show the physically reasonable assumptions
which can be made. These assumptions are also useful in the analysis of shells by
computer programs.
The results of analysis of an inner shell are summarized in tables (5-10) to (5-14).
Table (5-10)
~I 0
I 1/8 I 1/4 I 3/8 1/2 I
I
~o
!
I
0 1-2601 -4807 I -6281 -6798
30
20
i
I
0
J
1-2237 -2293
-4134 -5401
-2996
-5846
-3243
10
0
I 0
I 0
1--241241
507
-44
937
-57
1224
-62
+1325
Table (5-11)
Table (5-12)
40 0 0 0 0 0
30 -7224 -6674 -5108 -2765 0
20 -12496 -11545 -8836 -4782 0
10 -11992 -11079 -8480 -4589 0
0 0 0 0 i 0 0
Table (5-13)
.
M, (Kg-m!m)
40
30
0
0
-121
-53
-223
-98
-292
-128
I -316
-139
20 0 +110 204 266 +288
216 282 +305
I
10 0 117
0 0 -329 -608 -795 -860
94 Design and Analysis of Shell Structures
Table (5-14)
~o
~, ~2 ~, I ~2 ~, ~2 .
lt., , ~2 "P, ~2
30 -7224 +7224 -13352 1770 -19021 1-2381 , -22996 -4966 -24920 -5846
20 -12496 +12496 -14525 8792 -14629 4036 , -13106 -734 -11738 -3243
10 -11992 +11992 -8035 15299 -4131 1755] i -1179 18720 +19044 -62
0 0 0 507 27635 937 51063 i 1224 66717 +72214 +1325
Figure (5-27) shows variations of internal forces across the shell cross section. For the
transverse moment, M4I, the values of the table should be multiplied by 1000 (kg mim).
Also, for Nx a factor of 10000 (kg Im), and for Nell and Nxell a multiplication factor of 1000
(kg I m) should be applied.
crown
I I I
-
HO I '
E ...-
.>- t-r \ 11 1 I
( ])
200 ........ "l \ V-1~ I
Cl
"0
(])
./
,/
N~~ \ \ I
I
-E '- \1 o
" o.
C
ro r-- t--. /l i\ "-
üi r- r- JJ i \
'ö - 6 _4
_I" -12 -10 -8 -2 8
Figure (5-28) shows the quarter plan of reinforcing scheme of an inner shell from the shell
group in figure (5-26).
"lIa.15 tm
C rown
-........ ""- I
E
Cl
""''-....
r--- "-
~
.c 1"-
a.
ctl
'ö " mid-Iength
"C
c
Cl) " ~
f--- \2.81 m .1
Table (5-15)
N
x
Table (5-16)
Table (5-17)
K 40
0
0
1/8
0
1/4
0
3/8
0
1/2
0
I
Table (5-18)
4) Plan of reinforcement:
4l12C1t)] tm ~
crown ./'
1'2 Cl 2S c".,
~~~
""
I I /1
I I
11
end diaphrag rn
I
!
I
I
i
I
, I
fj~+~i
,,,,-, i I
!
i
,~ i I I I
t'0 rx: '" i
iI ~
i f
~ ['--,,~<~ I ~ I I
m/
f0
~"
~ l\:~
f>'~\:
0~ N
[}::
~~,,.
I
I
j
-*H
1 I
~ 11~
cJlI2 0115 tm
edge L.4> 1111']0 tft'\
Table (5-19)
I
10 0 -1677 -lB15
0 I 0
\-695
i :' 32 :
-1283
I
42B-----L 559 60S
I
Table (5-20)
~xz: i
Nx
0 1/8 1/. 3/8
Table (5-21)
~I 0
I
1/8 1/' I 3/8 1 1/2
I
I f)(g1Vrn )
I 40 I
0 -92 -170 -223
I -241
I 30
20
0
0
-33
55
-60
102
-79
133 1-::'
I ~o
0 67 124 162 175
I 0 -191 -354 -462 -500
Table (5-22)
1'~
;:<'('1 0 1/8 1/4 3/8 1/2 i "xl>
I [l'.g-'T,j
'0 0 I 0 i 0
!
3D
I -27':'2 -1484 I
20 I I
1J
Design of Concrete Cylindrical Shell Roofs 99
Table (5-23)
3) Reinforcement system:
'"
!~ t". .c
I"", ,,\ Öl
c
~"""'~"
,," " ~
-0
"", ",'~ I 'E
~"",
,>,,>~ ~,\, "
~,,'\ ,~
,
, I
E .,
~'"
~'\'\" ~"''''~
U o
'\' ~,\" I'} "
;1,8
~ I'
LI
~I'..
I I
edge.l 1 1
t/2
° '/8 '/4
,12'@30cm
31/8
12'ill~ • 12 '@ 20 cm
- = , ,4 "-t: 10·
"to 1'cil
12 r@ 15 cm
Problems
P 5.1 - Analyze the inner shell of the multi-cylindrical shell roof shown in figure (P 5-1).
The profiles of shells are semicirc1es. The shells are semicircular,of thickness t; the width
of the inner and the outer edge beams is 3 t, and the height of the edge beams is 5 t. The
width of the shells is one half of the roof length. The system is subjected to its own dead
load.
Edge beam
V /
Length of the shell
P 5.2 - Figure (P 5-2) shows a shell roof over a bus or train station. The length of the shell
is 10.0 m, and it is supported by arched diaphragms which cantilever out from the columns.
The radius of the circular shell section is 2.4 m. The intensity of applied load is 300 kg /
m 2 distributed over the shell area. Analyze this shell by using the beam-arch method, and
present a design of shell reinforcement.
Shell
I 10 cm I
DIaphragm
arches
~~
(a)
\ (b) I
5.4 -J. E. Gibson, The Design of Shell Roofs, E. and F. N. Spon Ltd., London 1968
5.5 - Design ofCylindrical Concrete Shell Roofs, Manual of Engineering Practice, No.31,
American Society ofCivil Engineers, New York, 1952
5.7 - C. B. Wilby, Design Graphs for Concrete Shell Roofs. Applied Scicnce Publishers,
Ltd., Barking, Essex, England, 1980
5.8 - Coefficients for design of Cylindrical Concrete Shell Roofs. Portland Cement
Association, Skokie, Illinois, 1959
5.9 - J. Chinn, Cylindrical Shell Analysis Simplified by the Beam Method, ACI J., Vol. 55,
May 1959
5.10 - V. N. Baikov, (ed.), Reinforced Concrete Structures, MIR Publishers, Moscow,
English Translation 1978
Chapter 6
6.1 - Introduction
In this chapter we will study the membrane behavior of shells of revolution with double
curvature. The shell types analyzed in the present chapter are a subclass of shells of
revolution having non-zero positive Gaussian curvature.Shells with non-zero Gaussian
curvature have non-developable smfaces. Hence they are stronger, stiffer, and more stable
than shells with zero Gaussian curvature.
Domes have positive Gaussian curvature. These types of shells are used to cover the roofs
of sport halls and large liquid storage tanks. The containment shield structure of nuclear
power plants have also dome-like roofs. Various pressure vessels are either completely
composed of a single rotational shell or have shells of rotation as their end caps.
Conical shells,with zero Gaussian curvature, are a member of the class of shells of
revolution. They are used to cover liquid storage tanks, and the nose cones of missiles and
rockets.
In this chapter, we firs derive the goveming membrane equations of shells of revolution will
be derived. Then, we apply them to the analysis of a number of shell problems including
dome-like roofs cones. Later on, we determine the membrane displacement field of
rotationally symmetrie shells. These results will be used in later chapters in which more
complete design and analysis of the shells of revolution are presented.
104 Design and Analysis of Shell Structures
The middle surface of a shell of revolution with non-zero positive Gaussian quadrature can
be described rather like the earth. Thus through any point we may take two seetions, one
perpendicular to the axis of revolution, one containing the axis. The first cuts the shell in a
parallel curve (circ1e of latitude) the other in a meridian (plane of longitude). At any point,
the radius of curvature of the meridian is called r 1, and the radius of parallel circ1e, r, is
projected value of another principle radius of curvature which has been denoted by r2.
(a) (b)
Parallel circles form the perimeter of the base of a cone the apex of which is the center of
curvature for r2. Due to rotational symmetry, the center of curvature of r2 always lies on the
axis of revolution. However, the center of curvature of fJ does not have to lie on this axis.
Denote the angle between the normal to the surface at P with the axis of revolution by <». We
also denote the horizontal angular position of P, from some arbitrary origin, by the angle 8.
The direction of the axis of revolution is assumed to coincide with the z axis.
Membrane Analysis of Shells of Revolution 105
I---------~
normal
H
,,
Referring to figure (6-2), the radius of parallel eirc1e, r, at point P ean be written as
ds = r 1 dq,
(6-1)
dr=ds Cosq" dZ~ds Sinq,
dr
r 1 Cos<!> (6-2)
dtf>
d: (6-3)
r 1 Sinq,
~
Combining the above relations, we obtain the following inter-relation between the surfaee
parameters f}, f2, and <1>.
1 dr r 1
~ ..,. = -r-
d'" Cotep
2 (6-4)
106 Design and Analysis of Shell Structures
Meridian
The free body diagram of figure (6-4) shows the intern al membrane forces, Nq" Ne, N$9. and
their differential variations, NeI> designates the meridional force, Ne the hoop force, and NeI>e
the membrane shear force; the quantities Pr, PeI>. and Pe represent the intensity of external
distributed applied loading, in the r, q" and e directions, respectively.
We write the equations of equiliblium in the <», 8, and r direetions. Beeause of the double
curvature, the membrane forees have projections in all three direetions and thus eontribute to
all three equilibrium equations. Figures (6-5) show the eontributions of Nq, and Na in
various directions.
(6-5)
o (6-6)
o (6-7)
The third equilibrium equation is obtained by projeeting all the forees in the direction
normal to the shell, i.e., in the r direction. By doing so, we obtain
108 Design and Analysis of Shell Structures
~+~=p
r r r (6-8)
1 2
Equations (6-6), (6-7), and (6-8) constitute the governing equilibrium equations of the
membrane theory for shells of revolution. These relations yield Ncp, Ne • and Ncpe. i.e., the
membrane force field in the shell.
Note that the meridional and hoop forces Ncp, Ne appear in all three equations. This indicates
that a doubly curved shell is a complex and efficient structure; all three fm·ces Ncp, Ne and
Ncpe contribute to carrying the load in any direction. The spatial interaction of internal forces,
manifested in their presence in all equilibrium equations, is indicative of an efficient and
profound behavior of doubly curved shells. This spatial collaboration is very rare in framed
structures.
In a number of important loading cases, such as the dead weight and internal fluid pressure
loading, geometrically complete shells of revolution have axisymmetric behavior.
Axisymmetric behavior is independent of the variable 8. The loading, internal forces. and
deformations can vary in the <1> direction.
d
(6-9a)
(r Ncp) - r1Necoscp - pcprr 1
dT
lr1 Ne
+--
r2
P
r (6-9b)
(6-10)
Since Pe = 0, the last (and already decoupled) equation would be identically satisfied, and so
we shall have Ncpe = o.
Membrane Analysis of Shells of Revolution 109
(6-11)
This gives the meridional membrane force, N</>; the hoop force, Ne. can be conveniently
obtained from the relation (6-9b).
The relation (6-11) has a clear and useful physical interpretation. To provide this
interpretation, we intersect the shell with a plane section nOimal to the axis of revolution at a
point having an arbitrary normal angle <jl.Figure (6-7) shows a sector of the shell lying
above this plane section; it also shows the effect of the lower part on this piece. which
consists of the internal meridional force uniformly distributed at the base of this sec tor.
Figure (6-6) shows the resultant of applied axisymmetric loading is also shown. denoted by
R.
Figure (6-7) shows a shell element and its meridional side view; an elemental area of the
shell is
We can verify that the differential of applied forces projected in the direction of the shell
axis of revolution has the value:
110 Design and Analysis of Shell Structures
If we now go back to the integral (6-11) we note that the integral is the resultant of applied
loads projected along the shell axis of revolution. We also note that the quantity 2 1t N<j> (r2
sin 2 <j» is the sum of reactive forces projected along that axis. The constant Cis the sum of
applied concentrated forces, if any, along the shell axis of revolution. Therefore, the relation
(6-11) is the equation of equilibrium for the global shell sec tor shown in figure (6-6).
",-
I I
---~
\,
I
\
I r ' \ dA
11
t
-" -- .
.... - -," /
/
P, 'OS+-~ si ••
p• • in.
(a) (b)
This procedure, called the method of sections, gives the membrane force field in the form:
R (6-12)
N<j> 21Tr 2 sin 2<j>
R
e=
N + + Pr r 2 (6-13)
21Tr 1 Sin 2 q,
In the following sections, the membrane theory developed in this section will be applied to a
number of shells with axisymmetric loadings.
Membrane Analysis of Shells of Revolution 111
The resultant. R, of applied dead load projected along the vertical axis of revolution is
R = 2n f:j>
·0
2 . 2
a qS1n:j>d:j>= 2na q(l - Cos:j»
The relations (6-12) and (6-13) give the membrane internal forces
Several interesting observations can be made concerning this solution. First, the expression
(6-14a) always yields negative values for N4> throughout the shell. Hence, the meridional
force in a dome under its own weight is always compressive. Secondly, the hoop force, Ne.
is compressive at the top, but changes sign somewhere along the meridian and becomes
tensile in the lower part of the shell. Ne is zero when:
(6-15)
1
1 + C"sl/l - cos:j> o
• I
_ oq -oq
(];f= I ±
51 52
T
Ne
(a) (h)
Figure (6-8) A spherical dome under its own weight, (a) the dome
section, (b) variation of meridional and hoop fOl-ces
112 Design and Analysis of Shell Structures
Figure (6-8) shows plots of the solutions (6-14), i.e., variations of meridional and hoop
stresses along the height of the dome. We observe that at <p = 0, N<j>= Ne = - aq /2. For a
hemispherical dome, the absolute maximum of the both membrane forces is (aq).
It is interesting to know that the ancient engineers were wen aware of this structural behavior
of domes. When building domes with masonry materials, such as adobe brick, fired brick,
and stone, which are relatively weak in tension, but strong in compression, they would
confme their dome sector to the compression zone or, for high domes, would reinforce them
in the tensile region. The hoop reinforcement would consist of wooden ties placed along the
parallel circles; when tied together they would form a closed strengthening ring capable of
absorbing tensile forces.
p p P cotQ
I \>
'.
\
'
\
\.;
\ t
(a) (b)
Figure (6-9) A spherical dome with a skylight and a ring at the top
The ring at the top region acts in compression and is meant to reduce the internal forces in
the shell body. The weight of this ring is applied to the shell as a unifOlmly distributed line
loading.
Assurne that the weight per unit width of the top stiffening ring of the spherical dome of
figure (6-9) is equal to P. Then, for a dome with skylight we have
R ; 2w J4> a2
q S1n 4>d4> + 2wP a sin ~
<1-0
Membrane Analysis of Shells of Revolution 113
Solution:
If the dead load has intensity w, relations (6-12) and (6-13) give the membrane fm·ces:
K
:x
N ~~~
8 4
z
2
r
C ~-f
z
Figure (6-10) ParaboloidaI dome of example 6.1 and the
equation of its melidionaI curve
114 Design and Analysis of Shell Structures
The maximum compressive meridional force under combined loading occurs at the base of
the dome. Hence, K = (2 x 6.0)/2.4 = 5.0. Therefore,
N I
4> rnax -
- 10.0 X 2240 X 2.40[ (1 5 2)2
? +
~J
- .1+5~
100 X 6 X 5-
1
XTöö
2710.4
a~lmax ----'= = 2. 71 kg'~
10
The maximum hoop force under combined loading is tensile and occurs at the base; its
magnitude is:
=268.8-471.6+305.9=103.1 kg'm
1
103.1 x TIiö
0.10 kg'cm2
10
As we observe, the calculated values of maximal stresses are remarkably low for such an
structure. In particular, the maximum tensile hoop stress is very small and can be carried by
weak materials such as various masonry products. This is generally true for most shell
structures. Many historical masonry shells which still remain after many centuries testify to
this unique feature of shells.
Membrane Analysis of Shells of Revolution 115
Tbin shells have various applications as liquid storage tanks and fluid containers. In these
applications, varieties of shell shapes may be envisaged, designed, and manufactured. In this
section we present a membrane analysis of three types of these containment vessels.
Tbe applied internal pressure in the tank of figure (6-11) filled with a liquid of density r is,
P~ = 0 , Pe = 0 , Pr = ya (1 - Cos 9) (6-17)
At the apex of the shell (~ =0), the denominator of the above expression vanishes resulting
N<I> infinite. Physically, this force must be actually zero at the apex. To have a finite value at
the "regular" apex point, we evaluate the constant C so that this condition is satisfied. Hence,
we evaluate the expression in the nominator at <I> = 0 and then set it to equal to zero. If we do
so, we find C = 1/6. Substituting this value in (6-18), and making use 01' the formula (6-
9b )we obtain
2
=~ 1 - Cos ~
N~ (l + 2 Ces ~)
6 1 + Ces ~
(6-19)
2
=~l - Ces ~ (5 + 4 Ces $)
Ne 6 1 + Ces $
The expressions (6-19) are valid for the part of the tank lying above ~o. For the lower part,
we must fmd another proper value for the constant C which assures the finiteness 01' Nc)l at ~
=~o. By a manipulation similar to the previous one, this value comes out to be C = 5/6.
Hence, for the lower part of the shell we find
2 2
N~
=~ 5 - 5 Ces ~ + 2 Ces 4
6 1 - Ces ~
2 2 (6-20)
=~ 1- 7 Ces <1>+ 4 Ces p
NO 6 1 - Ces q,
116 Design and Anal ysis of Shell Structures
N. ll.O~;a,J
(a) spherical tank (b) membrane forces
Variations of membrane forces in the liquid tank are plotted in figure (6-11b). Note that the
internal force quantities are discontinuous at the ring section. This discontinuity is due to
the presence of the supporting ring. In fact, due to presence of the ring element, a loca1
bending field develops around the ring section. The membrane theory is of course incapab1e
of determining the bending forces.
According to membrane theory, the difference in internal forces in the upper and lower parts
of the shell, in figure (6-11a), must be absorbed by the stiffening supporting ring. The
discontinuity in Ne creates in the ring a compressive force of magnitude:
2ya2 Cos<j>o
-----
3 Sin<j>o
Also, the sum of the meridional forces at the ring section exerts a vertical component equal
to:
2ya2 S· n.
-3- lll'l'0
A pre1iminary design of the ring can be carried out on the basis of these quantities.
"(. o
f
t,t----------1
•
A'
I
H
+:~B'
The liquid pressure loading at the bottom spherical part of the tank is,
)
(6-21)
Substituting these values into the formula (6-11), and integrating from q, = rrJ2 to q, = 0, we
obtain
(6-22)
where Ncjl9 is the value of Nq, evaluated at q, = rrJ2. It is, on the other hand, equal to the value
of Nq, at the level A. Their common value can be found by using the formula (6-11) again
but with H =0 and integrating it from q, =0 to q, =rrJ2. It can also be found by the method 0/
sections applied to an arbitrary section x-x. The resulted value of Nq,o, found by either of
these two schemes, is
(6-23)
(6-24)
118 Design and Analysis of Shell Structures
The expressions (6-25) and (6-28) give values of N4> and Ne for 1t/2:::; 4>:::; a. For a:::; 4>, the
effect of the line load coming from the support reaction must also be incorporated.
The total support reaction, V, is equal to the weight of the liquid in the tanle Hence
(6-29)
V = ( 3"2 no 1 T ":rtrH
'
+ 3"2 no",) -y = (3"4 )
7T<1 + , ) -y
TIlrH
Using the relation (6-12), we add a term to the expression for N4> from relation (6-24), to
find (for a:::; 4», the following expression:
.y = -ya' (COS3 4> _ cos' 4> + ~ _ !!. cos' "')
. '" ,m' '" 3 2 6 a 2
4 na)-y na'-yH
+ - +
3 2na ,m' '" 2na sm' '"
or
N
<I>
-
-
-ya' (2 cos' <I> - 5 cos 4> +
"""6 I - cos <I>
5) + 2- --ya' H
a
<I»a
(6-30)
Membrane Analysis of Shells of Revolution 119
From expressions (6-30) and (6-31) we see that in the region a ~ cjl the membrane forces
Ncjl and Ne are always positive. This is true since the shell bellow the ring support is actually
"hanging" from the ring and thus carrying the load by tension. For 120° ~ cjl, Nej> and Ne are
given by the expressions (6-25) and (6-28).
Having obtained a general solution to cylinder-sphere pressure vessel, we now consider the
following special cases :
(1) Wa = O. In this case, we have the complete spherical tank which was treated before. In
this case, Ne is always positive (Le., tensile ) whereas Ncjl is negative in the region 120° ~ cjl, as
seen from (6-25).
For the metallic tanks buckling of the shell is a problem and so the compressive forces must
be taken seriously. On the other hand, all metallic materials can sustain tensile stresses
efficiently. Therefore, in the design of a metallic shell it would be desirable to place the
support at such an angle that both Ncjl and Ne are tensile throughout the shell body.
(2) Wa = 1. In this case, in the region 112.5° ~ cjl, the meridional force Nej> is compressive. For
Hla much larger than unity, Ncjl would be compressive throughout the region 90° ~ ej>.
Comparing this case with the previous one, we conclude that to ins ure the existence of a
tensile Ncjl field, the angle 0., determining the location of the support, should vary from 90 0
(for very large H/a) to 120 0 (for H/a = 0). The expression (6-28) shows that the
corresponding hoop forces Ne will also be tensile.
In this liquid tank problem, as in the previous example, the internal forces at the ring support
location are discontinuous. The role of the supporting and stiffening ring is to absorb the
differences between the internal forces. Due to the incompatibility of defOlmation at two
sides of the ring, some bending field develops in that region. This state of affairs is
demonstrated in figure (6-13).
lJ
ßl.-
I N~l
Net vertlcal
load
N<-:'-1~nb.lanCed
hortzon'al
~ componen.
SuPPOrt
Pressure vessels are used to store pressurized liquids or gases; they must resist high
internal fluid pressures. Pressure vessels are normally built 01' metals and I or composite
materials. In such constructions, the stresses due to the weight of the vessel are negligible
compared with the internal pressure-induced stresses.
Pressure vessels are normally shells of revolution. In this section, we will determine the
internal membrane forces in axisymmetric pressure vessels with arbitrary meridional
geometries.
Consider apressure vessel having an arbitrary meridional shape of radius rl, as shown in
figure (6-14). If the internal pressure is assumed to be equal to p, then the applied force
components will be
=P (6-32)
P
e =0 P<j> =0 P
r
1
(6-33)
This integration can be performed without assuming a specific shell surface; using the
expression (6-9b) we obtain:
(6-34)
For the special case of a spherical pressure vessel of radius a, the expressions yield N4> =Na
=pa/2. For a cylindrical pressure vessel having r1 = 00, and r2 = a, we find N4> =pa/2 and Na
=pa.
Membrane Analysis of Shells of Revolution 121
Shells structures can be subjected to loadings wh ich are not axisymmetrie. Examples of
nonaxisymmetic loadings are: wind forces, earthquake effeets, soil pressure on buried pipes,
and temperature gradients in composite and I or metallic shells.
aN~ . ON pa
r2 "ej> S~n ej>+(r 1 +r 2 )Nej> Ces ej>+ r 1 ae (6-34a)
r2
aN~e
~s~n
.
ej>+ 2r 1 N4>0 Ces ej>- r 2 a0 3. ap (6-34b)
= - r 1 r 2 (Pe Sin ej>+ a0 r )
For a distributed loading we ean expand the loading funetions, Pcp, pe, and Pr, in terms of
Fourier series.These expansions have the following forms:
P0
0
co
r Pan Sin ne + E
o
..
.L
q0 Ces
'n
1~0
(6-35)
1
co co
p r p
r = 0 rn
Ces n0 + r q Sin n0
1 rn
For known loadings, the so-called "Fourier eoeffieients" pcpn,pen,qcpn, ... , ean be determined
using Fourier series analysis.
Equations (6-34a,b) have solutions which are separable in e and <1>. For eaeh value of n there
are two different solutions: one in whieh Pcp, Pp,N~ Na are funetions of<l> multiplied by eos
ne, while pe. Ncpa are funetions of<l> multiplied by sin ne; allother in whieh eos ne and sin
ne are interchanged. Both solutions are found in the same way; for the first we write
p
ej>
= P ej>n Ces n0
(6-36)
122 Design and Analysis of Shell Structures
where, Ncpn.Nen, and Ncpen are, in general, functions of <1>. Substituting these expressions into
(6-37) and cancelling the common factor of cosne in (6-34a), sinne in (6-34b) we find
d.~1m r1 N pen r1
+ (1+-) N Catljl+n. - =r, (-P +P Cat CP)
d<j> r2 Ijln S~nljl r 2 J. cpn rn (6-38a)
(6-38b)
For a simple model of wind forces, acting on the shells of revolution, we assurne the
following distribution,
dN
~+
ci;,
(6-40)
dN _
~ + 2 C
d~ at",
.• N _n_
~en+ Sinq,'~n
t, ( P n
a - en+ Sin<j>
P)
rn
(6-41)
Membrane Analysis of Shells of Revolution 123
dU n a (_P _ P + n+~os4 p )
dcl> + (2Cot cl> + Sin cI» U 8n cl>n Sl.n cl> rn
(6-42)
dV + (2Cotcl>- _n_)V = a(P _ P _ n-Cospp )
dcl> sin cl> 8n cl>n sin cl> rn
dU (6-43)
d4 + P(~).U + q(cI» o
(6-44)
lA l + -t' cos-q,)]
.~
• t , .. "..
.1..,-"-.1....1.::. .......
U Pa (CoS1'-
Sin-4
l-Cosc!> 1 3 1 (6-45)
V [B l - Pa (Cos4- - 3 Cos 4) J
S'l.n 3",_
Returning to the relations (6-41) and multiplying the resulting expressions, for the actual
field variables Nq,n and Nq,9n, by cos e and sin e, respectively, we obtain
Al +B, Al-B l
Cos8 I_ _~ + 4
)J
')
N4 - 2 - Cos4 + Pa(cos-~ 1/3 Cos
3 L 2 ~
Sln y
Al-B Al+ß l
N
48
~~
.i
1-2- l+ - 2 - Cos4 + Pa(Cos y - 1/3 Cos
3
,) J
Sin c:>
The integration constants Al and BI can be determined by imposing the physical condition
that Nq, and Nq,9 must be finite at <I> = O. Hence, after some algebraic manipulations, we
obtain the following final solution to the problem
Pa (2 + Cos ~ ) (1 - CoscjJ) Cos PC 8
N~ = - "'3 (1 + Cos q,) Sl.n 4 os
Plots of variations of internal forces in this shell are presented in figure (6-15). Note that
because the loading and the behavior of the shell is not axisymmetric there is a nonzero
membrane shear force field. as weIl as normal membrane forces. Therefore. the meridional
curves and parallel circles no longer represent the principal directions of the internal
stresses. For the present case. the plan form of the stress trajectories is shown in the figure
(6-16).
----:.~- -~-lJ
L-------r-----~-- - - - -O.6e7,,0 - - -po
Having found the three components of membrane forces in the c)I and e directions we can
use the well-known transformation formulas to find the membrane forces in an arbitrary
direction. Specifically. we can determine the principal forces and the directions of principal
stresses. Using such relations we can plot two orthogonal families of curves indicating the
principal directions. These are called the stress trajectories. Figure (6-16) shows the plan
of the stress trajectories for a hemispherical dome subjected to lateral wind fOl·ces.
---- --
----
~
-
-+-
-
Figure (6-16) Stress trajectories in a spherical dome subjected to lateral
wind loading
In figure (6-16), the solid and dashed lines indicate the compressive and tensile principal
stresses, respectively. Thus, the behavior of the dome under wind fOl·ces can be conceived to
consist of combined arch and coole actions.
Membrane Analysis of Shells of Revolution 125
Tbe displacement vector in a rotational shell of double curvature generally has meridional,
hoop, and normal components. If the applied loading is symmetrieal, then the hoop
component of the displacement vector is zero. In these truly axisymmetric problems there
are only the displacement components along the meridional and nOimal to the shell are to be
determined.
Consider an infinitesimal element, AB, taken from the meridional section of the shelL Tbis
element is deformed into A'B', as shown in the figure (6-17). The positive meridional
displacement, v, is taken in the direction of increasing <1>, the positive normal displacement, w,
is taken inwards.
The change of length of element AB is composed of two parts: one part arises from the
meridional differential displacement, (dv/d<P) x d<p; the other from the normal displacement,
(w) x d<jl. With the adopted the sign conventions, the total change of length of element AB iso
The meridional strain is obtained by dividing the above change of length to the undeformed
length of the element (rl d<l». So, the expression for meridional strain iso
1 dv IN
E
q,
---
r 1 d.J; r 1 (6-47)
As we see, the meridional component of strain is also affected by the normal displacement,
w, as well as the meridional displacement. v. This result has a simple physical interpretation
which is offered as follows:
Suppose that a spherical membrane "balloon" is inflated by an inner pressure. The balloon
will expand symmetrically so that only there is a radial component 01' the displacement, w.
126 Design and Analysis of Shell Structures
The meridional strain is equal to the change of diameter divided by the Oliginal diameter, Le.,
(wIr). In this example, dv/d<jl = O.
To determine the hoop strain, we consider a hoop element of the shell. Figure (6-18) shows
the change of radius, ~r, of the parallel circle passing through this element. Refening to this
figure we find the following expression:
Since the circumferentiallength change is proportional to the the change in the radius, so the
hoop strain is
1
(v Cos ~- w sin ~)
r
[ .
:J
(6-49)
t.r r0
--'- --I
I
A I
I
~ I
I
I
Figure (6-18) Shell displacement components leading to the change
of radius of a typical parallel circle
v Cot ~ (6-50)
Membrane Analysis of Shells of Revolution 127
Having obtained the kinematie relations and equilibrium equations we now write down the
third group of goveming relations, i.e., the constitutive relations. If the shell is linearly
elastic and isotropie, the two dimensional elastic eonstitutive relations, for a loeal state of
plane stress, are
1
c (Nq, - vN o )
Et
" (6-51)
1
- .
Et (Ne - vN
q,
)
1
v Cot ~ = Et (6-52)
Suppose that we have carried out a stress analysis the shell. Thcn we know the membrane
forees, Ncp and Ne. Furthermore, we ean express the shell radii of eurvature, 1"1 and r2,as
funetions of <p. This means that the right-hand side of equation (6-52) is a known funetion
of <p. Representing this funetion by f (</I), we may write the general form of equation (6-50)
as
dv
v Cot ~ = f (~)
d<jJ (6-53)
dv .
d -- S~n~-v cos~
(__
v_) =~d~t__________ 1 (dv CA.)
Sin~ d~ - v otg ~
d~ Sin~ Sin2~
Therefore
dv d v
d4- - v Cotg Ij'= Sin4- d~ (Sin 4»= f(q,)
or
d v 1
- ( - - - )= - - - f(q,)
d~ S~n ~ S~n ~
Having obtained the meridional displaeement funetion, v, we ean now use either of the two
relations (6-47) or (6-49) to find the normal eomponent of displaeemcnt, w.
128 Design and Analysis of Shell Structures
Consider the spherical dome of figure (6-8) subjected to its own weight. The membrane
forces in this dome were obtained earlier and are given in relations (6-14). To determine the
deformation field in this shell, we use the relations developed in section (6-9). Using the
expressions (6-14), and noting that r1 = r2 = a, we rewrite the equation (6-52) in the
following form:
2
dv a q (1 + \I) 2 (6-55)
v cot l' Et (Cos <p - 1 + Cos 4»
d<p
2
A (Cos ~ - 1+Cos ~) A (6-56)
We substitute this expression into the general solution (6-54) and perfOlm the integration
on the variable <I> to obtain:
,'rcos q, 2 ]
= A J lsw 4> - (1+Cos.p)Sin4> dq,
=.."\
J
:CSd;+COS 2 q,-2
Sin~(1+Cos4>1 dq,=
[J Sin<p f Cosq,-1
A ~+cos.;, dq,+ Sin",(1+Cos")
~ ~
~
A
fiJj.L
Sin4>
d4> = 1n (1+Cos4»+ J Sin</>Cos{1+CoS4>1
-1
d<p
(
J
Cos</>-1
Sin</> (1+Cosq,)
d.p= f
2!
-2Sin 2 = f -Sin 3 <P/2 d4>
2Sin.! cos! ·2cos 2! 2Cos <P/2
2 2 2
Therefore
With v(eIl) determined, we can find w(eIl) from either of the two relations (6-47) or (6-49); the
latter gives:
_ _a__ N +v Cot ~
(6-58)
Et e
In particular we may find the horizontal displacement at the base of the dome ilH (at eil = a).
2
~ :..- \J
(6-59)
~H= r E S= Et ( - - - -COS(l)
l-=os-l
Sina
a 6. am. t = 10 cm a = 60 0
2
Kg/m
q 150
2
Kg/Cm
E 2 X 10 5 \) = 0
Solution:
Using the expression (6-57) in (6-58), we have
2 2
W= - i~ (l+~OS l' - Cos 1') + \i cos.p[ ln (l+COS~)-l+~OSq,]
+C COS .p
and since
H- -lnl.S]
2
C aE :
130 Design and Analysis of Shell Structures
therefore
2
w- a q { (_1_ _ _ Cosq,)+ cosq,[ln (l+cosq,)-1+C01S
- Et - l+Cosq, 't'
J +
2
(J -ln1.5)cos<j>}
2 [
w = aE~ +~ + (ln 2 -1/2 ) - (101. 5- 2/3)
J (6.0)Lx150xO.954
= ..o..:::.~--=;:-=..::...:;~"""'-" 0.03 nm
2 X lo5x 10
This dome has a relatively large span of 2a = 12.0 meters In spite of this, the central
deflection is indeed very smal1. It would be instructive to compare this negligibly small value
with relatively high central deflection of a beam element having the same span and loading
conditions. We would conclude that a dome is several orders of magnitude stiffer than a
corresponding beam structure. As was pointed out before, this high degree of structural
stiffness is typical of all shell forms.
The membrane field of internal forces in domes consists of a meridional force, a hoop force,
and a membrane shear force, figure (6-l9a). For axisymmetric loading of domes, the
membrane shear is zero throughout and the internal force field consists of meridional and
hoop forces only, figure (6-19b).
axis of revolution
meridional direction
hoop direction
(a) dome with non-axisymmetric load (b) dome with axisymmetric load
For axisymmetric loading of domes, the stress trajectories, i.e., the directions of principal
normal stresses, will coincide with meridional and hoop curves; the shear stress is
identically zero along these directions, figure (6-19b).
Figure (6-20a) show the stress trajectories for a symmetrically loaded continuously
supported spherical dome. As we have seen before, from our membrane analysis of domes,
the meridional force is compressive throughout the shell, while the hoop force has a change
of sign from compression to tension. In this figure, the compression field of principal
stresses are plotted by solid curves while the tension stress trajectories are sketched by
dashed lines.
Figure (6-20b) shows the stress trajectories for a spherical dome with four concentrated
supports under vertical symmetrie loading. This figure reveals the flow of forces towards the
supports and the resulting stress concentration near the point supports. A bending field will
develop at these supports to compensate for the shortcomings of the membrane behavior.
The structural behavior of domes can be conceived as the interaction of two mechanisms:
(1) arch action of the shell along the meridional direction; (2) ring action of the shell in the
hoop direction. The interaction of these two mechanisms gives rise to an efficient spatial
behavior of the doubly curved shell.
Sometimes domes are provided with edge supporting and I or stiffening rings. For example,
when there is only a vertical support (such as a supporting wall) the horizontal thrust must
be absorbed by a ring, figure (6-21). To stiffen a dome the designer may place a stiffening
ring at the intersection of the dome with other structural elements.
132 Design and Analysis of Shell Structures
Figure (6-22a) shows a dome roofwith an edge ring. Figure (6-22b) shows a liquid storage
tank with a cylindrical wall, a dome roof, and a stiffening ring at the intersection ofthe two
shell types.
I
- - -1- - -
---1---
I
(a) dome roof with ring (b) dome-cylinder shell with ring
Figure (6-22) Domes with rings, (a) a dome with supporting ling,
(b) a combined cylindricalliquid storage tank with a stiffening ling
The spatialload-bearing behavior of domes can be weIl contrasted to that of the arches.
Figure (6-23) gives a graphical comparison between an arch and a dome. For example, the
effect of boundary shear effect in the dome is local whereas the bending effect in an arch is
global. Figure (6-23) show the variation of the bending moment caused by the edge shear in
the shell and the bending moment diagram in an arch under edge force subjected to a
distributed loading.
MA = bending moment in arch
MD = bendmg moment in shell
surfaee loa~
~ ~M"-O
./ ..
(a) an arch
~ ~
mendional toree "
hooPto22f~~0
~...
~~
d
- edge etteets
(b) a domleal shell
The overall membrane behavior of domes with or without rings is graphically represented in
figures (6-24). All these domes have distributed supports and are subjected to axisymmetric
verticalloading.
The arrows in the figure (6-24a) show the sign of membrane field in a high rise dome with
no ring and only a vertical support. Tbe tangent to the meridional curve at the lower edge is
assumed to be vertical. The meridional force remains compressive. but the hoop stress
changes sign. In this shell, the hoop tension is taken by the shell itself and the edge
meridional force is carried by the vertical support.
The membrane field (metidian and hoop stresses) in the low rise shell of figure (6-24b) is
totally compressive. At the edges of this sheIl, the inclined meridional force is carried
through the support, which is assumed to sustain vertical as weIl as lateral thrusting forces.
Tbe equilibrium requirements.of membrane behavior are satisfied for this shell.
Tbe support oflow tise shell of figure (6-24c) can only carry vertical forces. Therefore, the
horizontal thrust developed by meridional compression must be carried through the shell
itself mainly by thc mechanism of hoop action.Some tension will be induced in the lower
parallel circ1es, as demonstrated in figure (6-24c). This is obviously contrary to the
predictions of membrane theory; there must be a, some bending field in the lower part of
this shell to satisfy the equilibrium requirements.
The low rise shell offigure (6-24d) has a supporting ting at the edge together with a vertical
support; the hotizontal thrust is totally carried by the ring.
Conical shells, as the name implies, have as their middle surface a complete cone, a truncated
cone, or a sector of a conical surface. Conical shells are sheIIs of revolution with zero
Gaussian curvature. In spite of this feature, these shelIs are non-developable and thus offer a
great resistance to buckling and collapse. Conical shells are used in roofing and flooring of
liquid retaining tanks and reservoirs; as for the ends of pressure vessels; the nose cones of
spacecrafts and missiles; the roofing of astation or astand.
134 Design and Analysis of Shell Structures
In order to perform the membrane analysis of a conical shell, we consider the conical shell
of figure (6-25). One of the principal radii of curvature, i.e., r\, is infinite, and the meridional
angle, <1>, is constant. We define a new variable, s, as the meridional distance from the apex,
and introduce the following change of variables
4> =
(6-60)
r 2 = S cot4>
Using these new definitions in the membrane equilibrium equations, (6-6), (6-7), and (6-8),
we obtain the following goveming equations for conical shells
o
(6-61)
o
d
(N S S) - NE) - Ps s
ds (6-62a)
NE) P S Cot 4>
r
(6-62b)
Membrane Analysis of Shells of Revolution 135
~
ds
(N
5
5) =- (P
5
- Pr Cot ,p 5
Expressions (6-63) and (6-62b) give the closed fonn solution to the membrane force field
in axisymmetrically loaded conical shells.
As an example, we consider the mushroom-type conical shell shown in figure (6-25a). This
shell rests on a single column support at its apex. The loading consists of the dead weight
with intensity p. The load components are,
and therefore,
P L 2 _52
N
5
= ----
2 S 5in CI
(6-66)
Figure(6-26a) shows the membrane forces, Ne and N s. The meridional force, N s, becomes
infinite at the apex. This is what the membrane theory predicts. Actually, due to concentrated
support at the apex, there is some bending field at the apex. Figure (6-26b) shows the sign
of the internal forces in this shell.
6pf
(b)
If this conical shell had a continuous. rather than a point. support at its lower edge. figure (6-
27). then the expression for Ne would remain the same. but the melidional force would be
(6-67)
Figure (6-27) shows the sign of the internal forces for this case. The comparison of the two
figures (6-27) and (6-26b) is instructive in the sense that it demonstrates the influence of
supporting conditions on the conical shell behavior.
Membrane Analysis of Shells of Revolution 137
Problems
P 6.1 - Consider the pointed dome shown in figure (P 6-1). This dome has a continuous
edge support and is acted upon by its own weight The meridian is a circular arch of radius
a. The angle of normal to meridional curve at the shell apex with the vertical is $0. Hence,
the principal radii of curvature are,
q=a
and
r S~n <1>0
r 2= -
Sin-<p = a(l- s'n")
• 'I'
Derive the following expressions for the membrane forces in this pointed dome.
ep
N.= -
"
Pa
(Sinep-Sinep) Sinep
-- J
<10
(Sinep-Sin<p )dep
0
(Cosep -c8sep)-(ep-ep )Sinep
o 0 0
Pa .. .
(Sl.nep-Sl.nepo)Sl.n<p
Also, plot the membrane force field variation with the angle f, and thus reproduce the
diagrams shown in figure (P 6-1).
P 6.2 - A circular cylindrical grain silo is suspended from its top and has a conical bottom
(hopper) as shown in figure (P 6-2). The thickness of the shell is t and the unit weight of
shell material is g. The conical hopper is open at its bottom. Find the membrane forces in
both shells. Explain why a ring may be needed at the junction of the two shells.
P 6.3 - A conical concrete liquid retaining tank having a dome roof is shown in figure (P 6-
3). The tank is filled with water with specific weight of g, and the conical part is subjected to
a line load, q (N/m), from the dome roof above as weIl as being subjected to its own weight.
The unit weight of concrete is assumed to be gc. Derive appropriate expressions for the
membrane forces in this conical shell.
--===t===-..
//}:W
6.4 - Figure (P 6-4) shows a steel cylindrical pressure vessel with conical end caps. Using
the membrane theory alone, design the thickness of steel required to resist the internal
pressure of p = 4.0 bar. The yield stress of the steel is 4000 kg / cm2. Where may
additional strengthening be required.
6.1 - M. Farshad, Shell Structures, (in Farsi), Shiraz university Publications, Vol. I, 1986,
Vol. 11, 1987, Shiraz
6.4 - V. S. Kelkar and R. T. Sewell, Fundamentals of the Analysis and Design of Shell
Structures, Prentice-Hall,INC.,NJ., 1987
6.5 - D.P. Billington, Thin Shell Concrete Structures, McGraw-Hill Book Co., revised
edition, 1982
Bending Analysis ofAxisymmetric Shells 141
Chapter 7
7.1 . Introduction
By the term "axisymmetric sheIls", in this chapter, we me an all doubly curved as weIl as
conical shell forms which have an axis of symmetry and are loaded in an axisymmetric
fashion. Shells of revolution such as domes, single sheet hyperbolic paraboloids (used in
cooling towers), and conical shells fall in this category.
Domes, being doubly curved, are non-developable, sync1astic shells; they are the strong, stiff
and, stable. The membrane forces dominate the internal force field in a dome. Geometrical
and force incompatibilities will cause some bending field to develop, but this bending field
is loeal and is eonfined to the vieinity of geometrical variations, concentrated forces, and
membrane-incompatible boundary conditions.
In chapter six, we discussed the membrane theory of shells of revolution with positive
Gaussian curvature, inc1uding domes. In the present chapter, we develop a simplified
bending theory for axisymmetric shells subject to axisymmetJ;c loading. As a result, we will
obtain the influence coefficients which are useful in the force method ofaxisymmetric shell
analysis.
142 Design and Analysis of Shell Structures
Sometimes axisymmetric shells, and in particular domes, are provided with edge rings. The
analysis of domes with edge rings will also be carried out in this chapter. With such an
analysis, one can analyze combined shell structures composed of cylindrical walls with
dome-or cone-like roofs which are sometimes connected to each other by means of
stiffening rings.
The analysis scheme developed in the present chapter can be applied to the design of dome
roofs; cylindricalliquid retaining shells with dome roofs; pressure vessels; containment
shells of nuclear power plants; as weIl as other types ofaxisymmetric shells.
The equations of equilibrium consist of three force equations and three moment relations.
Due to axisymmetric conditions, and assuming the applied load in e direction, Pe, to be
zero, the force equation of equilibrium along the e direction is satisfied identically as are the
moment equations of equilibrium about the rand 4> directions. We are left with three
equations of equilibrium which we will now write down.
Bending Analysis ofAxisymmetric Shells 143
Because the shell has double curvature, the internal forces have projections in all directions.
For example, when writing down the equilibrium of forces in the cj> din~ction we should take
into consideration the contribution of the shear force, 0cjl. as weIl as the membrane forces,
Ncjl and Ne. Figure (7-2) shows the projections of membrane and the bending shear force in
the cj> and r directions.
"
"
~~~/
1,,/o..;dQ/
I
(a)
The force equations of equilibrium are, with the help offigure (7-2), written as folIows:
d
(N~r) - Nor1cos.p- r Q 4> +rr 1 P.p = r: (7-1)
d.j.
d(Qq,rl
+ Nor 1 Sln 4>+ ~ - rr 1 Pr = 0 (7-2)
These equations have a more general form than the membrane equations which were derived
in chapter six. For instance. the load-bearing contribution of the bending shear force. 0cjl.
appears in the equilibrium equations.
"
-.../
'\
\~.
I \
I \
\
To derive to moment equation of equilibrium about the e axis, we make use of the figure (7-
3). This figure shows the spatial configuration of shell element and the projections of
bending as weH as twisting couples. Some useful relations among the geometrical
parameters represented in this figure are
AB
AB
du= Cos-jldO
The moment equation of equilibrium for the shell element about the e axis is
o (7-3)
Equations (7-1), (7-2), and (7-3) constitute three relations among six unknown force
quantities N$. Ne. M$. Me. and Q$' This means that the shell is statically indeterminate and
three more relations are needed to find the internal forces. These additional relations are
provided by the kinematic and constitutive equations.
Kinematic relations relate the strains and the change of curvature with the components and
gradients of the dis placement vector. In this axisymmetric case. the displacement
components are the displacement normal to the shell sm'face, w, and the displacement
component along the meridian, v. These quantities are. in general, functions of meridional
parameter, <1>. Due to axial symmetry, the hoop displacement is identically zero. In chapter
six, we have already obtained the following strain-displacement relations.
v w
- - cot <j> - r (7-4)
r 2
:2
as shown in the figure (7-4). In this figure, AD and A'D' represent the undeformed and
deformed meridional sections, respectively.
The angle which the tangent to the meridian at A makes with the initial direction of the
tangent line is designated by a. This change in meridional tangents is brought about by
both v and w displacement components. Figure (7-5a), shows the effect of meridional
displacement on the infinitesimal rotation of the tangent. This part of rotation is designated
by 01 = v I r1. Figure (7-5b) shows the effect of normal displacement on the angle of
meridional rotation. This contribution is easily seen to be equal to 02 = (dw) I (r1 d<\l). The
total angle ais
v d'N
-+
r 1 r1d<ji
(7-5)
The angle between the meridional tangent and the tangent for a neighboring point D, figure
(7-4), is simply equal to that of point A plus the differential change in a. Thus, if we
designate this angle at point D by ß, we have
r (7-6)
Figure (7-4) shows that the central angle of the deformed element, A'D', is (d<!> + ß - a) and
so its length is equal to
A'D' = ri (d~ + ß - a)
in this relation, r'1 is the meridional radius of curvature of the deformed element. If we
neglect the second order effects of membrane stretching, we may assume AD = AU', so
r :;d<P = ri (d~ + ß - a)
hence,
_1__ _ dp + ß - a
r i - r ld<P
1 I (8 - ('()
(9 r' r l r 1 d<jl
1
(7-7)
d (~ ~)
+
'<1> rId<jl r l r1d<jl
To determine the change in the radii of parallel circles, i.e., the curvature change in the hoop
direction, we consider the shell element shown in figure (7-6); the initial and defOlmed radii
of curvature of the hoop element, AB, are designated by r2 and r'2, respectively.
Because of the axisymmetric behavior of the shell, the radius of curvature has the same
rotation at A and B, namely the angle a. From figure (8-6), we can easily deduce the
following relations:
AB : r; Sin (~ + a)d9
If we neglect the second order effect of the membrane change of length, we may write
1 1
X0 = r' r 2
2
Substituting for a from relation (7-5), we find the expression for the change of curvature in
the hoop direction
(7-8)
Relations (7-4), (7-7), and (7-8) constitute the kinematic relations for axisymmetric behavior
of shells of revolution. The strain components (in two perpendicular directions) in a shell
layer located at a distance z, from the middle-surface, can generally be expressed as follows:
E * E - zx
x 1 1 (7-9)
E* E - ZX
Y 2 2
For shells of revolution, the indices x and y, are usually interchanged with <p and 9.
Moreover, very often the notations X1 = Xcp and X2 = Xe are used interchangeably for the
redundant forces in the shells of revolution.
148 Design and Analysis of Shell Structures
* =_E_ * +VE *
0.p 2 (E.p
o )
I-v
tI
(7-10)
* E * .
oe = - - 2 (E e +VE.p)
I-v
Now, if we substitute these expressions into the relations (2-1), relating internal stresses to
internal force resultants, and use the kinematie relations (7-4), (7-7), (7-8), we obtain
Et 3 [1 d v dw v v dw
M.p - - ( - + - - ) + - ( - + - - ) c o t 4>] (7-11e)
12 (1_v 2 )
r 1 d4> r 1 r I d4> r 2 r 1 r1d<jl
(7-11d)
These are the axisymmetrie eonstitutive relations for linearly elastie shells of revolution. It is
to be noted that in deriving these eombined eonstitutive-kinematie relations, we have also
used the kinematie relations for axisymmetrie shell behavior.
(7-12)
Bending Analysis ofAxisymmetric Shells 149
Where t is the shell thickness, and E,v are the elasticity parameters.
From the physical point of view, D signifies the membrane stiffness and K designates the
bending stiffness of the shelL
The relations (7-12) show that the bending constitutive relations can be decoupled from the
membrane constitutive relations. This conc1usion follows the assumptions that we had made
regarding the equality of undeformed and deformed lengths of the shell elements.
Relations (7-1), (7-2), (7-3), and (7-12) form the governing equations of linear elastic shells
of revolutions with axisymmetric behavior.
If we substitute the constitutive-kinematic relations (7-12) into the equilibrium equations (7-
1) to (7-3), we obtain three simultaneous differential equations for three unknown functions
v, w, and Q<\l. We could eliminate the variable Q~, signifying the lateral shear force, from
these three equations. The result would be two simultaneous differential equations for the
two displacement components, v and w. At this stage, it is more convenient to introduce the
following new variables:
* 1 dw (8-14)
V=-(v+-)
r 1 d,j>
(8-15)
u* = r Q
2 <P
The variable V* signifies the rotation of tangent to meridional curve i.e., the angle cx. The
variable U* is directly proportional to the shear force, O<!J.
T
Figure (7-7) Base forces on a horizontal section ofaxisymmetric shell
150 Design and Analysis of Shell Structures
If we write the equation of equilibrium of the segment shown in figure (7 -7), we obtain
N - Q", Cot .p ~ - -1 U
* Cot .p (7-16)
<P 'I' r 2
d(Q r)
r
1 0
N Sin tjJ =- N r
:j>
-
dcj>
p
1 d 1 dU *
(7-17)
N
o = - -r 1 -dljJ (Q r) = - -
cj>
-
r 1 d:j>
We have thus succeeded in expressing the membrane forces in telms of new variables, U*
and V*. In order to write the governing equations in terms of these new variables, we
proceed as folIows:
dv r1 (7-18a)
w = - vN )
r,+
(N -
d<j> Et 0
(7-18b)
v Cot .p - w = E~ (Na - VNtjJ)
To eliminate w between these equations, we subtract the second from the first to obtain
(7-19)
(7-20)
Bending Analysis ofAxisyrnrnetric Shells 151
We can easily eliminate the tenn dv/d<)l between (7-19) and (7-20). The result is
v + ciw = r
Li<j>
v' = ~()tcP
1 Et
[(r1+vc,lN
_",
- (r~+'JrllN)J
~ f
(7-21)
I r ~
- ~
,14,
[---,,-
L:t
(tJ,-')!l
tJ l'
l]
If we substitute the expressions N<)l and Na, from relations (7-16) and (7-17), into these
equations we obtain the following:
(7-22)
This equation is one of the two desired relations between U* and V*. To obtain another
goveming equation, we substitute for M<)l and Me from relations (7-11c) and (7-11d) into the
equilibrium equation (7-3). By doing so, we obtain the second relation in the following
fonn:
1
d
r-(-=.)
r')
T
r?
-=. cot <j> +3
r2 dt
1 dV *
r 1 'd4> r 1 r 1 r 1t ~ d<j> (7-23)
1 3vcot<j> dt r1 2 u'
,,' ____
( v- -+ - cot <jJl ,
r 1 t d<j> r2 K
We have expressed the goveming equations ofaxisymmetric shells as two relations (7-22)
and (7-23) in tenn of new variables U* and V*.
Equations (7-22) and (7-23) are valid for thin axisymmetric shells with varying thickness.
For a shell with constant thickness, we have dt/d<)l = O. In such a case, the differential
operators in the left-hand side ofthose equations become identical,namely
L( •.• l
(7-24)
With this notation, the two basic equations (7-22) and (7-23) can be written in more
compact operator forrns as follows
L(U * l + - v U* = Et V* (7-25a)
r 1
U*
K
(7-25b)
152 Design and Analysis of Shell Structures
This pair of second order differential equations can be combined to yield the following
fourth order differential equation with <I> as its independent variable. Operate on (7-25a) by
L to obtain
*
LL(U*) + vL(Q....) EtL(V * ) (7-26)
r 1
* v V*- -U= - - (
*
v L[ ( U*) v u *] - U
*
L(V) =- +- -
r 1 K r lEt r 1 K
• Et *
LL(U*) + VL(~) - ~ L(U*) - - U (7-27)
r1 r 1 K
The following fourth order equation on variable V* can be obtained in a similar fashion.
LL(V * ) vL(-)
v* +v- *
L(V )
Et
- v* (7-28)
r 1 r 1 K
These equations are valid for constant thickness axisymmetric shells having arbitrary
meridional curves. For special shell types (for example spherical shells, cones, and toroidal
shells), in which the meridional radius of curvature (rl) is constant, we will have the
following:
L(..!:!...) = J:... L(U)
r1 r 1
4 Et (7-29)
j.l = --
K
Operationally, this differential equation can be decomposed into the following components:
Bending Analysis ofAxisymmetric Shells 153
Therefore, the problem of bending analysis ofaxisymmetric shells with constant thickness
and constant r1 reduces to solving the equation (7-32). As we shall see, the solution to this
problem has a broad range of applications; it can be applied to edge-effect analysis of other
axisymmetric shells with arbitrary meridional shapes.
t vertical reaction
Calculations based on more exact theories and experimental results show that the influence
of boundary conditions dies out rapidly away from the edges. In fact, at a distance located at
approximately 20° away from the edges, the shell "does not feei" the bending effects
brought about by the edge forces. The membrane force field predominates in the inner part
of the shell.
The set of figures (7-9) show the spatial variations of internd forces and shell deformations
due to distributed edge shear and bending moment. These results could be quantified by
bending analyses and I or actual measurements of shell behavior. As we see, all internal
forces and displacements due to edge forces "damp out" rapidly away from the boundary
region.
154 Design and Analysis of Shell Structures
.---------
N, N. M. M. Q t;.r
f1
~ l J l ~ J lJ
- +
.'20·
--H
1~ T l l t ~ 1J
- +
200
,":' ""
..ii.
Even in the edge zone differences in shell configuration have little effect on the bending
behavior. This means that we can determine this behavior approximately by using, say, a
spherical shell, figure (7-11); we choose spherical shell because it is easily analyzed. Figure
(7-9) shows the implication of this argument and its application in practical shell analyses.
$-
I
t
.-20· I
$"f(.f.
(a) (b) (c)
spherical shell ,
actual shell _____ _
(7-34)
In Geckeler's approximation, of the above equations, only the functions themselves and their
derivatives (gradients) of higher order are retained and the lower order derivatives are
discarded. Recall that the functions V* and ~ represent meridional rotation and lateral shear
force, respectively. The GeckeIer approximation is based on the physical argument that the
boundary effects are localized and thus the higher order gradients are of greatest
importance. Hence, in each equation, only the highest order derivative of the main function
and the zero order derivative of another, coupling function, shall be retained.
By eliminating the variable function V* we obtain a fourth order differential equation on the
function Q<!l'
o (7-36)
in which
4 2 a 2 (7-37)
\ = 3 (1 - v ) (tl
156 Design and Analysis of Shell Structures
We recall that the coordinate parameter <)I was originally taken as the angle between the
normal to the shell surface and the vertical direction. Thus, the lower edge of the shell would
be identified by the value <)I = 0:. For shells with convex meridional curves <)I would decrease
towards the shell apex.
The last two terms in the general solution (7-38) increase as <)I increases,i.e., as we move
away from the edge. But we argued that edge effects must decrease as we move away from
the edge. Therefore, we set C3 =0 =C4, and obtain
(7-39)
For convenience, we introduce a change of coordinate \jf = 0:-<)1, figure (7-12), and replace
the integration constants Cl and C2 by two new constants C and y. With these alterations,
the general solution (7-39) can be rewritten
-A1jJ
Q = Ce Sin(A1j! + yl
~ (7-40)
Using the relations (7-15) to (7-17), we can now write down the expressions for Y*, Nep, and
Ne
(7-41)
(7-42)
To obtain the expressions for bending moments Mep and Me, we use the relations (7-11c),
(7-11d), and (7-14), set rl = r2 = a and neglect y* compared with dY* I dep. We obtain
*
_ ~ dV =~ cc- AljJ SinC\ljJ +y+ 4 )
1f (7-44)
MIj> a dlj> *>..{2
For our future use in the analysis ofaxisymmetric shells, we also need the expressions for
shell rotation and displacement. The horizontal displacement, i.e., the change in radius of
parallel circles is
(7-46)
If we neglect the term containing U*, compared with dU* I dcp, we will have
Sinej> dU * a
- Et dcjl = Et Sin (Cl -1jJ) x (7-47)
)'/2 ce->"w sin (),ljJ + y -~)
4
Relations (7-41) to (7-47) approximate the bending field of forces and displacements in a
spherical shell subjected to edge shear and bending moment. All of these fields have the
decaying form that was demonstrated qualitatively in figures (7-9).
158 Design and Analysis of Shell Structures
As a useful by-product of this analysis, we now obtain the flexibility influence coefflCients
ofaxisymmetric shells, i.e., the displacements due to unit edge forces.
f6~
tr
(a)
!'~
(b)
6H sign convention
Consider figure (7-13b) in which the shell is acted upon by a unifOImly distributed edge
moment Mn. The boundary conditions are:
(7-48a)
(M ) =M
.p <V = a Cl
(N ) =() (7-48b)
tp <V = a
If the condition (7-48b) is used in(7-43) it gives Y= o. Substitution of (7-48a) into (7-44)
gives
c a
Having obtained the two integration constants C and y, we can write down the expressions
for internal edge forces and edge displacements due to the edge moment Mn. These values
are tabulated in the third column of table (7-1). Specifically, for Mn = 1.0 we shall get the
bending moment flexibility influence coefficients.
Bending Analysis ofAxisymmetric Shells 159
As another basic solution, we consider the shell of figure (7-13c); the edge conditions for
this shell are
=a
o
(7-49)
- l! Cos n
=a
A = _ 1T 211 S ina
C
4'
Having found these constants, we can write the expressions for intern al edge forces and
edge displacements due to a distributed edge shear force, H. These results are tabulated in
the second column of table (7-1). Again, for H = 1.0, these expressions give the flexibility
influence coefficients of the shell due to a unit edge shear force.
}I"-~~ I L~Ma
I
N.p -l2cot (u-lji) Sina e-\~'x II -2 A
-
a
cot (u-~') e
-\tj>
x
M
a
J\ Slnae
- \~)
sln().Ij;)H 12 e -l,ljisin(AIj;+ ~)M
.p 4 CL
2
.!a;\sin 2).2 s1n r.t
II 11
Et
" l!
Et
M
Cl
4).3 M
I\a 2\2 sina a
Et
H
, --
Eat
I
160 Design and Analysis of Shell Structures
The force method ofaxisymmetric shells analysis consists of: (1) the membrane analysis,
(2) the bending analysis for the edge and / or boundary effects, and (3) superposition of
membrane and bending analyses to satisfy the compatibility requirements.
In the force method, the redundant unknown quantities are the bending edge forces and / or
the forces between two shell segments. Once these forces are determined, the other
unknowns such as internal forces and shell displacements can be obtained by superposition
of the membrane and the bending fields. The flexibility influence coefficients are useful for
deriving the compatibility relations and for determining the unknown redundant edge forces.
In the following section, a sam pie example will be worked out to show the force method of
axisymmetric shell analysis and also to demonstrate the relative quantitative contribution of
the edge effects to the total field of forces and displacements.
v 1/6 = 0.167
2
q 440 Kq/m
r a Sin a = 13.35 m.
Solution:
Let us designate the redundant edge shear force and edge bending moment by the symbols
Xl and X2, respectively. In this problem, these are the ben ding constraints which are needed
for deformation compatibility and are to be determined by the application of the force
method.
a2 1+v .
D = ~ - Cos Cl. ) S ln Cl.
10 Et I+Coso.
(28.4)2 xq (1+0.167 -Cos28)Sin28=- 997..9...
D10 = E x 0.1 I+Cos28 E
D =~(2+v) Sina
20 Et
28.4q (2+0.167)Sin28 289 l
D20 = E x 0.1 E
Here, D10 is the membrane horizontal displacement ofthe edge parallel circle and D20 is the
membrane rotation of meridional curve at the edge of the shell. Figure (7-15) shows the
schematics of membrane deformation and adopted sign conventions.
(a)
To find the bending field of action, we first calculate the value of parameter A which was
defined by relation (7-37). Its numerical value for this problem is,
4/ _)
). = V 3 (l-v:l) (~)- = 22
t
The influence coefficients necessary for bending analysis can be calculated by the help of
table (7-1). If the i-th deformation component due to unit edge force Xj = 1.0 is denoted by
Dij, then the appropriate flexibility influence coefficients are,
2a:\Sin 2 a 2754
Dll = +--
Et E
", ' 1 , {
2).2 sina 4544
D21 = +--
Et E
'{ ",:
4544
0 21 = + - -
E
X
2 = 1
4).3 14997
D22 - --
Eat E
The compatibility relations, requiring zero horizontal displacement and zero meridional
rotation at the edge, are as folIows:
= - 113 Kg-mjm
X 2= - 0.258 q
The negative sign in the value obtained for the edge moment indicates that the actual
direction is contrary to wh at was assumed in the beginning of this problem. We also note
that the edge shear force, Xl, comes out to be positive. This means that the membrane shell
has a tendency to move inwards and the effect of edge shear is to bring it back outwards,
Bending Analysis ofAxisymmetric Shells 163
so that the actual edge conditions can be satisfied. This result differs from what we would
have expected from the analysis of a planar arch.
Once the redundant edge forces are found, we can use relations (7-42) to (7-47) to
determine the internal forces and displacements due to edge effects. These calculations are
summarized in tables (7-2) and (7-3).
Table (7-2)
, i: ':::;t (:!-1'~I
\~I re - lo:~)" '0/1 c
.s.:.n(i,~'- ~) o:n()~··4)rH(·'~f)
I -
degree
'I' f
radla1l degree radian
'Y' Sln ::t
I
;.-
I u :.:'8 0 0 1. 000 0 0.707 b.707 1.000 0.4695
I "
5 0.0873 2.36 110 1.92 0.147 0.940 0.906 0.423 0.342 0.4695
10 o. '- ~ 45 I :;. :8 220 3.84 0.021 -0.643 0.097 -0.996 0.766 o.4E95
I I I
I
,I -" o. 34~: I - -" I 440
I
7.68 0.00041 0.985 0.574 0.619
iI- o . 174 0.4695
I :a 10 .• 667! . i
I elei I
10. '7sl 0 -0.970 -0.515 -0.854
I! O.
I
~42
I O.4E95
Table (7-3)
~
x: coefficient
N~ N8
29 J 0 0 0 0 0
Using the results of part (1) and part (2), we can present the bending and the membrane
field of internal forces in the shell for a number of meridional angles, as shown in table (7-
4)
Table (7-4)
T Kg/m 2
4.N/ m(Q;440 ) -6.33 -6.23 -6.18 -6.28 -6.26 -6.25 -6.25
TI Kgl m2
8.N e ~Q;440 ) -1.12 -2.24 -3.66 -5.82 -6.3S -6.25 -6.25
9.M membrane)T-m 0 0 0 0 0 0 0
1>0 m.
T-m
10.M1> (lI=xi - - 0 +0.05 +0.07 +0.03 0 0 0
1 m
T-m I
110M (M='2)- -0.11 -0.10 -0.07 -0.01 0 0 0
4>2 m
T-m
12.M.p - -0.11 -0.05 -0.00 +0.02 0 0 0
m
Using the numerical values of table (7 -4), we can plot the variations of the internal forces in
the shell on the meridional curve. The set of figures (7-16) show variations of meridional
force, the hoop force, and the bending moment in the shell of figure (7-14). In these
diagrams, the membrane field of forces is plotted and compared with the results of a more
comprehensive bending analysis.
Bending Analysis ofAxisymmetric Shells 165
meridional force
I
_---·-'.2~ Tim
Ne hoop force
... I
---~J 0.0 T.m/m
M$ bending moment
As we see, the influenee of edge forees is quite loeal and damps out rapidly as we move
away from the boundary zone. At an angle of about 5° the effect of edge constraint is quite
insignifieant
The maximal values of internal forces could be determined from table (7-4). The maximum
internal force quantities and their locations are as folIows:
(N ) -6.33 Tim. at 1jI=00 T-m
q. max M 0.11
q. m
(Ne)max -6.25 Tim. at ljI =28° Mq. 0
Problems
P 7.1 - Consider A conical elastic shell with a central angle a, the base radius a, and
thickness t. The lower edge of this cone is fixed against rotation and translation. This shell
is subjected to its own weight with intensity q. Perform a complete analysis of this shell.
P 7.2 - Consider the spherical tank of figure (6-11). Determine the bending field which
would develop around the supporting ring of this shell.
P 7.3 - The presence of the upper ring in the spherical shell with the top opening of figure
(6-9) is expected to produce some bending field in this region. Even in the absence of the
ring a bending field may be produced around the free edge. Verify if there is a bending
field there. If so, determine the bending field at the top of this shell.
P 7.4 - Find the bending field in the interseetion of the cylindrical wall and its conical
hopper of the silo structure shown in figure (P 6-2).
P 7.5 - Consider the conical water tank with a domed roof of figure (P 6-3). Perform a
complete analysis of this shell structure.
P 7.6 - Determine the bending force field between the cylindrical wall and the conical end
of the pressure vessel shown in figure (P 6-4).
Bending Analysis ofAxisymmetric Shells 167
7.1 - M. Farshad, Shell Structures, Vol. I, 1986, Vol. 11, 1987, Shiraz University
Publications,shiraz
7.2 - D. P. Billington, Thin Shell Concrete Structures, McGraw-Hill Book Co.N.Y.,
Revised edition, 1982
7.4 - S. Timoshenko and S. Woinowsky-Krieger, Theory 0/ Plates and Shells, 2nd edition,
McGmw-Hill Book Co., N.Y., 1959
Design of Reinforced Concrete Domes 169
Chapter8
8.1 - Introduction
Domes have synclastic shell surfaces with positive Gaussian curvature. They are strong and
structurally stable. Dome roofs can be constructed from steel, various fiber reinforced
composites, and reinforced concrete materials. Precast shells made of these materials have
also been constructed successfully.
Reinforced concrete domes are used to cover large spans of stadiums, memorial buildings,
meeting halls, and other large assembly halls. They are also used to cover the roofs of
liquid retaining structures, silos, as weIl as the roofs of containment sheIls of
nuclear power plants. The end caps of concrete containment structures and press ure
vessels are also made of these axisymmetric shells. Parts of shells of revolution and I or
various combinations of these shell types can also be conceived and designed.
A dome is often provided with an edge ring at its lower edge and / or with a ring
somewhere along its parallel circles. Figures (8-1) show some cases in which edge rings
are used with domes.
I
---- -~-:..-:.. :...
I
(a) dome roof wlth ring
Edge rings stiffen the shell and / or provide lateral support for the shell structure. The
lateral support action of the rings is specially needed in cases where there are only vertical
supports and thus the lateral thrusts are to be absorbed by the structure itself. For
combined shells, figure (8-1b), the stiffening ring between two shells acts as a
strengthening member which absorbs part of the bending field created by the curvature
change from one shell to the other.
In metallic and composite shells, rings are needed to strengthen the shell against buckling.
They are also required to distribute the line loadings and to transfer the shellloads to the
supporting elements.
Edge beams in a shell structure create some bending field in the vicinity of the ring. This is
due to the difference in stiffness between the shell and the ring and the ensuing violation of
the membrane assumptions.
From the structural analysis point of view, a force field composed of shear force and
bending moment as weIl as membrane forces would exist between the shell and its edge
beam. The magnitudes of bending effects would be such that the defonuation compatibility
requirements are satisfied. These forces of dome-ring interaction are shown in figure (8-
2).
Design of Reinforced Concrete Domes 171
{
I
i----
I
Figure (8-2) Interaction between an axisymmetric shell and its
edge ring
The generalforce method of structural analysis, outlined and used in previous chapters, can
be used in the analysis of domes with rings and also domes with cylindrical walls. The ring
in a dome acts as a tie capable of absorbing the horizontal forces; it is a deformable body
integral with the shell and must be analyzed. along with the shell.
A dome-ring structural system has its counterpart in linear framed structures. A dome with
a ring resembles a framework having a tie at its lower base. The tie in the "dome-ring"
structure can act in tension as weH as in compression.
In order to understand the combined behavior of a "dome-ring" system, we use this
analogy. In doing so, we utilize the knowledge and elementary analysis of framed
structures. We shall therefore first discuss the force method as applied to a simple frame.
, q
~ C 0 0 ~
I.
A (d) 8
X,_--.....;..,.-'-_..,...,..r-.
~-1=
X,
Figure (8-3) Force method of analysis applied to a POltal frame with a tie
172 Design and Analysis of Shell Structures
Figure (8-3a) shows a portal frame with a tie. A general approach to analysis of such a
frame is depicted by figures (8-3b) and (8-3c). This frame is statically indeterminate having
one degree of indeterminacy. The unknown force in the tie can be chosen as the redundant
force. Figure (8-3b) is the associated statically determinate frame, Le., the frame with the tie
removed. The magnitude of redundant axial force in the tie is such that the compatibility
requirement between the tie and the frame is fulfllied.
The statically determinate frame is structurally analogous to the membrane shell with the
edge ring removed. Figure (8-3c) shows the influence of a unit horizontal thrust. Finally,
figure (8-3d) shows the effect of tie deformation on the tied-frame behavior. These two
latter figures correspond to the effect of edge forces, Le., the edge ring on the shell
structure.
The compatibility relation for the tied-frame of figure (8-3a) is
or
giving
(8-1)
f
-0 0
xl
of + Df
1 1
(a) (b) (c)
(d)
R
°20 D
OD
21 °22
0:' 022R
Oll
°ll
TI.
/2
Figure (8-5) shows the details of the decomposition of the dome and its related
deformation parameters. Figure (8-6) demonstrates the decomposition scheme of ring
analysis as weH as the deformation parameters to be calculated in the course of the ring
analysis.
This analysis can be applied to a shell with a ring. Figure (8-4a) shows a dome with a ring.
Figure (8-4b) depicts the same dome without the ring, acting as a (statically determinate)
membrane shell. The membrane deformations in the dome and the deformations in the ring,
due to membrane forces, are also defined alongside this figure. Figure (8-4c) demonstrates
the edge forces and corresponding flexibility influence coefficients related to the dome.
Finally, fi~ure (8-4d) shows the ring together with related influence coefficients. In all these
figures, D ij and DR ij refer to influence coefficients related to dome and ring, respectively.
The corresponding membrane deformations are denoted by DDio and DR io •
l
~
l
' /
)(2
X,
~~t X,
Xl
x,
N. N. 0 0 H. 0 0 0 0
0 '0 J 0'0
•
0 11' 0 1'-- 0 12 J
0 12
a
'\e
_
@ .
Ncx.e
""COSex
+ .
;1-YOXI
XI
0" "
'0 ,0 10 Oll11 J
R 11
O,t=OZI ' D1112
Based on the foregoing discussion, we may now state the stages of any "dome-ring"
analysis problem as foHows:
(1) Analysis ofthe ring under hoop force, unit radial force, and unit torsional couple.
(2) Membrane analysis of dome for distributed forces as weH as bending analysis of
dome for unit value of edge effects.
(4) Superposition of membrane and bending effects to find the total force and
deformation in the dome-ring structure.
x
y (b)
[!J~~h
+-
(c)
_ : • nng cross section
J
Figure (8-7a) shows that the hoop force and the hoop stress are:
T = H.r a
T
=-
(8-2a)
6 AR
(8-2b)
Design of Reinforced Concrete Domes 175
t. e = 21Tr. E e
21Tr 2nr (8-3)
t. e = EA • T = - Hr
R EA R
Therefore, the change of radius of this ring due to a radial force is expressed as follows:
t.H = ~. r
_ Ae
t.H =- X r
(8-4)
21Tr 2
t.H = -rT = -~ H
BAR EA R '
Now consider the free body diagram of half ring shown in figure (8-7b). The equation of
moment equilibrium about the x-axis is
so
M (8-5)
x M .r
II
o
Figure (8-8a) shows the deformation of a seetion of this ring under the action of twisting
couple Mn.
6Mx
-b"hl
(a) (b)
Referring to figure (8-8a), and using the classical formula for ben ding of beams, we find
the bending stress and corresponding hoop strain and change of ring radius as folIows:
MXY
o = (8-6)
IR
MxY M . - h/2 6M
* x x
E*
e (s e) max I = =--
EI R y=+h/2 3
Ebh /12 Ebh 2 (8-7)
nr
. -6r 2M
-r (ce) max = - - -
a
Ebh 2
Note that c1assical beam theory gives a bending stress which varies linearly with the height
of a ring section. This is shown in figure (8-8b).
Assuming linear variation of torsion-induced hoop strain in the height of the ring section,
and using simple bending stress formula, we find the torsion-induced change of ring
perimeter to be
(8-8)
(8-9)
Due to difference in the radius change, each section of the ring would undergo the
following torsional rotation:
M (8-10)
"
Relations (8-4) and (8-10) give the radius change and cross-sectional rotation of the ring
under the unifonnly distributed radial force Hand twisting couple Ma . These relations are
rewritten as folIows:
Cl
2
-r 12r-
c\1! = Ebhl! --M (8-11)
3 (i
Ebh
Design of Reinforced Concrete Domes 177
Mo<
c:- r
o o
0 0
(a) (b)
4 ? a 2
\ = 3 (l-v-) ( t )
Figure (8-lOa) shows part of a "dome-ring" structure resting on a vertical support. Figure
(8-1 Ob) shows the forces of interaction between the dome and the Jing.
r
h
1 (a)
, tv r
(b)
Figure (8-11a) shows the membrane field of "dome-ring" interaction. The deformations
caused by these sort of interaction and the adopted sign convention are shown in figures
(8-11b) and (8-11c), respective1y.
Figure (8-11a) shows that the components of meridional force at the base of the shell are
The vertical component is absorbed by the vertical support while the horizontal component
is taken by the ring. The radial displacement of the ring due to this horizontal component is,
2 2 (8-14)
,\ _r_·_ H = _r_ N Cos a
oH EA R 0 EA R a
In the general case, the meridional force acts on the ring seetion with an eccentricity. Thus,
assuming an eccentricity of e, we find that the torsional couple, induced by the membrane
force, acting on the ring is Moa. = Na. e. The radial displacement of the ring due to this
couple, derived in the previous section, is
,
2:.......'L '1 e
EI
k
..( (8-15)
Design of Reinforced Concrete Domes 179
We seek the radial displacement of the ring at the "dome-ring" junction. At this point, we
have
Y h -d'
o 2
in which
t
d' ~ ~os a
Since d' is usually very small, we may use the approximation Y0 =h / 2. Therefore, the total
radial displacement of the ring is,
2
R r r 2Y oe
AR
H
DlO~ - - Cos
EA
Cl + EI} N
Cl (8-16)
R R
r 2e
AR
'1
R
D 20 -N
EI R '1
e.... (8-17)
For a ring with rectangular cross seetion, the above relations hecome
L2Y e
r 2N 'l
"10
()
[) ~ \\=-os '\ t - - . , - )
!:LJh
(8-18)
, II
1 'r - e N,!
Il~O
P-
Lbh
1 C (8-19)
For example, if a spherical dome is acted upon by uniform dead weight of intensity q. then
as we know, from chapter 7,
As another step in "dome-ring" interaction analysis, we subject the dome and the ring to the
edge forces H =Xl and Ma = X2, separately. Figure (8-12) shows that the ring deformation
due to force XI, applied at Y0, is
180 Design and Analysis of Shell Structures
2
"'ll! = -E-x
EA R 1
r 2Y o 2
1I2H = - - X (e y )
EI R 1 o
2
II = _ _
r_ y X
Cl EI R 0 1
Figure (8-12)
,. 2Y OX7
llu = - - - -
EI R
SO, for X2 = 1 we have
R r 2y o R (8-24)
0 12 = - EI = 0 21
R
2
R r (8-25)
0 22 = EI R
Therefore the ring influence coefficients, i.e, the ring deformation for unit radial force and
unit twisting couple, observing the sign convention of figure (8-13), are
(8-26)
(8-27)
2
°R22 =--
12r
Ebh3 (8-28)
Figure (8-13)
Design of Reinforced Concrete Domes 181
At this stage, we are prepared to combine the influence coefficients of the dome and the
ring to determine the influence coefficients for the "dome-ring" system. The system
influence coefficients are,
D R
D ll = Dll + Dll
D k
[) L~-~ D12 + D 12 f) (8-29)
21
D R
f)22= Dn + Dn
In the following relations, we adopt the sign conventions shown in figure (8-14).
---t--+_6H
(a) (b)
Figure (8-14) Sign conventions, (a) for the ring, (b) for the dome
(8-30)
t,D~ Do,O= ~ IN ,~ (r 1
+'Jr ) - NO (L"?+vr1l]
:: _
u _ L:t (8-31)
*
(8-32)
(8-34)
(8-35)
__ aq_
cl 1 +('os cl (8-36)
(b)
(8-40)
(8-41)
(8-42)
,
o
......
/ \" \
/ \ \
-
/ / \
I.... / \ \
'- I \..... ....
....J R
D,~ °'2
(a) (b)
Figure (8-17) Positive sign convention for the ring influence coefficients
Having obtained al1 necessary influence coefficients and membrane deformations. we are
now prepared to app1y the final relations of the force method to the "dome-ring" system.
We must satisfy the compatibility relations which express express the continuity of radial
displacement and rotation at the "dome-ring" junction. These are
(8-43a)
(8-43b)
Using the parameters defined in relations (8-29) we write the compatibility relations as
(8-44a)
°12 x2 + 10
°llX 1 ~ D ~
0 D R
°lC/ 0 10 + 0 10
+ D22 X2 + 0 20 0 I< (8-44b)
°12 Xl
~
°
°2U= 0 20 t D20
184 Design and Analysis of Shell Structures
By solving these linear simultaneous algebraic equations, we determine the two unknown
redundant forces Xl and X2; they are
°22°10 - °12°20
xl H (8-45a)
°22°11- 0i2
°ll °20 - °12°10
x2 M= -
a 2 (8-45b)
°22°11 - °12
The final step in the "dome-ring" problem is to combine the bending field induced by these
forces with the membrane field.
Domes have doubly curved, sync1astic, nondevelopable surfaces, and are generally very
stable. Nevertheless, the buckling must be considered in the choice of dome thickness.
Construction requirements also play a major role in the choice of shell thickness for thin
reinforced concrete domes.
In chapter 13 we discuss the buckling of shells, and give formulas for the buckling loads of
domes. We can use those data in the actual design of reinforced concrete domes.
The relative dimensions of concrete domes vary with the particular situation, service
conditions, and the means of construction. We present so me general guidelines for the
initial design and quantity estimates related to concrete domes.
Figure (8-18) presents a design chart representing the appropriate rise to span ratios of
concrete domes. Table (8-1) gives the ratios of the span, thickness, central angle, rise, and
radius of spherical domes.
Design of Reinforced Concrete Domes 185
1
2°,5 - I I
I I
I T
I
04
1
J
03
,
I
4
1 ,
~2
,
~il :tJrD
'
~ I'
6
h1
I
- - 0,1
1 8
I0
Rise _
h i
l =-- T
Span 30 &0 ~o
120"' l = Span of the shell
D t a f a
section meter cm deg meter meter
r f7\
30 0 7 5 30 4 02 30 0
45 6 22 21 22
40 0 7 5 30 5 36 40 0
45 8 29 28 29
Lfj
45 · 0 9 0 30 6 03 45 0
(7.5) 45 9 32 31 82
55 • C 10. 0 30 7 37 55 0
(9.0) 45 1l. 40 38 90
60 • C 1l. 5 30 8 . 04 60 0
(10.0) 45 12. 43 42 43
186 Design and Analysis of Shell Structures
In this seetion, we present the design of a reinforced concrete shell roof composed of a
dome with a ring all resting on a vertical support. The dome is assumed to be part of a
sphere with half central angle Il = 28°. The geometrical details of dome shell connection are
depicted in figure (8-19). The purpose of this problem is to design the reinforcing steel in
the dome and the ring. The reinforcement steel has an assumed allowable tensile stress of
1200 kg I cm 2, and
a = 29.0 m
t = 10 Cm = O. 1 m
h=45 cm=0.45m
b; 20 Cm; 0.20 m
2
q = 300 kg/m
E; 2 x 10 5 Kg/cm 2
v = 0.0
Dome
T ~~
-~~
()I.=2a
.
I
E
"
..
C>
on
I
"
I
..::.
Ring
r= \1.61 m
I. b=20c:m
~
Design:
To design this shell structure, we must first determine the internal force field in the "dome-
ring" system. For this purpose, we employ the method developed in the earlier sections of
this chapter. The numerical calculations based on the force method of shell analysis are
carried out in the following.
t
d' = 2 x Ces Cl = 5 x 0.88 = 4.41 Cm Sin Cl = 0.47
Ces Cl = 0.88
y =~-d'=22.50-4.41~18.09 Cm
e 2
b'=~-..!..-SinCl= 10.00-2.35=7.65 Cm
2 2 Cm
h
e= ( 2 - d ' -b'tga )Ces Cl =(22.50-4.41-4.07)0.88= 12.34
A4=3(I_v2)(~)2=3(29.0)2=25.23x104 ~ 1.=22.41
t 0.1
Now we calculate the numerical values for the flexibility influence coefficients and the
membrane deformations. By direct substitution of assumed data in the appropriate
formulas we obtain:
2
D 1 2a\Sin 1 2x29.0x22.41x (0.47)2
-E
(~
Dl l = =! 2871.21
E t 0.1 E
D
0 12 =
1 n 2 Sina 1
----=-
2x(22.41)2 (0.47)
=
1
E 0
4720.76 = [ln
E t E 0.1
3
0 1 41. 3 1 4 (22.41) =.l... 15523.43
0 22 = - - - =
E at E 29.0 x 0.1 E
2
Y 2 2
oR = .l...(1+ 12 e E-.=.l...
7)
+ 12 x (18.09) ) (13.61) 2
11 E bh E (1 (45) 2 .45 x.20
R 1
o
12
=--
E
1 12x (l3.61)2xo.1809 1
-E 22063.2
E
0.20x (0.45)3
1 12(13.61)2
E =.l... 121963 5
0.20 x (0.45) 3 E .
188 Design and Analysis of Shell Structures
2
DD = a q (1+V Cosa )Sina
10 Et 1+Cos a
= (29)2 xq 1 1
ExO.1 (1+0.88- 0 • 88 ) 0.47=- T 1375.88 q
D ~ . 29q 1
D 20 = Et (2+v)Sl.n a = E;C""Q.T (2) 0.47=T 272.60 q
2
~( -aq_l - _ ...9...
Ebh 1+Cos ci - 69935.62 E
R 12r 2 e
D20 = - Ebh 3
qa
(- 1+Cos a ) = + 232158.86 T
If we substitute these values in the parametrie solution of the compatibility relations, Le.,
expressions (8-45), we fmd
11 _ _ ..:..(
- .::..13.::..7:...4:.:8:.:6,-"•.;:..9.::.3:...)..:..(-_7:...:1:.:3,-,,1:;:1.;:..• .::..5.::..0:..)_-_(:..-..::1:..;.7.::.3.;:..4.::..2~.4~4:.!)....;(:.::2.;:..3.::.2.;:..4~31:-.,-4;...6..:...) q =+h.24q
(137486.93) (8920.58) - (-17342.44)2
(8920.58)(232431.46)-(-17342.44)(-71311.50) -0.904'1
/>1
a (137486.93)(8920.58)-(-17342.44)2
M
a
= -0.904 x 300= -271.2 kg-m
m
Design of Reinforced Concrete Domes 189
Having obtained the edge forces. we can now use the expressions in table (9-1) to
determine the bending field in the dome. The appropriate expressions are
-
a
Sinll e
-AtjI
Sin AtjI
=
H + ,,2 e
-AtjI . Tl
Sln(AIj!+ -4) M
A II
- 2 A Sin ae
-AIj! Tl 2 -AIj! .
Sin (AIj!- - ) H - - ' - A e Sw (AIj!- ~)
212 M
2 a 4 ('(
N = - 12 Cot (a-Ij!l sin a e -.\tjlSin (A1j!- ..2!....)!I - Q cot (a-Ij!) e -'~J
~ 4 a
Here '" = a - <p. If we substitute the numerical values in the above relations. we obtain the
following expressions for our problem.
To determine the complete internal force field. we must add to these bending forces the
internal membrane forces. The calculations related to determination of bending and
membrane fields are summarized in tables (8-2) and (8-3).
Table (8-2)
'i'
I H:~\-~ • Ro.l.J. Sin.\~· Sil\(Af'~) Sln ().~- %) TI
Slll(Al}-'I) cot~J (Cf-IV ) ~
-.\4'
1--.--
O.\J U.U 1),\) +-n.71 -1. t)() -lJ.71 tl . Hel tl.UD
._-
3. lJ ~). U') t Ij. ')": t-U. 'J 3 -0.3') .0.3tl +2. J4 '0.33
o.J.u u.Il.l -u. 37 -u. 'J.! Tl). Y 3 -.0.40 t.2. ')0 +0.03
Table (8-3)
[~- total fjeld
bending fjeld membrane field
! <p'
I
'f
0
M., 1 Ne N-:> M",
N-
0
N ••
.p M.p Ne N.p
-- ______ L
I
2H 0.0 -271-=.:J 29998.84 1652.8 0.0 -3061.2 -4620.4 -271.2 26936.8 -2967.6
f---- -- - ----- .-'--
25 3.0 214.52 6269.4 1- 55 •58 0.0 -3321.1 1-4563.8 214.52 2948.3 -4619.4
f---- l---- -_._--
22
'--r--
6.0 -7.29 -1369.05 -223.43 0.0 -3552.1 -4514.4 ~; .2iT~~9;;:~1~ _._-
-4737.8
19! 9.0 -2.02 -927.4 -53.55 0.0 -3754.2 -4471.8 -2.02 ' -4681 1-4521 •
'---r--- -
16 12.0 -7.9 -91.01 14.15 0.0 -3927.1 -4435.9 -7.9 -4526.9 : -4421.8
13 15.0 -1.7 166 • 93 12.13 0.0 -4070.6 -4406.5 -1.7 -4003 -4394.4
-2961.6
'V. 47371\
N
e
Ik,/", I
The reinforcement design of the dome is based on finding the hoop and meridional
reinforcements. The hoop reinforcement (per unit width of the shell) which would be
needed in the lower part of the dome is determined with help of diagram (8-19b). The
required reinforcement is:
2
26941. 7 22.45Cm Im
1200
We note from diagram (8-19b) that beyond the meridional angle", = 5° the hoop stresses
are compressive. Hence, at the upper part of the dome there would be no need for any
calculated reinforcement However, a minimum reinforcement is placed there to control the
shrinkage of concrete as weH as to absorb the stresses caused from temperature changes.
We see from diagram (8-19a) that the meridional force is compressive throughout the sheH.
Hence, from the structural viewpoint, there would be no need for a designed reinforcement
along the meridional direction. However, we provide the sheH with a minimum amount of
shrinkage and temperature steel which, in this case, is chosen to be 14 mm diameter bars
placed 40 cm apart.
In addition to the mesh-type reinforcement, the dome must be provided with bending
reinforcement at the lower region. This bending reinforcement is determined with the help
of diagram (8-19c):
272.3 x 100
0.875 x 7 x 1200
Having designed the reinforcement for the dome, we now determine the reinforcing steel
for the ring. The bending moment at a typical section of the ring is
3691.0 x 100 2
As = 0.875x40x1200 8.79 cm
In addition to this, we should strengthen the ring for torsional effects; for this purpose, we
provide the ring with top reinforcement as weH as transverse closed stirrups.
Figure (8-21) shows the plan of dome reinforcement designed according to this procedure.
Note that, for the presently assumed axisymmetric loading, the reinforcement pattern is
axisymmetric, but for a compact presentation, all three types of reinforcements are shown in
a single figure.
Figure (8-22) shows the detail of ring reinforcement as weIl as detail of "dome-ring"
connection.
192 Design and Analysis of Shell Structures
The dome under design consideration must also be checked for buckling strength. In
chapter 13, we will present a number of formulas for the buckling loads of the shells of
revolution shall be presented. Those guidelines will show that the thickness of 10 cm of the
dome is quite adequate to withstand buckling.
Design of Reinforced Concrete Domes 193
Problems
P 8.1- The concrete dome roof of a hall is supported on a cylindrical wall (or beams over
columns) around its circumference as shown in figure (P8-1). A 2.4 m-wide annular slab
surround the dome, acting as an overhead canopy. The loadin~ on the dome is 3000 N/m 2
over the surface area, and the loading on the sI ab is 500 N/m . Dimensions for the dome
and the slab are shown. Poisson's ratio is assumed to be zero, and modulus of elasticity for
concrete is assumed equal to 2000 N/mm2 .
(2) Determine the steel reinforcement needed for this dome. The allowable stress of steel is
assumed equal to 150 N/mm2 .
10 cm
r~ß~
·1·
2.4.0 m 9.0 m i
Figure (P 8-1)
P 8.2- Determine the steel reinforcement for the concrete dome of example 7.8. Use the
steel bars with the an allowable stress of 1500 kglcm 2.
194 Design and Analysis of Shell Structures
8.1 - M. Farshad, Shell Structures, (In Farsi) Two volumes, Vol.: I 1986, Vol. 11: 1987,
University of Shiraz Publications, Shiraz
8.2 - W. Flügge, Stresses in Shells, Springer Verlag, New York, 1973
8.3 - D.P. Billington, Thin Shell Concrete Structures, 2nd ed., McGraw-Hill Book
Company,NewYork,1982
8.4 - V. S. Kelkar, and R. T. Sewell ...Fundamentals o[ the Analysis and Design o[ Shell
Structures, Prectice-Hall, INC., New Jersey, 1987
Analysis of Shells with Arbitrary Geometry 195
Chapter 9
9.1 - Introduction
Varieties of surfaces from which various shells may be designed and constructed are
practically infinite. With the advent of such materials as reinforced concrete, prestressed
concrete, ferro-cement, fiber-reinforced concrete, composites, and reinforced plastics, the
varieties of shell geometries have been further increased. The choice of a particular surface
geometry for the shell depends on the functional, structural, and architectural requirements.
Simple shell geometries, such as cylindrical forms and the shells of revolutions, can also be
combined to give more elaborate shell forms suited for specific purposes. The set of figures
(9-1) show some examples of combined shells and / or shells having new geometrical
shapes. The structural design and analysis of combined or arbitrarily shaped shells,
however, requires a methodology of its own to which the present chapter is devoted.
In the first part of this chapter, we will determine the membrane equilibrium equations for
shells with general geometrical forms. Then, to solve these equations and to determine the
membrane forces, we present an analytical technique, called the method of stress functions.
In a later section of this chapter, we develop a simplified theory of shallow shells. This
theory can be used in the bending analysis of many shells. In particular, this theory will be
applied to the analysis and design of Hyperbolic Paraboloid shells which will be treated
in the following chapter.
196 Design and Analysis of Shell Structures
y \,
horizontal plane
~
/r
angle of AD with Y
We denote the unit vectors along the x, y, and z axes, of the global Cartesian coordinate
system xyz, by L....J., and Js, respectively. The position vector, r, of the corner point A can
be written as
(9-1)
198 Design and Analysis of Shell Structures
Since the unit vectors 0, j, ls) are constant the partial derivatives of r in this system can be
expressed as
ar
(9-2)
ar
3r -, .
- Idx
,IX
C-,z) • dx =/ 1 + pL dx
= ,11 + 3x
(9-3)
and
ar / 2
dy = /1 + (~) dy =)1 + q 2 dy (9-4)
la~1 3y
In these relations
3Z dZ
p =- q = oy
dX
In figure (9-2), the quantities p and q have the following geometrical interpretation:
p=tanq,
(9-5)
q = tan '"
so that
A'B' 1
Cosc/> = AB = -;::::;==:
I 1 + p2
(9-6)
A'D' 1
Cos1jJ = """"Ai)" = --;:::,;=:=::;
I 1 + q2
Analysis of Shells with Arbitrary Geometry 199
The spatial angle, co, between the generally non-orthogonal sides AB and AD, of the spatial
shell element, can be determined by calculating the inner (scalar) product of their respective
vectors. So
ar ar az (9-7)
(a~) • (a~) = ax
ar ar ar ar
(a~) • (a~) \a~ \. I a~ Icosw (9-8)
11 + pZ 11 + qz COsw = M
we obtain
Tbe value of elemental area, dA, can also be calculated by forming the vector product of two
vectors representing the sides AB and AD. The resulting expression is,
We now have an the geometrical ingredients and can proceed with the membrane analysis of
shells with arbitrary geometry.
In order to carry out membrane analysis of sheIls with arbitrary geometry we use a weIl-
established methodology in which the stressjullctiOIl solution scheme is employed.
200 Design and Analysis of Shell Structures
shel/ element
horizontal projection
Figure (9-3) Free body diagram of an arbitrary shell element and its
horizontal projection.
The two sets of forces, Le., the actual membrane forces ( Nx.Ny.Nxy ) and their plane-stress
projections (Nx, Ny, N xy ), can be related using the shell geometry. Referring to figure (9-
3), and using the relations obtained before, we can write the relation between Nx and its
horizontal projection ( N x ) as
- Cosp
Nxdy = Nxdy CoS\jl
Analysis of Shells with Arbitrary Geometry 201
or
(9-12)
(9-13)
and
Moreover, the actual distributed applied loads ( Px, Py, pz) can be related to their
corresponding horizontal projections ( Px, Py, p z ). If the projected area is designated by
dA', then the relation between the actually applied fm·ces and their hOlizontal projections
wouldbe,
P.x,y,zdA' = -x,y,z
';) dA
or
p = P 11 + pt + q2 (9-15)
Y Y
We now have the relations between the actual membrane forces and their horizontal
projections. Hence, if we could determine the force field of horizontal projection of the shell,
then we could, in principle, use these relations to fmd the actual membrane force field for the
givenshell.
Thus we have converted the problem of spatial shell analysis to that of a plane stress
analysis together with another problem which, as we shall see, is much easier to handle
analytically than the original shell problem.
202 Design and Anal ysis of Shell Structures
(9-16)
aN aN
f.F
x = 0 => _x_ + ~+
ax ay p.x = 0
aN aN (9-17)
EF
y=0 => ---.:t.. +:....E... +
Cly aX. p.y = 0
To write down the third equilibrium equation, Le., the equilibrium of actual forces in the
vertical (z) direction, we go back 10 the actual shell element ABCD of figure (9-4). We now
write the vertical equilibrium relation for the actualforces ( Nx.Ny. and Nxy ) but we try to
express this equation of equilibrium in terms of the projectedfictitious forces N x, Ny, and
N xy•
The vertical projection of the actual membrane force acting normal to the side AD, in terms
of its fictitious counterpart, is
Analysis of Shells with Arbitrary Geometry 203
NX.AD.tang<P = N (~ .d 1 ·)-12--
x /1 + r,L
=N
x
1 1
~
02+ odv
- dZ
= N -dv
1 + p2 • - x:ix-
c
Similarly, the vertical projection of the membrane force normal to the side Be is
- dZ - dZ
N -dy + -d (N - ) dxdy
x dX dX x dX
Cl [ N
-;:;- ~Z
- (~) -J dxdy
oX x ~x
In a similar fashion, the resultant of vertical projections of forces normal to the sides AB and
CD are found to be
_d -
(~
3z dxdy
.,,--)
oy y oy
The vertical projection of the membrane shear force acting on the side AD is
-N .N).tang<P = N
- dZ
."..-- dy
xy xy oy
N ~ d + 1-ö~' ~) dxdv
xy dY Y dX xy dY -
..L[N
dX
(dz)1 dxd
xy dX Y
The resultant of vertical projections of shear forces acting on the sides AB and CD is
similarly
204 Design and Analysis of Shell Structures
The contribution of externally applied forces to the equilibrium equation in the vertical
direction is (Pz dx dy ).
Summing up the projections of internal force resultants and the external forces in the
vertical direction, and keeping track of proper signs, we obtain the following vertical
equilibrium equation for the actual shell element
a2z
aN ClN
-
N ~2: + 2N ~+N + ( x + -Et) dZ
X "XL xy ax3'! y ay2 ax ay Clx
(9-19)
oN aN
+ (äf + äfY) ;~ + Pz = 0
If we use the relations (9-16) and (9-17) in this equation, we obtain
(9-20)
(9-21)
(9-22)
The set of equations (9-16), (9-17), and (9-20) constitute the governing equilibrium
equations for membrane shells of arbitrary geometry. The relations (9-12), (9-13), and (9-
14) give the actual membrane forces in terms of calculated projected forces.
Analysis of Shells with Arbitrary Geometry 205
We solve the system of partial differential equations (9-16), (9-17), and (9-18) by the
method 0/ stress function. We introduce a generating function, <l>(x,y), called the stress
function and we define it as follows:
_ a2~ -
Nx = ay2 - J Pxdx
(9-23)
a2~
Nxy = - oxoy
These expressions satisfy equation (9-16) and (9-17) identicall y. The third equation, Le.,
equation (9-20) yields,
(9-24)
This relation is a second order linear partial differential equation on the unknown stress
function, <l>(x,y); it has variable coefficients which are dependent on the shell geometry. For
a shell with specified shape, they are known functions of x and y. The right hand side of
this equation is a known forcing function which depends on the applied 10ading as well as
the shell geometrical configuration. The formulation of the problem is completed by the
specification of the appropriate boundary conditions.
From a mathematical point of view, any second order partial differential equation can be
classified as being of hyperbolic, parabolic, or elliptic type, depending on the sign and
magnitudes of the coefficients of the second order terms; these depend on the shell
geometry. This classification has physical importance. Hyperbolic equations have the
property that the disturbances propagate, in the shell proper, whereas elliptic equations
describe non-propagating features. For a detailed discussion of this subject, the reader is
referred to standard books on partial differential equations.
The governing equations derived in this seetion will be solved analytically and I or by
numerical schemes. In the following chapter, we will apply this method to analyze one of the
most widely used types of shells, Hyperbolic Paraboloid shells. This method can also be
used to find analytical solutions to shells having Elliptic Paraboloids and Conoids as their
middle surfaces.
The analytical treatment of general shell equations is difficult Numelical solutions to shell
equations can be obtained by the Finite Element and / or Finite Difference methods.
Analytic solutions though crude be, demonstrate the useful interplay between various
parameters.
The bending theory of general shells can be simplified to yield equations which can be
solved analytically and numerically. One of such simplifications is the so-called "shallow
shell theory", based on the assumptions:
For practical purposes, many Hyperbolic Paraboloid sheHs and also Conoidal shells may be
considered to be shallow shells. This approximate theory is also referred to as Vlasov
Theory of shallow shells.
.. _--/
.,.~,
I,
~
... ,
...,+~ ••
••
Consider a shell with a general middle surface defined by the equation z = z(x,y) in an
orthogonal Cartesian coordinate system. According to our previous definitions, we have,
(9-25b)
a2 z
--=8 Torsion of the shell surface
axoy
With these definitions and related interpretations, we now proceed to delive the governing
equations of linearly elastic shallow shells.
aN aN
-2.+~+p. =0
ox oy x (9-26a)
aN aN yv (9-26b)
~+--="",+p. =0
ay ax y
00 oU_ l 2 2 (9-26c)
""X + ~
__ + N a z + 2N ~ + N d Z + = 0
dX ay x ax 2 xy axoy y oy2 Pz
M =-M
yx xy (9-26d)
208 Design and Analysis of Shell Structures
aM aM (9-26e)
-Y+~-a....=O
ay ay -~
aM (9-26f)
2+
ax
aM._ aM
Q =..-Y.+~
-y ay ax
(9-27)
dMx aM
o-X =-+~
ax ay
If we substitute the expressions for Qx and Qy from these relations into (9-26c) we obtain
e:x =au
--rw
3x (9-29a)
e: av
=--tw (9-29b)
y ay
(9-29c)
Analysis of Shells with Arbitrary Geometry 209
In these relations, Ex, Ey, 'Yxy are the strain eomponents of an arbitrary point in the shell
thiekness. As we note, in the present approximation, the influence of normal displaeement
eomponent, w, (whieh in this approximate theory is assumed to be the same as vertical
displaeement) is highlighted. In other words, the gradients of other components are
neglected in eomparison with those of the funetion w.
The relations between the ehange in eurvature and also the twist of the shelI, on one hand,
and the displacement gradients, on the other hand, are,
a2w
Xx = aX2
(9-30)
Et
Nx = ~ (e; +-.)e;) (9-31a)
- \) x Y
Et
Ny '" 1 _ ~2 (Xy +" Xx) (9-31b)
Et
Nxy "'Nyx '" 2(1+'J ) Xxy (9-31c)
In these relations, the parameters D and K are the membrane and bending sfifjnesses of the
shell, respectively. .
D- Et
-1 - \.12
210 Design and Analysis of Shell Structures
By combining the three types ofbasic relations, Le., equilibrium, kinematic and constitutive
relations we obtain the goveming equations of Vlasov theory of shallow shells. The
synthesis procedure is as follows:
By differentiating both sides of relations (9-29a) and (9-29b) twice, with respect to y and x
respectively, and after adding up both sides of resulting relations, we obtain
(9-33)
(9-34)
At this stage, we consider special loading types in which only the applied loading has a
vertical component Pz and the other loading components are identically zero. This is
common for practical roof shell design problems.
To reduce the number of goveming equations, we now introduce a sU·ess function, cI>(x,y),
and we defme it in the following fashion:
N
x
=~
oy2 I
N
xy
=-~
oxoy
(9-35)
(9-36)
Ifwe make use of constitutive relations (9-31), the strain displacement relations (9-29) and
stress function relations (9-35) in the above relation, we obtain
V"cp + EtVk2 W = 0
(9-37)
Analysis of Shells with Arbitrary Geometry 211
In which
(9-38)
Now, substituting relations (9-31d), (9-31e), and (9-31f), and also relations (9-30) and (9-
35), into the equation (9-38), we obtain
or
(9-39)
Equations (9-37) and (9-39) constitute the goveming equations of Vlasov theory of shallow
shells. Equation (9-37) expresses the compatibility of deformations, while equation (9-84)
relates to the equilibrium of extemally applied and intemal fm·ces.
These equations for shallow shells contain more special theories as their offsplings:
(1) Theory of Hat plates - For a flat plate, the initial curvature is zero. In this case, the
goveming equations of shallow shells become deeoup1ed and take the following forms:
(9-40)
These equations govem a laterally loaded thin plate as well as the same plate loaded by in-
plane forees and acting in plane stress.
(2) Membrane shallow shells - In this ease, the bending stiffness of the shell, K, is
assumed to be zero. The goveming equation of this membrane shell is delived from shallow
shell equations to be,
(9-41)
In the following chapter, the theory of shallow shells will be applied to analyze and design
Hyperbolie Paraboloid shells.
212 Design and Analysis of Shell Structures
Problems
P 9.1 - Figure (P 9-la) shows a conoidal shell roof with windows which is used for roofing
of factories, assembly halls and other places. A conoidal surface is an anticlastic surface
having negative Gaussian curvature. A conoid is A mied surface that can be generated by
sliding a straight line on two different curves called the directrices. As a special case, one of
these directrices may be a straight line, Figure (P 9-1 b).The general equation of a conoidal
surface is,
where fI (y) and f2(y) are the equations of plane directrices, y is the vertical coordinate x is
the longitudinal, and L is the length of the conoid. In a special case that fI (y) =Ü and f2(Y) is
a parabola, then
z= _ fr 1 _y2]
TI b2
Where fis the rise of the parabola and b is the half-width of the shell
(1) Derive the appropriate goveming equation for the membrane field in parabolic conoidal
shell subjected to uniform loading of intensity qo.
(2) Determine the membrane force field in the parabolic conoid. Show that the membrane
shear force has the following expression:
~
xy
oll>
3XoY =
-
qo
a
eh + a 2x 2 y 2
y
+
n
2y
here:
2Qo 2f
n = - a =
a """ill2
I ~'
/
\/'1"
::/// z
(a) (b)
P 9.2 - Consider a cooling tower shell wh ich has a hyperbolic paraboloid of one sheet
as its middle surface. This surface is generated by rotating a parabola about a non-
intersecting vertical axis. If the axis of revolution is identified by the veltically oriented z-
axis, and if x-y axes characterize the horizontal plane, then the equation of this surface can
be written as
( x2 + y2 ) / a2 -z2/ b2 = I
where a and bare two constant parameters of the surface. The x-y plane is chosen at the
section B-B; the so-called the throat of the cooling tower.
(I) Write down the expressions for two principal radii of curvature, r\> r2 , in terms of the
constants a and b and the angle <1> of the shell normal with the vertical ( see figure (P 9-2».
(2) Derive the expressions for the membrane fOl'ces, Nq. and NB, in this cooling tower
subjected to its own weight. The shell is assumed to have a unifol1n thickness.
214 Design and Analysis of Shell Structures
9.1 - M. Farshad, Shell Structures, (in Farsi), Shiraz University Press, Vol. I, 1986, Vol. 11,
1987
9.5 - V.Z. V1asov, General Theory of Shells and its Applications in Engineering, NASA
Technical Translations, NASA TIF-99, 1964
Design of Hyperbolic Paraboloid Shells 215
Chapter 10
10.1 - Introduction
Hyperbolic paraboloid shells are doubly curved shells with negative Gaussian curvature;
they are called HP or Hypar shells, and a subclass of them are called Saddle-type shells.
Hyperbolic paraboloid shells are structurally efficient and many constructional and
aesthetic advantages: they are used to cover large spans, vast roofed areas, and a variety of
other roofed spaces, figure (10-1); they are used as foundations for special structures; they
can be prefabricated simply.
The theoretical tools for the membrane and bending analysis of HP shells were prepared in
chapter nine. In the present chapter, the solution methodologies of that chapter will be
applied to the analysis and design of hyperbolic-paraboloid shell structures. In addition to
quantitative analyses, we will also present qualitative discussion of the overall structural
behavior of these shells.
In a later section of this chapter, results of simplified theory of shallow shells, developed in
chapter seven, will be applied to predict the bending field developed around the edge
members of hyperbolic paraboloid shells. Based on this theory, practical formulas are
presented which give the analytical expressions for bending forces in the HP shells.Some
design guides for HP shells are given in this chapter, and a sam pIe design example of a
reinforced concrete HP shell is carried out in detail.
216 Design and Analysis of Shell Structures
tt
(10-1)
in which
We can easily verify that the intersection of vertical planes, parallel to the x'z and y'z planes,
with this surface would create plane parabolic curves. On the other hand, the intersection of
this surface with horizontal planes. z =constant. generates hyperbolas having equations
The name hyperbolic-paraboloid thus originates from this feature of the surface that has
families of hyperbolas and parabolas as its intersections with horizontal and vertical
planes, respectively, figure (10-3).
218 Design and Analysis of Shell Structures
If we intersect the surface with the particular horizontal intersecting plane z = 0, we obtain
(10-2)
or
(.L._E)(.L. +E) o
If; ~ n; fft (10-3)
The above equation represents two intersecting straight lines which pass through the origin,
0, while remaining on the hyperbolic parabolic surface at aII their points. Hence, these two
lines could be considered to be the straight-line generators of the HP surface. These lines
are, in fact, the asymptotes of horizontal hyperbolas. The tangent of the incIination of these
lines with the x'-axis is
tanw
(10-4)
Ibl
lai
Now, we choose the directions of these asymptotic lines as a new horizontal coordinate
system, and we designate their directions by x and y. In figure (10-3), two sets of
coordinate systems, one being tangent to generating parabolas and the other being oriented
along the straight line generators, are shown. If the angle between two coordinate systems
is designated by 00, we can write the relation for coordinate transformation as foIIows:
x' (y- x)CoS'.LJ
If we substitute these expressions, for the new coordinate variable, into (10-1) we obtain the
equation of hyperbolic parabolic surface in the coordinate system xyz. This equation is
expressed as
(10-6)
In a particular case that fl = f2, we have, tan ffi= 1. In that case, the equation for the surface
assumes the following simple form
2
z=-xy=-xy
1 (10-7)
f2 C
By inspecting this equation we observe that the result of intersecting this surface with
horizontal pI aces are hyperbolas having the coordinate axes x and y as their asymptotes.
A very interesting feature of the hyperbolic paraboloid surfaces, as seen in equation (10-5),
is that by intersecting the surface with vertical planes, parallel to x or the y axes (i.e., y =
constant or x = constant), we obtain the equations of straight fines. This means that we can
generate a hyperbolic parabolic surjace by sliding a straight line on two other non
intersecting lines.
Figure (10-4) shows part of a HP surface generated in this fashion. Obviously, the
generated surface is doubly curved. It is also non-developable and has non-zero negative
curvature. Surely, by intersecting the surface of figure (10-4) with a pair of orthogonal
planes, we shall obtain two intersecting parabolas.
vertical planes
a
I
~y
From what we have said here we conclude that a HP surface can have straight lines as its
generators. This feature is shown in figure (10-5). If the x-y plane (Indicated by x-y axes
and two orthogonal dashed lines) is horizontal, then the corner point of the shell with
horizontal coordinates x = a and y = b, is seen to have a rise of z = f. Inserting these
=
coordinates in the equation (10-7), we obtain c abI f. So, the equation of this surface
would read as
z =(f/a b) xy.
generating line
/
/~
i ~
01
Figure (10-6a) is a single module shell which, in practice, could be situated on two supports
located diagonally at the lower corners. Figure (I 0-6b) is a four module shell which is
usually supported at four corners; this type of shell is sometimes referred to as hipped
hypar shell. Figure (1O-6c) shows an inverted umbrella shape or mushroom-type shell,
since it is supported at its middle point by a single column.
Generation of a HP shell surface, by straight lines, has practical significance: it means that
the forming of concrete HP shells can be carried out by using separate pieces of rectilinear
shoring.
In some schemes of prefabrication, use can be made of the foregoing geometrical feature of
straight-line generated HP shelJs. For example, a low cost construction scheme can be
developed in which the molding of concrete precast shells could be achieved by a heap of
soil. In this method, a heap of shapeless soil could be very conveniently shaped by means
of a long straight plank sliding over two straight directrices. in this fashion, an "earthen
mold" on which the concrete shelJ can be easily cast would be economically generated.
Figure (10-7) shows a picture of a case in which this method was applied successfully to
construct a number of precast concrete HP shells.
r~·-;lF
Jf,l~ '\.\~
~~~" . "
,'\n-,~ ,• 'i'
I '~&
-~:'
• . . ." . . . . lo";
~".
~
I,
I
. ,- .~
• ~.
,~ ,./ -' .
In this section we will analyze the behavior of HP shells having straight generators as their
straight edges by using the membrane theory developed in previous chapter.
Consider a straight-edge hyperbolic paraboloid shell with the mid-surface equation of the
fonn,
(10-8)
We use the governing membrane differential equation (7-24). If we evaluate the second-
order derivatives of z(x,y) and substitute them into the equation (7-24), we obtain the
following equation:
(10-9)
To carry out the analysis further, we must now specify the applied loading. Suppose that
the shell is subjected to a load uniformly distributed with intensity p on the horizontal
projection. Snow loading is one of this type. In this case, the equation (10-9) becomes
(10-10)
Here, Fl(x) and F2(Y) are two integration functions to be determined from the boundary
conditions. Now, the internal forces are given by the equation (9-23):
-
N
1 ab
=--cp=--p
xy 2 2f
(10-12)
1 N = d 2 F2 (10-13)
N =--cp
xy 2 X dY2
which shows that Nx is a function of y only, and Ny a function of x only.
Design of Hyperbolic Paraboloid Shells 223
A HP shell may have a variety of edge conditions. Sometimes HP shells may be provided
with the so-called edge members. In a majority of practical cases, edge members are
comparatively stiff in their own plane, but can not sustain loadings applied in the lateral
direction, figure (10-8). In such cases, either or both of the normal forces, N x and Ny, would
vanish at the boundaries normal to their direction so that Fl (x) and F2(Y) would be assumed
to be identically zero. Consequently, the normal membrane forces are identically zero
throughout the shell and we are left with a pure shear membrane force field existing
throughout the HP shell.
(a) (b)
Consider the HP shell roof shown in figure (1 0-8a), composed of four HP shell segments.
The whole shell structure rests at its four corners on vertical column supports. The applied
loading is assumed to be uniformly distributed with intensity p on the horizontal projection
of the shell.
The state of stress at a typical element ofthe shell is determined using relations (10-13) and
the prescribed boundary conditions. Tbe internal membrane force field along each
generator line consists of a pure shear force of constant magnitude,
't = Nxy = - c p /2 = - a b p /2 f
Figure (l0-8b) shows the state of stress in an element of the shell. A more detailed picture
of stress and force distribution in the shell is presented in figure (l0-9).
Tbe principal stresses at any point corresponding to this pure shear are a tensile stress 't in
the direction parallel to OB and a compressive stress 't in the direction parallel to AC.
Note that the membrane field is inversely proportional to the shell rise, f. Tbus, according to
membrane theory, shells with higher rise value have more reserved strength and thus are
stronger than shells with lower rise. To arrive at a more definite conc1usion about the actual
224 Design and Anal ysis of Shell Structures
strength of the shell, however, the stability requirements must also be taken into
consideration.
According to membrane theory, there exists a distributed internal shear force system at the
edges of the shell of figure (10-9). To satisfy the equilibrium requirements of membrane
theory, these edge shears must be transferred to vertical supports by means of some
intermediate members. Two types of such members are needed: edge beams and ridge
beams; members such as AB and Be are edge bemm, and AD and CD are ridge beams.
edge beam
The internal membrane shear force, 'C, in being transfelTed to the edge and ridge beams
causes axial forces in those members. In the present case, figure (10-9) shows that the axial
forces in both the edge beams and the ridge beams happen to be compressive. In other
cases, either of these axial forces may be tensile.
The magnitudes ofaxial compressive forces in the boundary members of the shell of figure
(10-8) are
(10-14)
ab PY
(FEl oc = Nxy.s = Nxy ~ = - 2f cosß
At the corners of the structure, the edge beams produce axial thrusts as well as vertical
forces. The vertical forces are carried by the vertical column supports, but the horizontal
thrusts must be absorbed by tie members. Figure (10-8) shows horizontal ties carrying the
axial thrust in tension.
Design of Hyperbolic Paraboloid Shells 225
R = ab ~
2f Cosa
• Sina + ab2f cPosbß • 8inB C~ pab (10-15)
and the axial compressive force in, for example, the ridge beam CD is equal to:
As we see, the axial force the ridge beam is zero at the outer boundary (the edge) and is
maximum at the center.
Consider the hyperbolic paraboloid shell roof shown in figure (10-10). The assumed shell
dimensions are,
6.0 111
Solution:
The constant value of membrane shear force field is
1 ab 1 3x3
N
xy = - - 2 T P = - - 2 x - - x 400 = -1800 Kg/m
1.0
The principal stresses associated with this internal force can be obtained from the following
well-known formula:
whichgives
N = ta = ± N = +1800 kg;m
xy
This completes the membrane analysis of the present example. As we see, the level of
internal tensile and compressive stresses is quite low. We have observed the same structural
feature in other shell types.
Design of Hyperbolic Paraboloid Shells 227
According to the membrane theory, the state of stress in a HP shell element, oriented along
the straight-edge generators, is a pure shear which remains constant throughout the shello
The principal stresses make 45° with the straight-line generators; one principal stress is
tensile, the other compressive; both are constant throughout the shello
shear
tension
/ comprossion
(a)
(b)
In HP shells having edge members, the applied load is normally transferred to the edge and
the ridge beams which, in turn, carry the induced axial forces to the shell supports.
Figure (lO-ll) shows the membrane behavior of HP shells with straight edge boundaries
and edge beams. Figure (lO-lla) shows the intern al stress field and the shear stress
transfer to the edge beam.
Figure (lO-11b) demonstrates the mechanism of load transfer from the shell to the edge
beams. This sheII is assumed to have simple continuous supports, suitable for the
realization of membrane conditions.
Figures (10-12) show the flow ofinternal forces from the shell body to the edge members
of the shells depicted in figures (10-6).
We conclude that the structural action of the HP shell arises from the interaction of two
mechanisms: (I) a cable action of the shell along the directions of principal tensile stresses,
and (2) an arch action along the lines of maximum compressive stresses.
A HP shell is an anticlastic surface, Le., it has negative Gaussian curvature. The cable and
arch mechanisms interact in two "opposite" fashions: the applied load induces
compression in the "inner-built" arches while it also causes tension in the "internai" cables,
figure (10-13). The applied force is thus distributed between these two "in-built" structural
elements; the over-compression of the arches is hindered by the cables that tend to "puH
up" the arches at each point; this increases the elastic stability of HP shells.
The sign of the axial force developed in the edge and ridge beams of HP shells depends on
the shell configuration and the supporting conditions. Figures (10-14) show the state of
axial forces in the edge and ridge beams of two HP shells.
(a)
The HP shells considered so far had square horizontal plans; other shapes are possible.
Figures (10-15) show some examples of HP shells with straight line boundaries having
various horizontal projections; they are all composed of basic HP shell units having
arbitrary quadrilateral shapes.
So far we have analyzed the behavior of the hyperbolic paraboloid shells with straight edge
generators. There are HP shells which have other configurations and boundary formations.
There are many shapes of HP shens that have curved, parabolic edges. One may combine
simpler HP shells to construct a number of useful shell configurations. Figure (10-16)
shows a HP shell with curved edges.
Figure (10-17) demonstrates the membrane behavior of the HP shell shown in figure (10-
16). Again, the arch and the cable actions interact efficiently. The curved edges act as
compression arches which carry the shear forces, transferred to them by the shell, to the
shell supports. There is some boundary field at the support which is compensated by the
bending mechanisms ofload transfer.
)I
Note that all these conclusions are based on the predictions of membrane theory. In actual
situations, there will be some bending field. The magnitudes of the bending fOl'ces and the
range of penetration of the bending field into the shell will depend on the source of the
bending field as wen as the geometrical properties of the specific shell.
The stability of HP shells must also be considered in their design. However, as we have
pointed out before, the membrane theory gives a very useful overall picture of the shell
behavior. Some designers have designed large HP shells using onIy membrane theory
together with their intuitive knowiedge of HP shell behavior.
232 Design and Analysis of Shell Structures
The results of this section were obtained by Loof who has used Vlasov theory of shallow
shells to find the bending field of moment and shear in the square HP shells with straight
line generators.
Loofs results for two different boundary conditions in a square HP shell shown in figure
(10-18) are as folIows:
f -1
~ = +1.732 ga(1:) (7-86)
In these formulas, My and Qy are the bending moment and the shear force in the mid-
length of the shell edge. The parameter g represents the intensity of uniformly distributed
verticalload; fand t denote the rise and the thickness of the shell, respectively.
Consider part of a HP shell structure with simply supported edges as shown in figure (10-
15). At the hinged edge, the ben ding moment is zero but the non-zero trans verse shear
force is,
~ = +0.577 ga (~)-1
)(
The maximum bending moment occurs at a seetion located at a distance Yl from the edge,
where
These formulas show that the bending moment is reduced by increasing the shell rise, and
increased by increasing the shell thickness.
Figure (10-19) shows a log-log plot of the variation of dimensionless bending moment in a
HP shell with dimensionless parameter (f I t). The limiting value for very small quantities of
(f I t) corresponds to the bending moment in a flat plate.
fit
Figure (10-20) shows the variation of bending moment versus the dimensionless length
parameter. Figures (10-19) and (10-20) show that the differences in boundary condition
have little effect within the shell body.
-O~\l
-0.5
-0-\
iS -03-
~+I-02
I"': -01
---1
_----'1"'.1"'-3
or-~r---~----~--~--~2',I==:~~~~=====-~4'
0.1 -
02 +0167
Hyperbolic paraboloid shells are among the most favoured shell types. Since they can be
generated by straight lines, they can be formed by using rectilinear planks or prefabricated
forms.
A simple state of stresses in a membrane HP shell is another feature which can be used to
great advantage. As we have learned, the internal force system in a membrane HP shell
consist mostly of a pure shear wh ich gives rise to two principal stresses, one being tensile
and the other compressive_
Concrete HP shells need tensile reinforcement; ideally this should be placed along the
directions of principal tensile stresses, figure (10-22,1), but for ease of construction is often
placed along the straight line generators, figure (10-22 II).Figures (10-21) show the
probable cracking patterns in HP shells; against which the shell must be reinforced.
The edge beams and point supports induce some bending field in the HP shells. The extent
of bending field is dependent on the geometrical parameters of the shell as weil as the edge
conditions. Reinforcements are needed to strengthen the shell against the bending field
arising from these sourees. To achieve an efficient transfer of loads, from the shell to the
edge members, one must design the system in such a fashion that the monolithicity of the
whole shell structure is assured.
We now present some guidelines for the choice of appropriate dimensions for the
preliminary design of two HP shell types; a hyped shell and an inverted umbrella type
reinforced concrete shell.
Figure (1O-23b) presents appropriate proportions for the preliminary design of the
reinforced concrete HP shell, of figure (l0-23b); this having a generally rectangular plan
and resting on four corner supports. This graph gives appropriate proportions of rise to
long span ratio with long span, for a number of short span values.
Figure (10-24) and its related table (10-1) give some guidance in the design of an inverted
umbrella type shell. This table gives appropriate dimensions for the edge members.
In designing of hyperbolic paraboloid shells we must consider the buckling behavior of the
structure. The arch strips oriented along the compressive diagonal are compressed, and
therefore may buckle. Moreover, the axial force in the edge members of the shell are, in
most cases, compressive forces. So, the edge members of a HP shell could also be prone to
buckling. Hence, the shell thickness as weIl as the dimensions of edge members should be
controlled to satisfy the stability considerations.
In chapter 13, of this book, the stability of shells is discussed. In the same chapter, some
formulas for the critical buckling loads and buckling modes of hyperbolic paraboloid shells
are presented. Use may be made of those results in the actual design of HP shell
Structures.
236 Design and Analysis of Shell Structures
0.4
~~/v I
c:
Y
,:;;,1:f
0(5:.
I
Cl!
a.
(/) 0.3
/'
Cl
c:
o
V I
i
....J
I", ~
E V\0.,.Ü'
I/ 1'\~li~Q1"
Cl
·äi
.r::.
a.
i=
0". _ I - \~
0.2 I- f-. &
\ ~o
~
<$I" ~!- ':?
0
'?
/ \~<$I"
~
~l?':?
V
~
\ '6>
':?
0.1 I
Om 30 m 60 m 90 m
Lang span (m)
... '~
edge beam
L
.~ ., .
:dJ l1
~.r; ()
mo .-··
I' ." .
~iJ,
. .-
I
.. ' ... ~ f1J?' .' - -.. . .:. : .
. . '.... ... '\ :r~~"":'.:: .. ~ r- ridge beam
.; ' . . . ......."
t-" I-:-; .-. ',.. 22
:~: .' '; •• ~ .. ..,J
L h 8
max Dmax 18 mln D
mln
I
I
m m rm Cm Cm Cm Cm rm Cm
12 ') 50 8 15.5 8
II :: I·::::
I
:::: I:::: : :: :: 7~ 2~
tl_Lt ,~l
18 'J.D 9 L) 13115.,> fj I I
I 24 I 18115.~ i I I
I :: ! , : : I
'1----2~4112.~J 15 5
I fj 2 25
! '>-1
tl
, 9
I 33 18 1
".,~
. 1
I J0 I 9. 0 1 LI • r, 70 20 1 1 ') .:. 1 8 11 J 40 I
jO 110.~> I 9 GO 20 1
I
15 . 5 8 35 35 I
30 15.11 I 9
____L_l
43 20 1 15 . '> B 3') 35
'------+----'..' -"-----'--~IL___ _._l _
Design of Hyperbolic Paraboloid Shells 239
As a design example. we treat the reinforced concrete HP shell shown in the figure (10-24).
The thickness of the sheIl is 10.0 cm.; the assumed uniformly distributed verticalloading is
p = 400 kg/m 2; this inc1udes the dead weight of the concrete ( wh ich is assumed to weigh
2.5 tl m 3). The steel reinforcement as weil as the edge members of this shell are to be
designed. The allowable tensile strength ofthe reinforcing steel is 1400 kg/cm 2 •
FEI 1".
a 3[lö
cosa =-1800x-3- = -5692.1 kg
!TaX
3
Pt = Force In the tle = F
E
I .Cosa = +5692.1x- = 5400 kg
!TaX {lö
P : Force In the column = p.a 2 = 400 x (3) 2 = 3600 kg
c
240 Design and Analysis of Shell Structures
Considering that the edge beams cannot resist torsion, we may assume that the edges can
rotate freely, and use Loof formula for hinged edges. Accordingly, the edge shear force and
maximum bending moment are
M \ =+0.149 pa2 (.!.-) -4/3
Y max t
Y1 = 0.55( ~ ) -1/3 a
f -1
~ = +0.577 pa(T)
Therefore
_ 1 -1/3
Y1 - 0.55 (0.1 ) x 3 = 0.77 m
1 -1
~ = +0.577 x 400x 3 x (0.1) = +69.24 kg/m
To design the reinforcing steel mesh, to be placed along the straight li ne generators, we use
the following relations:
1400 kg/cm L
J = 1 - k/3 0.86
Design of Hyperbolic Paraboloid Shells 241
Now, if the parameter d, called the effective depth, is the distance from the point of action of
the resultant of the concrete compression to the centroid of steel reinforeement, then by
applying simple staties we find the area of steel reinforeement as
M 24.9 x 100 2
As =f Jd = 1400xO.86x7 = 0.295 cm jm
s
AS 0.295
P = b.d = 100x 7 = 0.0004< Pmin = 0.0035
Since the calculated steel ratio is smaller than that prescribed by ACI codes on concrete
shells, the prescribed minimum reinforcement is provided for this shell. The provided steel
cross sectional area is,
P min 0.0035
AS = Pbd = 0.0035xl00x7=2.45 cm 2
Since the provided reinforcement is high er in value than this magnitude, hence the design is
also satisfactory from the bending point of view.
To design the edge elements, we assume an overall dimensioning of (15 cm x 20 cm) for
each of the edge and ridge beams. The intensity of dead weight of the edge elements is,
d = 17 cm
242 Design and Analysis of Shell Structures
M 93.75x100
A s = f Jd = 1400xO.86x17
= 0.433 an 2
s
We compare this calculated steel ratio with the minimum required reinforcement:
0.433
P = 15x 17 = 0.0017 < Pmin = 0.00502
A S = 0.00502 x 15 x 17 = 1.28 an 2
We design the shear reinforcement for the edge beams as folIows. First, we plot the shear
force diagram in the edge beams as shown in figure (10-25). According to ACI code of
practice of reinforced concrete, the design shear is to be taken at a distance equal to the
effective depth of the member from the support. This design shear force is shown in the
following figure.
d: .7 c m
The routine calculations for design of shear reinforcement are caITied out as folIows:
v = w~ = 75 x {lo 118.9 kg
max 2 2
Vu 105.84
v u =O.85J:x:l = O.85x15x17 =0.488 kg/an 2
Design of Hyperbolic Paraboloid Shells 243
~v = 4.19 kgjcm 2
c
v < v
u c
Minimum shear reinforcement:
3.52b S
Av . = -f--
nun y
5 ~ d/4 when vu
- vc > 1. 06!f:.. = 1. 06/250 =
C
16.76 kgjan2
17
5<2=8.5 5 = 8 an
(2A ) x 2800
s AVmi.nXfy= 5.~10 2xO.79x2800
84 cm
max 3.52x b
3.52x 15 3.52x 15
Figure (10-26) shows the reinforcement detail ofthe edge members.
~12 Scm
The design of ridge beams can be carried out in a completely similar fashion. Figure (10-
27) shows the reinforcement detail of one of the four lidge beams
Having designed various part of this shell structure, we can now put the designs together
and present an overall design sketch for the whole system. The design sketches for this
exarnple are presented in figures (10-27) to (10-29).
244 Design and Analysis of Shell Structures
20cm 14 cm
N <1>12
A
1
J
I.
I.. scm .. 1
15 cm
,"
Figure (10-27) Reinforcement detail of the ridge beams
r---
: 1 1 -.J ..
r4h
I' I
E I' I
::: 1
1I
1 1
- - -c_=cc_~ ~=-tl~
~
:':
-- -=--C~~1j
:11: -- ---- ---1
I !
e> 1' 1 I
:;
n I II
IJ
~'O[~fiT41 tr-~
'1';;~
E
: 1
w_
I ,I •
-'---
Problems
P 10.1 - Consider the relatively shallow reinforced HP shell of figure (P 10-1) resting on
four corner supports. This shell has a unifonn thickness. The weight per unit surface area
is pz.Show that the expressions for the projected membrane fOl·ces in this sheIl.are,
Figure (P 10-1)
Here k = c / a b and f1(y) and f2 (x) are functions resulting from the integrations.Using
these expressions, find the expressions for the true membrane stress resultants in the shell.
LI = 40 m, L2 = 80 m, h = 15 m
Figure (P 10-2)
246 Design and Analysis of Shell Structures
10.1 - M. Farshad, Shell Structures, (in Farsi), Shiraz University Press, Vol. 1,1986, Vol.
11,1987
10.3- A. M. Haas, Design ofThin Concrete Shells, John Wiley & Sons, New York, 1962
10.4 - C. Faber, Candela: The Shell Builder, Reinhold Publishing Corporation, New York,
1963
10.7 - V.N. Baikov, (ed), Reinforced Concrete Structures, MIR Publishers, Moscow, 1968
10.9 - V. Gioncu, Thin Reinforced Concrete Shells, John Wiley & Sons Ltd., Chichester,
1979
Chapter 11
11.1 - Introduction
Folded plate structures are composed of a number of flat plates connected to each other.
They have many uses: in roofing oflarge spans in an architecturally appealing appearance;
as box girders in bridges and overpasses; as bunkers in silo structures; as sheet piles.
In this chapter, we present the analysis and design of prismatic folded plates. Although
folded plates can be analyzed by numerical schemes such as the Finite Element method, but
the analytical calculations of this chapter aid the understanding of the structural behavior of
folded plates.
250 Design and Analysis of Shell Structures
There exist a large variety of folded plate structures used in different engineering and
architectural applications. Figures (11-1) show some examples of the used of folded plates.
~
tf
~
~
Figure (11-1) Examples of folded plate structures
Analysis and Design of Folded Plates 251
Folded plates are degenerate shells: in the flat plate parts the curvature is zero, while at the
corners, i.e, at the plates junctions, the curvature is infinite. The simple geometrical
configuration of folded plates makes the manufacturing of metallic, composite, and
polymeric folded plates quite simple. Forming of reinforced concrete folded plates is much
easier than for other shell forms. Thus there are great economic advantages in the
constructional aspects of folded plates as compared with more complicated curved shell
configurations.
A simple paper experiment, shown in figure (11-2), demonstrates the structural efficiency
gained by forming an otherwise structurally inefficient flat piece of paper into a folded plate:
placing the material further away from the axis of flexure, makes the system much stiffer
than a flat plate. The folded plate shape of the figure (1l-2b) is a truly load bearing stiff
structure.This structure is a prismatic folded plate. Other nonprismatic efficient shapes
could also be generated with the same piece of material.
(a) (b)
By folding a flat plate into a folded plate, we completely change the internal force system as
weIl as the load carrying mechanisms. In a folded plate structure, in-plane (membrane)
forces are generated in both directions; The internal force field is truly spatial and the
applied force is mainly carried by the generated in-plane fOl·ces, not by bending in a flat
plate structure. In this sense, a folded plate is aform-resistallt structure. Its strength and
stiffness to weight ratio are much higher than those of flat plate and planar framed
structures.
Consider the folded plate roof structure shown in figure (11-3); it can be characterized by
the plate dimensions, slopes, lateral span, and the length on the sU·ucture. So far as the span
(b) length (1) relation is concerned, the prismatic folded plates can be c1assified as being
long (Vb> 2), short (Vb < 2), and intermediate. This c1assification is similar to the one
presented for cylindrical shell roofs.
252 Design and Analysis of Shell Structures
I-- b
Long folded plates resting on end supports behave like beams, figure (11-4). In short folded
plates, the lateral action, i.e., the action of the folded strips becomes important.
-----------_ .....
The behavior of a intermediate prismatic folded plate is truly spatial; it behaves like a beam,
but it is modified by the action of the transverse folded strips. The longitudinal load transfer
emanates from the action of inclined plates This is referred to as the plate action. The
lateral load transfer takes plates by folded strips acting as a continuous beam and
undergoing bending deformation; this is called the slab action of the folded plate, figure
(11-5).
Analysis and Design of Folded Plates 253
Figure (11-5) Plate action and slab action of a prismatic folded plate
The structural behavior of a prismatic folded plate, resting on end supports, can thus be
described as folIows:
(1) First, the applied loads act on transverse folded slab strips causing these strips to
undergo bending as a continuous beam resting on flexible supports located at the
intersection of the inclined plates. The supporting reaction, R, comes from the longitudinal
plates, figure (11-6).
(2) The transverse folded slab strips transfer their loads to the end supports by means of the
longitudinal plates action of the folded plate. The vertical reactive force, R, will act as a
longitudinally distributed applied force on the incIined longitudinal plates. Each plate will
receive a share of this distributed force. To determine the share of each plate, the vertical
reactive force must be resolved into two components lying in the plane of two adjacent
plates, figure (11-7).
254 Design and Analysis of Shell Structures
(3) The totality of implicit transverse slabs and longitudinal plates act in such a fashion that
the integral behavior of the whole structure is preserved. This means that the action of each
individual element takes plate in conformity with the equilibrium and compatibility
requirements of the whole folded plate structure.
This qualitative discussion on the behavior of prismatic folded forms the basis for analytical
determination of internal forces and deformations in folded plates presented in the
forthcoming sections.
Prismatic folded plate roofs are normally supported by the end diaphragms resting on end
columns; the end diaphragms can be solid, or may be in the form of end frames and / or
trusses.
There are many prismatic folded plate sections from wh ich a design choice can be made;
folded plate sections having tri angular and trapezoidal shapes are very common since the
forming of such sections is repeatable and their construction is relatively easy; folded plate
sections made of a number of slanted plates forming a prism are favoured. They
approximate cylindrical roofs by a number of flat surfaces; their forming is simple
compared with a curved cylindrical surface.
Figures (11-8) suggests initial choices for geometrical parameters of reinforced concrete
folded plates. Folded plates mayaiso have stiffening and / or supporting edge beams, and
the general design considerations described for cylindrical shell roofs are applicable.In the
Figures (11-8) present appropriate relative dimensions for lateral span and slopes of
inclined plates in a folded plate roof.
In prismatic reinforced concrete folded plate roofs, the length of the folded plate ( L ) is
usually greater than the lateral span ( b ) depending on the architectural planning. The rise
of a folded plate is normally chosen in the range of VIO to V7. The choice of a particular
folded plate geometry depends on architectural and functional requirements.
Analysis and Design of Folded Plates 255
rs: ,~Z
Table ( 11-1) Appropriate proportions for two types of reinforced concrete folded
plates
Reintorcemenll . I
per kgl Thic~ness, I
Rise (MeIer) I Widlh Length
horz.projection Centlmeler Max Min I (Meter) i (Meter)
0.11-0.08 10.0 1.2 0.8 : 4.5
,
12.0
:
I
~
0.19-0.13 10.0-15.0 1.8 1.8 1.2 18.0
0.26-0.18 10.0-15.0 2.25 1.5 7.5 I 22.~
• 8 J 0.36-0.28 12.5-15.0 3.0 2.0 9.0 30.0
--- - - - " .- . .-
.:1
0.14-0.11 7.5 1.5 0.75 6.0 12.0
,~
I
0.21-0.14 7.5-9.0 1.8 1.2 7.5 18.0
I,
0.28-0.18 7.5-10.0 2.J 1.5 9.0 22.5
I
Table (11-1) presents some appropriate proportioning of two types of reinforced concrete
folded plates, inc1uding the average thickness, and the approximate magnitude of steel
reinforcement (per unit horizontal projection). This table may be used in dimensioning
reinforced concrete folded plate roofs and in the initial estimation of required material.
There are several methods by which the folded plates and, in particular, prismatic folded
plated could be analyzed. In addition to numerical schemes, such as Finite Element method,
the folded plates could be analyzed by a number of analytical methods. One of these, is the
arch-beam method which, in principle, is similar to the beam-arch method of cylindrical
shell analysis. Another method, is the limit analysis technique employed for ductile yielding
structures. A third method, of elastic analysis of folded plates, is based on combined lateral-
longitudinal behavior of folded plates described in seetion 11.2. We shall employ this
analytic methodology to analyze the folded plates of the f0l1hcoming seetions.
A general comprehensive analysis of elastic prismatic folded plate consists of three parts:
the preliminary analysis, the corrective analysis, and superposition. For the initial
design of folded plates, the preliminary analysis proves to be sufficient. The corrective
Analysis and Design of Folded Plates 257
analysis is only needed as a refinement in the later stages of design, specially for relatively
flexible folded plate structures.
The transverse slab analysis determines the support reactions. These reactive forced are
oriented along the vertical directions.
The transverse slab analysis yielded longitudinally distributive reactive forces. These
reactions were actually provided by the longitudinal plates. Thus, the action of transverse
strips would be forces equal in magnitude to these reactions but opposite in direction of
these reactions. Therefore, loading on longitudinal plates consists of components of reactive
forces determined in the previous step (with their signs changed) resolved in the plane ofthe
adjacent plates.
The analysis of individual longitudinal plates can be carried out by means of classical beam
theory; this determines the linearly varying bending stresses in each plate, and the deflection
of each plate in its plane. This deflection must satisfy the boundary conditions at the ends of
the folded plate structure.
The analysis of each longitudinal inclined plates yields a ben ding stress and deflection field
independent from the stress and deflection fields in the other plates. In the folded plate
structure the plates are connected along the plates junctions; it is this connectivity that
creates an efficient folded plate structure; this interaction imposes certain constraints on the
stresses and deformations in the individual plates.
258 Design and Analysis of Shell Structures
The plates are matched for equilibrium and compatibility; the bending stresses at the points
of intersection of two adjacent plates must be equal; the deflections of two adjacent plates
must be compatible.
The preliminary analysis of folded plates gives an equilibrated force field and a compatible
deformation field for the whole folded plate structure assuming immovable lines of plate
junctions. In reality. the plates junctions are not totally constrained in space; and there will
be some lateral movement of the lines of plate intersections which in turn will affect the
stress and deformation fields.
This phenomenon resembles the "sidesway" appearing in framed structures. The lateral
movement of plate junctions would alter the stress and deformation field in the folded plate.
The magnitude and influence of such movement would depend on the features of the folded
plate structure.
The corrective analysis. taking into account the movement of plate junctions, can be carried
out in similar fashion as to the sidesway analysis of frames. The procedure is to apply unit
lateral displacement to each junction and to determine the influence of these unit deflections
on the structure.
Having obtained these influence coefficients we write the compatibility relations for each
junction. In writing such compatibility relations we use appropriate geometrical relations at
the junctions.
The corrective analysis. like the preliminary analysis. has three stages:
The results of preliminary and corrective analyses must be superimposed to yield the
complete force and deformation fields in the monolithic folded plate structure. The
superposition must be carried out so that the compatibility requirements at the plates
junctions are fulfilled. By writing down the appropriate compatibility relations. the actual
lateral displacements at the junctions of adjacent plates are determined.
The results of corrective analysis. assuming unit displacements. are then used to fined the
corrective internal forces produced by these displacements (settlements). Superposition of
corrective analysis and preliminary analysis would then yield the complete solution to the
folded plate problem.
Analysis and Design of Folded Plates 259
Figure (11-10) shows the action of longitudinal plates; forces equal and opposite to the
reactions, Ri, are applied and are resolved into two components along the adjacent plates.
Figure (lI-lOb) shows the adopted positive sign conventions are also indicated; the
parameters cIli indicate the orientation of inclined plates with respect to the horizontal
direction.
Figure (11-11) shows the bending stresses obtained from the analysis of the longitudinal
plates. The bending stresses at the line of junction of two adjacent plates are, in general,
unequal. Since adjacent plates are monolithically attached to each other, the bending stresses
at the junction must be equal. To achieve this equalization there must be an equilibrating
shear force, t 12, at the junction. This shear force is to be determined by a process of
redistribution of bending stresses, as outlined in the following section.
"
Figure (11-9) Part of a folded plate section showing the transverse
slab action, distribution of the transverse bending moment, and reacti ve
forces at the plates junctions
z:}l
'0
Figure (11-11) Bending stresses in the indi vidual plates and the junction
shear resulting from plates matching
Consider two adjacent plates in a prismatic folded structure, numbered (n) and (n+l) in
figure (11-12). The components of applied in-plane forces, figure (11-10), produce bending
moments in these plates. The bending moments produce linearly vmying bending stresses.
According to classical beam theory, the maximum stresses occur at the extreme fibers of
each beam, which are located at the plates junctions. In the figure (11-12), these maximum
values are indicated by O"n and O"n+l.
I
(n'l Plate (n+1)
[nI -------It.--.. .
• t"" Pl n
~
-K"
In a monolithic folded plate. the stress in the common fiber located at a plate junction must
have the same value when considered as a fiber in either of the adjacent plates; the difference
between two stresses <Jn and <Jn +l vanishes by virtue of a compensating shear stress. 'tn• at
the junction.
Figure (11-12) shows this shear stress and its resultant. N n • over the length of the plate. If
we consider each longitudinal plate as a relatively deep beam then. figure (11-12) shows that
the bending stresses are produced by the shear forces as weil as the junction reactions.
Thus. every one of these longitudinal plates is acted upon by a resultant axial force.
produced by edge shears. a bending moment. again due to edge shears. and the bending
moment produced by the junction reactions.
To achieve a stress matching between two adjacent plates. we should calculate the total stress
in the longitudinal junction fiber caused by the ben ding and shear actions. for each plate.
and so determine the required equalizing stresses.
x
N
n-1
;: - JX 'LI n-1 d.x, N
n
= 1o T dx
n-tl
(11-1)
o
The expressions for complete longitudinal stress at the common fiber. calculated for plates
(n+ 1) and (n) are. respectively
N Nn + 1 Nn+1 hn+1
n
---+
(11-2)
M N h N N Nh (11-3)
o,n n-1 n + n-1 _ ..2:!. _ ~
-~- 2S A A 2S
n n n n
In these relations. (h n• An) and (h n+l.An+Ü are the (thickness. cross-sectional area) ofplates
(n) and (n+l). respectively. The quantities Sn and Sn+l are the section moduli ofplates (n)
and (n+ 1), namely
(11-4)
Now, we match the two plates. The equilibrium constraint is that the bending stress at the
common junctional fiber must be the same; by equating (11-2) and (11-3) we obtain
Nn - 1 N N N
p:- + 2{--.!! + _n_) + n+1 = (11-5)
n An An +1 An+1
This equation relates the shear forces at three subsequent junctions (n-1). (n) and (n+ 1). and
resembles the well-known three moment equation for continuous beams.
262 Design and Analysis of Shell Structures
c
---
Plate (n+1)
To determine the share of each adjacent plate in the balancing stress, Le., to find the
distribution lactor, we proceed as folIows: Consider adjacent plates (n) and (n+l), again
as shown in figure (11-13). By imposing the condition of stress equality at the junction B
weobtain
N
_n_+ (11-6)
~l
Analysis and Design of Folded Plates 263
4Nn -4N
--+0 = __n_+ o (11-7)
An+1 n+1 An n
1 1 (11-8)
4N ( - + - - ) = (a - CI )
n ~ An+1 n n+l
This relation shows that the unbalanced stresses an and C1n+l can be equalized by addition
of certain shear forces as indicated in the above relation.
4N An+1 (11-9)
D = - __n_ = - (0 - 0 )-::-~~_
n An n n+1 An + An +1
(11-10)
Figure (11-14) shows the share of each of two adjacent plates from the unbalanced
--- --
equivalent shear force.
"........
//
/
/ / p/.€ite / -- ~
/ 1f'/~7) //
611 ,-
/
,-
,-
"
G
11.,.,
I G
n
I
I
- ... ...JI
Figure (11-14) The share of each of two adjacent plates in the balancing
shearforce
264 Design and Analysis of Shell Structures
Using relations (11-9) and (11-10), we obtain the corresponding distribution factors, D n
and D n+l
1
An+1
Dn+1 = ba -;-1-":':+'-_~_""';1== (11-11)
~ An+l
1
A
D = -ba n
n -1-+~
An An+1 (11-12)
Here ~<J = <Jn-<Jn+l is the equivalent unbalanced shear stress. The coefficients Dn and
Dn+ 1 are called shear distribution factors at the junction n of the folded plate structure.
We note that the share of each plate is proportional to its shear compliance, i.e., the inverse
of shear stiffness of each plate acting as a beam. This becomes evident if we would multiply
the cross-sectional areas, appearing in the nominators and denominators of (lI-lI) and (11-
12), by a common factor of the shear modu1us, G. As wc know (by ignoring a common
shape factor), the quantity GA is the shear stiffness of a beam section.
The condition of equilibrium requires that the plates in which the unbalanced forces are
distributed must be in equilibrium. To satisfy the equilibrium requirements, part of the share
al10cated to each plate should be transferred to the other edge of the plate. This event is
similar to bending moment transfer in continuous beams.
Figure (11-13), and the equilibrium requirements in each plate, show that the shear stresses
carried over to edges a and b are
(11-13)
(11-14)
The transferred quantity in each plate is, in magnitude, one half 01' the share of that p1ate
from the unba1anced stresses. Therefore, taking the algebraic sign into consideration, we
conc1ude that the carry over factor for each p1ate is equal to -1/2; this is the same carry
over factor for uniform continuous beams.
Analysis and Design of Folded Plates 265
(1) perform the preliminary analysis, including transverse slab and longitudinal analyses,
and redistribution of unbalanced shear forces to arrive at a stress compatible force system in
the constrained folded plate. The method is similar to moment distribution, but is called
shear distribution.
(2) Perform the corrective analysis. Introduce unit deflection at each junction, carry out
the related force analysis along the lines indicated for the preliminary analysis. We must
assume an appropriate longitudinal distribution for lateral displacement at each junction. For
example, for a simply supported folded plate, a sinuosoidal distribution of det1ections and
forces is an appropriate assumption.
(3) Fulfill the deformation compatibility at the junctions by writing the kinematic relations
matching the adjacent plates, and so find the actual lateral dis placements of the
junctions. Multiply these quantities by the results of unit lateral displacements to find the
actual corrections.
(4) Superpose the results of preliminary analysis (part (1» and corrective analysis (part
(3» to fmd the complete force and displacement field in the folded plate stmcture.
In the following section, we apply this methodology to the analysis of a reinforced concrete
folded plate roof.
Consider the simply supported reinforced concrete single prismatic folded plate roof
structure shown in figure (11-15). Figure (11-16) shows the cross section of this folded
plate.This symmetric folded plate is composed of five plates, two of which act as the edge
beams for the structure; it rests on two solid end diaphragms.
The present folded plate can be considered as a spatial flat plate approximation to the
cylindrical roof designed in chapter 5; we would expect that the force systems in the two
structures to be similar.
The applied loading to the folded plate of figure (11-15) is assumed to be as folIows:
Dead weight of reinforced concrete (assuming a lOcm thick folded plate) 240 kglm 2
Weight of additional load 50kglm 2
Live load 100 kr;/m 2
o(3~=200 0 o.
4
I.'
I
2.60 m l
"
3,oom
i ~
Figure (11-16) Cross seetion of the folded plate of figure (11-15)
Analysis and Design of Folded Plates 267
The geometrical features of the folded plate stlUcture are given in tables (11-2) and (11-3).
Q)
ca h(cm) t(cm) A= th(cm2)
th 2
S= T(cm 3 ) cp Sincp eoscp
0::
1 90 15 1350 20250 90 1.00 0.0
0 0 0.00 !
I -
I
1 60 0.866 0.577
._---
2 20 0.342 2.75
3 20 0.342 2.75
Analysis:
To perform the analysis of continuous transverse slabs by means of moment distribution
method, we need the moment distribution factors; these factors relate to continuous
transverse strips acting in bending; they are not the shear distribution factors that we have
discussed be fore in relation to the unbalance shear distribution between the longitudinal
plates.
The moment distribution factors are proportional to the flexural stiffness (4 E I / h) of the
transverse strips. For a strip element with one end hinged, the distribution factor is
multiplied by 3/4. By doing so, we need not to carry the moment to the hinged end. In the
present case, the 10wer ends of two edge be am are free while their upper ends are free to
rotate, i.e., to be hinged.
The moment distribution factors are calculated in table (11-4). In this table, the junction and
plate numbers refer to the numbering chosen and shown in figure (11-16).
268 Design and Analysis of Shell Structures
!
2 11.2= 4 I 1.000
2 2 3 0.433
~1= (T) (4) = 3
I
3 3 0.567
I
I ~3= 4 x 3.05
_ 3
~
3 IS2- 4 x 3.ö5 0.500
3
IS4= 4 x 3.ö5 0.500
We consider a continuous transverse strip of unit width located at the mid-length of the
folded plate. At this stage, we assume that this strip behaves as a continuous beam on fixed
supports. The well-known moment distribution method can be used to find the bending
moments is this statically indeterminate beam.The distribution and carry-over factors are
given in table (11-3). With the help of these factors, the unbalanced moment distribution in
this continuous strip can be carried out.
Table (11-4) summarizes the moment distribution process Having obtained the moments,
each segment of this continuous strip can be considered and the shear fOl·ces at its ends can
be determined by statics. Finally, the end shear forces can be combined to find the reactive
forces at the junctions of strip segments, i.e., at the plate junctions. These values are also
calculated in table (11-4). Figure (11-17) shows a half profile of the folded plate and the
reactive forces at the plate junctions, as determined in table (11-4).
Analysis and Design of Folded Plates 269
1 2 3 Junction
0 10 12 21 :l 3 32 Member
o
Figure (11-17) Reactive forces at the junctions of a transverse strip
At this stage, we resolve the reactions to the forces obtained in the previous stage, along the
inclined longitudinal plates. If we designate the received force share of plate i at its edge j by
the symbol Pij, we have
PlO= 0
P21= 0
CoS413
P 22= 1336 Sina 2 1336 ~
0.342 = 3847 Kg/m
CoS41
2
P 32= -1336 sr--
ma2 = -1336 ~= -3383 Kg/m
0.342
_ Cos~3 _ 0.985
P33- 1144 sina2 - 1144 0.342 = 3294 Kg/m
270 Design and Analysis of Shell Structures
These component forces can be combined to yield the forces transferred to each
longitudinal plate. The result is shown in figurcs (11-18).
The longitudinal plates can, now, be analyzed as simply supported beams under uniform
loading. The maximum bending moments and bending stresses in these plates can be
obtained by the classical beam formulas. These ca1culations are perfOlmed in the following.
G)
P3= P32 + P 33 = 3294 - 3383 = -89 Kg/m -.!!-- ~
p ,t 2 2
M = _3_ = -89 x 20 = -4450 Kg-m
8 8
M 4450 x 100 2
0t= -Ob = -S- = 155040 = 2.87 Kgjcm
P ,t2 2
M = _2_ = 3847 x 20 = 192350 Kg-m
8 8
192350 x 100 2
0t= -Ob = - 150000 = -128.2 Kgjcm
P t2
M = _1_ = 782 x 20 2 = 39100 Kg-m
8 8
=
0t -ob
= _ 39100 x 100
20250
= -193 • 1 Kg/cm2
I
Analysis and Design of Folded Plates 271
Figure (11-19) plots the resu1ts of the stress calculations in each longitudinal plate, treated
independently from the other plates. As we see, the stresses at the common longitudinal
junction fibers are not equal. Therefore, there is an unbalanced shear force which is to be
distributed between two two adjacent plates. If the shear distribution factor for the junction i
of the plate j is designated by the Dij. then relations (11-11) and 01-12) give
3050
3000 +3050 = 0.50
0 __ =0.50
V=::=::;,-""":=:::::::::i+:J, 2 .17
'''.1 I,
113.1
I
I
Figure (11-19) Unbalanced bending stresses in the longitudinal plates
We should note that, due 10 symmetry, the shear stress at the junction :I is zero. So, no shear
carry over is to take place to that junction.
The unbalanced shear distribution process is summarized in table (11-5). Figure (11-20)
shows the final result of balanced stresses in the longitudinal plates.
272 Design and Analysis of Shell Structures
Table (11-5) unbalanced shear force distribution among the longitudinal plates
0 2 3 Junction
1 2 3 Member
-44.1
+23.5
____.-->+8 !I. 2
At this stage. the longitudinal plate are in equilibrium. The deformations of these plates
should also be compatible; this leads to some relations which we shall now derive.
Figure (11-21) shows the displacements of the longitudinal plates, Oj in their own planes.
These quantities are the displacements of longitudinal plates at the mid-Iength of the folded
plate. Figure (11-21) also shows the lateral displacements, ~i of the plates junctions, and the
conventional positive directions of these displacements are also depicted.
ES,, ---l0
( +-'b,
It can be shown that the in-plane displacements of the plates and the lateral displacement of
the plates junctions are related by the following kinematical relation.
(11-16)
113
274 Design and Analysis of Shell Structures
Now, as we know, for a unifonnly loaded simply supported beam having a rectangular cross
section, the relations between the bending stresses, maximum bending moment and
maximum deflection are as follows:
~1!,2·
Ö = if9EI
5(ab - at)S~2
Ö = 2 x 48EI
S 2
1=11
ab - at 5 t 2
Ö = h (4ifT)
In these relations, O"b and O"t are the bending stresses at the bottom and top fibers of the
beam, respectively. Using these relations together with the results of table (11-5), we can
fmd the displacement of plate i in the preliminary analysis. <>io. Thus. we obtain
Figure (11-22) shows the defonned state of the folded plate at its mid-Iength as predicted
by the preliminary analysis.
Ci)
/"
///
-- --
...,/-
--- -- -- --
The preliminary analysis was based on the assumption that the junction points were
immovable, whereas some lateral displacement, as shown in figure (11-22) was obtained in
that stage of analysis. In reality, junctions displace laterally. This displacement is not the
same as that obtained in the preliminary analysis. Some correction is required to achieve a
deformation compatibility of the adjacent plates. This is done in the "corrective analysis".
The corrective analysis consists of introducing unit lateral displacements at each junction, in
a sequential manner, and finding the force and deformation fields due to these individual
lateral "sidesways". Then, a set of displacement compatibility equations are wlitten which
would yield the true lateral dis placements. Note that, for each one of these unit lateral
displacements, an analysis similar to that of the preliminary analysis (consisting of slab and
plate analyses) should be performed.
In figures (11-23) and (11-24) the unit lateral displacements of plates number 2 and 3 are
depicted.
/
/
/
/
-r--.c..::.=---
-------
\
I,
I
Figure (11-24) Unit Lateral displacement ofplate 3
For a transverse slab, the end moments due to the unit displacements shown in figures (11-
23) and (11-24) are,
276 Design and Analysis of Shell Structures
Now, we repeat the moment distribution procedure for each of these two patterns. These
calculations are performed in tables (11-6) and (11-7). The con'esponding end reactions are
also obtained.
Table (11-6)
-') = lOTI
I
P 1 2 3 Distribution
10 12 21 23 32 Member
+168
0
+221 Distribution
Table (11-7)
I i
,
I Junction
b 1 2 3
10 12 21 23 32 Member
The results of calculations performed in tables (11-6) and (11-7) are depicted in the figures
(11-25).
IV • 1T X
Erv= CIo = PSm T (11-17)
= Elv" = - t1T 2
2 • 1TX
M(x) PSm - - (11-18)
R.
Pa 2 (11-19)
M=~
1T 2
(11-20)
278 Design and Analysis of Shell Structures
(11-21)
Now, having carried out the analysis of transverse slabs for unit lateral displacements, and
having obtained the above relations, we can perform the corrective analysis of longitudinal
plates. The procedure resembles the preliminary analysis. The details of the calculations are
as folIows:
Forß2 = 1 cm
Pu = 85.1 Kg/m
P 32 = 195.5
Cos $3
sin a 2
= 195 • 5 0.866 = +495
0.342 Kg;m
For ß3 = 1 cm
Pu = -125.5 Kg/.m
P21 = 0
0.985
P22=+413.5 0.342 = 1191 Kg/m
0.866
P32 = -413.5 0.342 = -1047 Kg/m
Figures (11-26) and (11-27) show these loadings applied to each plate.
",
IS.I ,""A'
:.;...r
~~~
~Platel~
Plate 3
Figure (11-26)
A,· ,
,~
;Y
~~
r~
u Plate1
Figure (11-27)
ForA2 = 1 cm
Fora3 =1 cm
a =- a = -(1659 + 1047) x 20 2 x 100 = -70 7 kg/ 2 Plate 3
b t '11'2 x 155040 • an
O'b
=_ O't = -125.5x 20 2 x 100
'11'~ x 20250
= -25 •1 kg/an2 Plate1
I'
Figures (11-28) and (11-29) show the results oflongitudinal plates analysis. At this stage.
the ben ding stresses at the junctions are not equal and should be equalized by a shear stress
distribution.
j
I
,
-~
'r .n.
-",
I i
"11.0
I ,;.. I
Figure (11-28) Bending stresses for a2 = 1 cm
-70.7
+
.'0.4
/
-25.1 i -JI.I
The results of unbalanced shear force distribution (in the cOlTective analysis) are presented
in tables (11-8) and (11-9).
Table (11-8) Corrective shear force distlibution • .12 = 1 cm
1 2 3 Plate
0 0.69 0.31 0.500 0.500 Distribution factor
-~ -~ -~ Carry·over factor
+17.0 -17.0 -15.2 +15.2
Stresses at -
+29.6 -29.6
fixed edges
0 +1.24 -0.56 +7.2 -7.2 0 Distribution
-0.6 0 -3.6 +0.28 0 +3.6 Carry over
+16.4 -15.8 -19.4 +22.7 +22.4 -26.0 Sum
0 -2.5 +1.1 -0.15 +0.15 0 Distribution
+1.3 0 +0.1 -0.6 0 -0.1 Carryover
+17.7 +18.3 -18.2 +22.0 +22.6 -26.1 Sum
0 +0.1 0 +0.3 -0.3 0 Distribution
0 0 -0.2 0 0 +0.2 Carryover
1 2 3 Plate
0 0.69 0.31 0.500 0.500 Distribution- factor
Carry-over factor
~ ~ -Is
Stresses at -
-25.1 +25.1 +32.2 -32.2 -70.7 +70.7 fixed edges
0 +4.9 -2.2 -19.3 +19.3 0 Distribution
-2.5 0 +9.6 II 0 -96 Carryover
°= ~
Ob - °t R.2
1T2
ForA2 = 1 cm
ForA3 =1 cm
(11-2:>'
Now, if we substitute the relations (11-16) into (11-22) we obtain the following
compatibility relations for the actuallateral displacements of plates jUl1ctions.
(11-24)
Li3 = -7.2508 + 5.012~2 - 11.2938~3
A3 = -1.62 On
284 Design and Analysis of Shell Structures
Having determined the actuallateral displacements of the plates junctions. we can obtain the
correction by multiplying the unit load corrections by the magnitude of actual
displacements.
Junctions 0 1 2 3
Plate 2 Plate 3
Preliminary an~lysis 0 +208 -344.5 +137 -274.7
Corrective analYSIS
62 = 1= 0 i +221 -110.5
1I2~-2.52an 0 -556.9 +278.3
!
t.3 -1= 0 I I -326.0 +539.5
"'3 = -1. 63c:l I
I
0 i
I i t528.1 -874.0
Total correctlon 0 -28.8 -595.7
Final results 0 +193.5 -373.3 -175 -870.4
Analysis and Design of Folded Plates 285
Tabfe (11-12) Shear force in the mid-Iength of the folded plate (kglm x 10 3)
Junctions 0 01
.. 1 12 2 23 3
,
I
,
!
To find the shear stress at each point of the folded plate, we need to know the longitudinal
variation of the shear force. We assurne that the shear force has the same variation as the
moment Mn and loading Pn. So, to find the shear stress, 't, we do as folIows:
(11-27)
(11-28)
in which No is the shear force obtained in the preliminary analysis. The shear stress,
according to preliminary analysis, is,
(1 _ 2x )
x- (11-29)
Plate L 3L L
A(x= 0) B(x= ~) C(x=T) D(X=a) E(x=T)
Junction 3 0 0 0 0 0
11~
N = Nmax Sin T X
=> T = -u:- Cbs
11 X
T (11-31)
In practice, we assume a linear longitudinal variation for the shear stress. On this basis, we
maywrite
4(N +N ) (11-33)
1" = 0 IIBX ~lIax (1 _ 2x )
tl. l
At this stage, we can determine the principal stresses at each point of the folded plate. The
appropriate known relation is
(11-34)
Figure (11-31) shows the plan of stress trajectories in the folded plate of figure (11-16). In
this figure, the compressive and tensile directions are depicted by the dotted and solid lines,
respecrively.
The results of this folded plate roof problem are similar to those obtained in chapter 5 for
the single simply supportcd cylindrical roof with edge beams.
The values obtained in table (11-13) could be used to design reinforcement for the concrete
in the folded plate roof. The steel reinforcement would consist basically of longitudinal bars,
transverse bars, and inclined bars placed at the corner region. The pattern of reinforcement
is similar to that of single simply supported cylindl1cal shell with edge beams.
288 Design and Analysis of Shell Structures
Problems
P 11.2 - A large spaee is to be eovered by a reinforeed eoncrete folded plate roof. Figure (P
11-2a) shows the overall geometrical dimensions of this roof. Figure (P 11-2b) shows the
transverse seetion of one of the repeating units. Analyze this structure and design the
required steel reinforcement. For your analysis and design, the following data may be
assumed.
Intensity of dead load: 300 kglm 2
Intensity of live load: 100 kglm 2 (on the projeeted horizontal sUlfaee)
f c = 200 kglem2, f y = 3000 kglcm 2
Es =2000000 kglcm2, Ec =200000 kglcm 2
~ ~/
tJ.~m ;> //?;~//:/J
/,/ /
" / / ~
/ /// / /
/'
l'lf
'0-r
I /~~~/~
1/ ")<~
h
\\ '.' b
.l.. -.//h ~- - l ' -
m / /
. ' -
I ' I! I' '<m
I f--9Cm-- i
__ --+_
-H''''--- I
I :
---t- ,o
1 1 '
r-"<~ -j--1tO m 1o Om ;:m--":
tao m
1--.---1.
1 .! I se", ! t
11.1 - M. Farshad, Shell Structures, Vol. I, 1986, Vol. H, 1987, Shiraz University
Publications,Shiraz
11.5 - "Phase I Report on Fo1ded P1ate Construction", Report of the Task Committee on
Fo1ded Plate Construction, Journal of the Structural Division, Proc. of ASCE, Dec. 1963, pp
365-406
Design of Liquid Retaining Shells 291
Chapter 12
12.1 - Introduction
Liquid retaining shell structures are made of steel, concrete, reinforced plastics, and other
reinforced materials; they may have circular shape or other geometrical forms; they could be
located underground, over of the ground, and / or on elevated towers. Circular containers
could be roofed by cones or domes; they could also have cones or flat plates as their base.
In the present chapter, emphasis will be mainly placed on the design aspects of reinforced
concrete circular containers.
In the first part of this chapter, we discuss general design considerations related to
cylindricalliquid storage shells. We will use the membrane theory of cylindrical shells
developed in chapter three and the bending theory presented in chapter four. We will also
develop a theory ofaxisymmetric behavior of circular cylindrical shells for quick reference.
In the later parts of present chapter, we treat circular containers with various top and bottom
conditions.
292 Design and Analysis of Shell Structures
Liquid retaining tanks may be classified according to shape, size, location, application, and
materials of construction.
(1) Configuration
Liquid storage structures may be cylindrical, rectangular, conical, combined, or some other
shape.
(2) Size
They may be small, medium or large. Volumes less than 50 m 3 are smalI, from 150 to 500
m 3 are medium, from 1000 m 3 to 2500 m 3 are Zarge, over 5000 m3 are very [arge.
(3) Location
Liquid storage containers could be buried (underground), ha1f-buried, situated on the
ground, figure(12-1), or be p1aced on e1evated foundations, figure (12-2) and I or towers.
Accordingly, they wou1d be called buried, half-buried, ground, and eZevated tanks.
(4) Application
Liquid retaining tanks could be water tanks, gas tanks, and I or press ure vessels.
Liquid retaining structures could be reinforced concrete containers, steel tanks, etc.
Elevated water tanks can have a variety of top and bottom systems. The set of figures (12-3)
show some of these arrangements. In overall shape design of e1evated tanks, the diameter to
height ratio is customarily chosen between one and two. For initial design, an average ratio
of 1.5 is recommended.
Design of Liguid Retaining Shells 293
r-------------------------- ·l
r:==-::t r=5II F--:J
IB
. ~
_.~
____~
_ __tmftI
_. -----4
~';:;6 r::==t
---------------'
~~
. .
r]
Gi J•
'\
I
!
I . /
1/f- ~ -
,(I ''.1 1
~ u.
/
X
xl 11\
(e) (f)
Elevated tanks, and specially reinforced concrete water tanks, can have various floor
systems; often the tank bottom must be stiffened. Moreover, the elevated tanks require some
sort of stiffening rib-ring which transfers the vertical forces to the elevated support and / or
strengthens the tank against the local effects of point supports. Figures (12-4) show some
arrangement of these flat and domed floor rib systems.
For larger elevated tanks, the bottom rib system must be more elaborate. Figure (12-5)
shows a network of beams which, together with a ring, form the stiffening system of a large
tank.
Design of Liguid Retaining Shells 295
There are many kinds of towers for elevated water tanks. The towers could be shells or any
other continuous structural systems (figures (12-6». In some cases, the tank and tower can
be a single monolithic structure; then the water tank is a natural continuation of its elevated
support. Figures (12-6c) and (l2-6d) show two models of this arrangement. The elevated
supports could also be in the form of an spatial framework and / or a spatial grid system.
Figures (12-7) show three examples of such systems.
,..-~-.
,, I ,, ~f~; .
.-------~
,
1~_~_.; '-1; I I
In addition to requiring structural strength and stability, liquid retaining structures should be
designed and constructed so that they do not leak. Therefore, the requirement of
"serviceability" would inc1ude the structural as weB as functional critelia, which in this case,
is the efficient containment of the contained fluid .
For reinforced concrete containers, there must be no cracking in the concrete. Thus,
cracking conditions put a limit on serviceability of concrete liquid retaining structures. In
order to satisfy this requirement, the stress in the concrete and the strain level in the
reinforcing steel must lie below certain limits. These limits are specified by some codes of
practice.
Design of Liquid Retaining Shells 297
,,.~fU'"
1 filler
- . --
wafer prooling steel collar
~
Ub ~
brtum~i~n~~fj!.i.::.,.j
~R'==
plasflC ring l=::::i:~~P'l~
plasflc JOint
4+ prestressed pipe
1I00rlng. asphalt
waterst~op
".:~'"
f+ l
I :•.• ' ....
I
.....
A. • a ••
;~
8
pool
waif 01 dry dock
Liquid retaining structures are subjected to hydrostatic fluid press ure, soil pressure,
temperature gradients, dead weight loads, and dynamic loadings arising from wind and
earthquake effects. In larger reservoirs, additional hydrodynamic loading due to the motion
of contained fluid could be produced; this motion called slushing.
The hydrostatic forces vary linearly along the height of the tanl<:. The wind and earthquake
forces are, however, more complicated to be described by simple analytical formulas. As a
298 Design and Analysis of Shell Structures
fairly good approximation for wind forces, we may assume the following expression for the
intensity of radial load:
(12-1)
Q(x,6) = q(x)Cos6
If we assume that the applied wind loading does not vary with height, then the expression
(12-1) reduces to
(12-2)
The applied force diagram corresponding to this approximation is shown in figure (12-14).
compression
The earthquake forces are body forces arising from the motion of the wall and the contained
fluid. The static fluid pressure on the wall is,
If we assume that the earthquake loading is proportional to fluid pressure on the wall at each
point, Then
In addition to these loadings , the buried containers could be subjected to fm·ces applied by
the surrounding soil and also the surcharge loads of vehicles and / or superstructures.
Design of Liquid Retaining Shells 299
In chapter 4, we developed the general bending theory of circular cylindrical shells and a
special theory ofaxisymmetric behavior of thin circular cylinders.
In the present chapter, that special theory will be used for the design and analysis of
cylindrical containers subjected to axisymmetric pressure and dead weight loading. We
recapitulate the basic features of the axisymmetric theory.
Consider a thin circular cylindrical wall of radius a, as shown in the figure (12-15). This
shell wall is assumed to be subjected to internal radial pressure as weIl as axisymmetric
vertical forces. The free body diagram of an element of this shell is shown in figure (12-16).
Q•• ~d.
d.
M.·~d
d. •
,.\,
Q
(b)
The equilibrium equations for the axisymmetric shell element of figure (12-16) are,
(12-5)
(12-6)
(12-7)
300 Design and Analysis of Shell Structures
du W
Ex -dx
- E
e " 'a- (12-8)
Et
N ... ~2 (E
x .1.- \I" X
+ \lEe>
(12-9)
(12-10)
As an approximation, we decouple the axial deformation from the hoop and the bending
fields. The constitutive relation for hoop force can be rewritten as
d 2w Et.3
M
x = K dx 2 K = 12(1- \/2) (12-12)
~= K d'w
Mo \i Mx dx'
(12-13)
d 2M
_-x_ + 1.... Ne = p (12-14)
dx 2 a r
and if we substitute the expressions (12-11) and (12-12) into (12-14). we get
(12-15)
Design of Liguid Retaining Shells 301
For a homogeneous shell with unifonn thickness, K is constant. So, we will have
(12-16)
(12-17)
The key goveming equation of a unifonn cylindrical shell in its axisymmetric behavior
becomes
d' P (12-18)
.-:!..
<bell + 4ß"w = --E.
K
This equation has exact1y the same form as derived in chapter 4. As we have seen, the
general solution to this equation is
In which, f(x) is a particular solution. From a physical viewpoint, this particular solution is
always the solution to the corresponding membrane problem, i.e., the membrane shell
subjected to the prescribed distributed loading.
For a circular cylindricalliquid retaining shell of height H, filled with liquid with density Y,
the particular solution (as found in chapter 4) is
(12-20)
The four integration constants A,B,C,D may be found from four boundary conditions (two at
each edge). However, as we have argued before, for actual water tanks the edge conditions at
two ends do not influence each other. Therefore, two of these constants (involving positive
exponential terms) are set to zero and the two remaining are found from two boundary
conditions. With this assumption, the solution to homogeneous equation would reduce to
Having this solution, we can study the influence of edge forces on the shell. This was done
in detail in chapter 4; we can repeat it
For the shell of figure (12-17), subjected to edge forces, Mo and Qo, the expres,~;"'ns for
constants C and D are,
(12-22)
(12-23)
(12-24)
(12-25)
(12-26)
Design of Liquid Retaining Shells 303
I writing down these expressions. the functional quantities ~, <1>. have been defined e. '"
for the sake of compactness. These quantities are all functions of the dimensionless variable
ßx:
(j!(ßx) e -Bx(Q:>sßx + Sinßx)
(12-27)
----------- - - - -_..---------
+11!xl =e-6x(O>sl!>< .. Sin:lx) w= t ~l ..'lo .ll>Xl1"00 811!xl1
"'(I\xl -e-I!>«O>sBx - Sinßx) <t..,,;;h{ 1
ax-2B'K 2ßHo "(Bx)+ 00 (B")
..
d'w -1
axr=1( [21!M0 (6xl- o.,~(Bxl 1
lIx 1/1 6 (
0
0.1
1.0000
0.9907
1.0000
0.8100
1.0000
0.9003
0
0.0903 Sx . ~ e C
0.2 0.9651 0.6398 0.8024 0.1627 3.0 ..0.0423 -0.0563 -0.0493 0.0071
0.3 0.9261 0.4888 0.7011 0.2189 3.1 -0.0431 -0.0469 -0.0450 0.0019
0.4 0.8184 0.3564 0.61"/4 0.2610 3.2 -0.0431 -0.0383 -0.0401 -0.0024
3.1 -0.0422 -0.0306 -0.0364 -0.0058
0.5 0.8231 0.2415 0.5323 0.2908 3.4 -0.0408 -0.0237 -0.0323 -0.0085
The fundamentals of the force method for shells analysis were outlined in earlier chapters;
the force method of analysis was presented in detail in chapter 4. In this section, the same
method will be used for the analysis of liquid retaining shell structures.
The basic steps in the formulation of shell equations according to force method are: solve
the membrane problem; obtain the flexibility influence coefficients; write the compatibility
relations; solve these equations.
(12-28)
d 1
dw w(x=O)=-2ß2K (2ßMo +q,)
To find the flexibility influence coefficients, we should consider unit edge forces and to
evaluate the edge displacement and edge rotation due to these unit effects. Figure (12-18)
shows the meaning and symbols of influence coefficients as weIl as the edge deformations
due to applied distributed forces, i.e., the membrane deformations.
. --f-•
"z =....=1 Kg-mlm
(b) (c)
(a)
The expressions for influence coefficients, as derived from relations (12-28), are
o - 1
11 - 2ß1K
(12-29)
For a liquid filled container, with radius a and height H, the edge deformations due to liquid
pressure are easily derived from the expression (12-20); they are
(12-30)
or
(12-31)
w - ~ ~e(Sx)
(12-32)
306 Design and Analysis of Shell Structures
The assumed hinged condition can be realized in concrete liquid retaining shells. Figure(12-
19) shows a design-construction detail of wall-floor connection in which this assumption is
fulftlled.
wall
water stop
reinforcement base(floor)
shear key
To determine the complete force field, the membrane forces must be superposed on the
solution given in (12-32). The schematic diagram of complete field is shown in figure (12-
20).
bending theory
membrane field
(a) (b)
(12-33)
By solving these two simultaneous equations for unknown edge forces Xl and X2. we obtain
the following general formu1a:
(12-34)
The bending field resu1ting from these edge effects can now be determined by substituting
Xl and X2 into the expressions (12-23) to (12-26). The hoop force Ne can be obtained from
relation (12-11). To find the complete force field resulting from these edge forces we must
add the membrane field to this bending field solution. Graphs of variation of intern al forces
in the shell are shown in figure (9-21).
in inner fibers
~
ompreSSion
bending theory
tension
compression in outer fibers'
In order to cast these expressions into more useful forms, we define a dimensionless
parameter, Z, as folIows:
(12-35)
}
or
(12-36)
Now, we define the dimensionless base shear force and base shear bending moment as
v -~- (12-37)
YBIiä
with this definition, the expressions (12-36) assume the following fOlms:
(12-38)
Design of Liquid Retaining Shells 309
Figures (12-22) show variations of base shear and base bending moment in terms of the
parameter z.
'X
-
....... ~ 0 .7 '
/ 05
0 .4
6
0.2
:Xz
0 .3
0.2
I 0.1
0 .1
0 o 1 2 3 4 5 6 7 ,0
01134567.
Z: _ _
H_ Z:_H_
(a)
~ v'TG (b)
Figure (12-22) VaIiation of: (a) base shear and, (b) base moments
with dimension1ess height parameter, z, in a cylindlical tank with fixed base
This case is realized whenever the wall-base connection is monolithic and the wall stiffness
is comparab1e to that of the base. In this case, the rotation of the wall must be equa1 to that
of its base. The wall is subjected to linearly varying hydrostatic press ure while the base is
resting on its tensionless foundation and is acted upon by the uniform pressure equal to q =
y H. Figure (12-23) shows a situation in wh ich part of the base has a tendency to be lifted
from its foundation in order to fulfill the compatibility requirement with the wall.
To analyze this case, we use an approximate method employing the moment-area theorem
well-known in structural analysis. Viewing figure (12-23), we note the the width, b, of lifted
"ring" region is unknown and is to be determined together with unknown redundant base
forces.
Using moment-area theorem for a radial strip of the circular base, we may write expressions
for edge rotation and lateral displacement of the base plate as folIows:
1 . !lL. (12-39a)
~AB - 3~ Mob - 24Kf
1 b2 _ !li:.-
wAB = 6Kf Mo 24Kf (12-39b)
Due to high axial stiffness of the wall, the lateral dis placement of the base plate at the edge
can be assumed to be zero. Hence, by setting the second expression equal to zero, we find
b=2~
Substituting this value into (12-39b), we fmd
~AB =3K-
f
1/M~ -=022
q
X
2
(12-40)
Having determined this expression, we can now use the classical plate theory to write down
the expressions for the influence coefficients of circular plate at its edge. These coefficients
are
f -a(1- v)
On = Et
(12-41)
f -1
°2-r2 = 3Kt q fJ
X2 b -1
D22 .. lK f ~ X2
Ci""
On the other hand, we had previously determined the influence coefficients for the
cylindrical wall (expressions (12-29)). Therefore, we can, now, combine two sets of
influence coefficients to determine the tlexibility intluence coefficients for the wall-base
system. These expressions are
w f
Oll =OU +On =- 2ßTR + a(l-v>J
(1 Et
(12-42)
Design of Liquid Retaining Shells 311
We can write down the compatibility relations for the case of wall-base system:
[ 1 _ a (1- \I ) ) X + 1 X = -ra2 H
~ Et 1 2ß2K 2 Et
(12-43)
~X
1
"S-K 1
-(-+-1 1 [;2
K 3K f
-lX
q 2
=ra-
Et
l
r
Figure (12-24)
M ~ Mo
DuetoMo:
W1 •
0
2ß!"'K dx ·-n (l2-44a)
dw2
% Qo
DuetoQo: W1 - 2ß1I{ dX - - 2ß2K (12-44b)
312 Design and Analysis of Shell Structures
I
wlx-O + w2x-O
1 - wl x-B - 0
Using relations (12-44a) and (12-44b), relation 0 0 = - ß Mo, adding up the resulting
expressions, we arrive at relations applicable to the case of the wall monolithic with roof
slab. In this fashion, the condition w = 0 at x = 0 will be satisfied. Then, the wall-sI ab
rotation at the top would be
(12-45)
In this case, the expressions for internal force field in the wall will be
M - Me
-ßx Cof·ßx
x 0
(12-46)
(12-47)
roof
i
1 x
tTs
t_ I-
wall base
(a) (b)
To perform an analysis of the wall-roof-slab system. we need the expression for lateral
deformation of a circular elastic plate. The deformation of plates can be determined through
the use of plate theory. Treatment of this theory is not given in the present text but is dealt
with in books on the analysis of plates. Only the result of such analyses will used in this
section. To make this presentation more useful, the results of circular plates analyses are
compiled in table (12-2).
Table (12-2) Internal force and deformation field in elastic circular plates
l&+ T+
M
2 ~(a2 _r 2 )2 mta2 - rl)(~ a 2 _ r 2)
w Cl C2 logf+Cl 21<11". V) (a' - r')
dw
ar ~t elf+ Cl!-
Mar
- i'fI"+Vr - itt (a.'l: - r 2 ) -m(ffi a2 - r2)
dw
ar ~+ ~l. _ C2 1._ -~
M.
- rlK<az - ]r!) - Ih (ffi a 2 - 3r2 )
j
c, -~
2M,.
-~ - ~i!~
c, 0 0 0
M,p' ~5+\I
C,
~ 64,; 1"'+V
i"i<fITV1 64K
.'
Mt -I(~+V~ +M. [a2 (1+v) -r2 (1+3v)]
+M. -8 n
M.
Et 3
in which k - S is the bending stiffness of circular plate.
s - 12(1- v 2 )
Now, since the wall-roof connection is assumed monolithic, the compatibility of rotations
must be fulfilled. Therefore we have,
:xa2 1
Ps a3 a
D (1 - v 2) - 2ßK • X 2 = 8k (1 + v ) + .-k-(,.;:l....,.+-v~) x 2
s s
from which we obtain
(12-48)
Having obtained the expression for edge moment, Xz, we can write the expressions for
displacement and force field in the wall and the roof slab; we consider two special cases:
(12-49)
2yKk ßa 2
~ s e -ßxCb~Sx
= D(l-v) 2Kßa+k (1+v)
S
(2) Empty container with loaded roof
p a3
w = S e -ßx Sinßx
8S[2KSa +K (1 + v)] (12-50)
s
Design of Liguid Retaining Shells 315
The force and defonnation fields can be detennined by superposition of the expressions in
cases (1) and (2).
The foregoing analytical results can be tabulated or plotted as design guides to the
proportioning of cylindrical containers. In addition to these analytical results, numerical
calculations can also be perfonned by means of, for example, the Finite Element method. To
provide so me examples of design tables, tables (12-3) and (12-4) are presented. In these
tables, the coefficients for the moments and the hoop forces in cylindrical walls, with various
loading and boundary conditions, are given.
Table (12-3a) Moments and hoop forces in cylindlical shell walls with various
loading and boundary conditions
!JDI
Moments in cvlindrical wall. M.
Triangular load
Fixed base, free top
M, • - co~f u}(1 ft Ib
)0' per h
Negat ..... e Sign Indlcales tension In the outside
l.0.5/1 I 0.6/1
m :
t'
"
.
0.711
"'"
0
'" I
•
I 0.8/1 I o.g/l I 1.OH
•
.
'
'
Sotlom
04 •. 0005 •. 0014 •. 0021 ••0007 -.0042 -.01$0 -.0302 -.0529 -.08'6 -.'205
08 .. 0011 .. 0037 +.0063 •. 0080 •. 0070... 0023 -.0068 -.022< -.0465 -.0795
1.2 .. 0012 .0042 •. 0077 ... 0103 .. 0112... 0090 .. 0022 -.0108 -.0311 -.0602
• 6 •. 0011 ... 0041 ".0075 •. 0101 ... 0121
".01" ... 0058 -.0051 -.0232 -.OS05
2.0 ... 0010 •. 0035 •. 0068 •• 0099 +.0120 •. 0115 •. 0075 -.0021 -.0185 -.0436
30 .. 0006 ... 0024 •. 0047 •. OO7t •. 0090 •. 0097 .. 0077 +.0012 -.0119 -.Ol33
40 • oool •. 0015 .. 0028 .0047 •. 0066 •. 0077 .. 0069 •. 0023 -.0080 -.0268
50 .. 0002 ... 0008 •. 0016 ... 0029 .... 0046 ... 00~9 .. OO~9 •. 0028 -.0058 -.0222
60 .. 0001 ... 0003 .0008 ... 0019 •. 0032 ... 0046 •. 00~1 ...0029 -.004' -.0'87
8.0 .0000 ... 0001 ... 0002 ... 0008 ...0016 ... 0028 ... 0038 ... 0029 -.0022 -.0146
10.0 .0000 . 0000 .000 • •. 0004 •. 0007 .. 0019 ... 0029 ... 0028 -.00'2 -.0'22
12.0 0000 .... 0001 ... 0001 ... 0002 •. 0003 ... 0013 ... 00'23 .. 0026 -.0005 -.0'04
140 0000 .0000 .0000 .0000 .000' ... 0008 .. 0019 .... 0023 -.000' -.0090
'60 .0000 .0000 - 0001 - 0002 - 0001 .. 0004 ... 0013 ... 0019 ... 0001 - 0079
c[iJLj
Tension in clrcular rings, Nil
Tnangular load
Fixed base. free top
\'~ - eoel ~ ..d/U Ib Pt:f tt
Po:.l1lve !>IYU ln(hcalt!S lenSlon
,
JI' f Top CoeH'lclenlS i.t pOint
-I
JJ/IOOH I 0_1lI i 0 '2H I 0 lH I 04H I 0 ~I/ I 0.611 I 0711 I 0 BH I 0.9H
o 04:.0 149 .. 013-04 .0120 .. 0.101
,0 OB21'O 066 .. 0 <Hg .. 0029 .. 001004 .000.
o• • 0263 .0239 .0215 .0190 .tJ lW .. 0130 .0.096 .. 0.063 .. 0034 .. 0010
1.2,.02B3 .. 0271 .0 2~4 .. 0234 .. 0209 .. 0 11$0 .. 0142 .. 0099 .. OO~ ,0016
1 6 !.O 2('~ .. 0268 .. 0268 .0266 .. O.2~ ... 0226 ,0185 .. 0134 .0075 ... 0023
20 .0234 .. 0 2~1 .. 0273 .0 28~ .0 2e~ .02704 .. 0232 .. 0172 ,01~ .. 0031
30 .0134 .0203 .. 0267 .0322 .0357 .0362 .0.330 .0262 .. 0.157 .0052
40 .. 0067 .. 0 164 .0 2~6 .. 0339 .0403 ... 0429 .. 0.409 .0334 ... 0210 .0073
;0 .0025 .. 0 137 .0245 .. 0346 .. 0428 .0477 .. 0469 ... 0398 .. 0259 .0092
~ 0 ;.0 OHi .. 0119 .0234 .0344 .. 04041 .0504 .0514 .0 -447 .0 112
.030'
b 0 ' -0011 .0104 .0 ~18 .0 33~ .0443 .. 0534 .0575 .0 !lc30
J
.. 0381 .0 1!11
100 -0011 .. 0098 .0208 .0323 .0437 .0 ~2 .. 0608 .. 0589 .0.4-40 .0 179
1101-0005 ·0097 .0202 .. 0312 .0.>91.0 >4J .0628 .. 0633 .. 0 -494 .. 0211
1~ 0 ;-0002 .0098 .0200 .. 0 ~O6 .0420 .. 0539 .. 0639 .. 0666 .. 05-41 .. 0.2"41
160 0000 .. 0099 .0199 .0304 .0412 .. O~ll .. 06-41 .0687 .0582 .. 0165
316 Design and Analysis of Shell Structures
Table (12-3b) Moments and hoop forces in cylindlical shell walls with various
loading and boundary conditions
6
Moments in cylindrical wall. M.
Trapezoidal load • 0 :
Hinged base. free top
M~ .. - eoel )( (tell' +
pI/I) fllb. pe' ft. I ':' I..:
,
Negative $19n mdu:.les tenSion In Ihe outsIde
0.' +,0020 •. 0072 ••0151 •• 0230 +.0301 •. 0348 ...0357 +.0312 +.01" 0
0.8 ... 0019 •. 0064 ...0133 ... 0207 ... 0271 •. 0319 ...0329 •. 0212 •• 0117 0
1.2 ... 0016 ... 0058 ... 0111 .. 0171 +.0237 ... 0280 +.0296 ... 0263 +.0171 0
'.6 .. 0012 ...0044 ... 0091 .. 0145 +.0195 •. 0236 •. O2~~ +.0232 +.01$5 0
20 •. 0009 ... 0033 ...0073 ... OIU +.0158 +.0199 +.0219 +.0205 +.QI4S 0
30 ... 0004 .. 0018 •. 0040 ... 0063 •. 0092 .. 0127 •. OI~2 .,0153 +.0111 0
.0 .. 0001 +.0007 •. 0016 +.0033 .. 0057 •. 0083 •. 0109 •.0111 •. 0092 0
~.O .0000 ... 0001 .. 0006 •. 0016 .. 0034 •. 00~7 •. 0080 •. 0094 ... 007 • 0
6.0 .0000 .0000 •. 0002 .0008 •. 0019 .0039 •. 0062 •.007• •. 0061 0
8.0 .0000 .0000 -.0002 .0000 .. 0007 .. 0020 +.0031 •• 0057 •. 0054 0
10.0 .0000 .0000 -.0002 -.000' .0002 + 001' •. 002~ +.0043 +.0045 0
'20 .0000 .0000 - 000' - 0002 .0000 +.0005 .0017 •. 0032 ... 0039 0
14.0 .0000 .0000 - 0001 -.0001 -.000' .0000 .0012 ... 0026 .0033 0
'60 0000 0000 .0000 -.0001 - 0002 - 000<4 .0008 .... 0022 •. 0029 0
i1~
Tension in circular rings, N tt
Triangular load ___':1 :.
Hinged base, free top
"'e :; 1.. . I:! ! :.
eDel ... wllH Ib per f1
PoslllVe sign .ndledles tenSion 'WH r-: ~ , .
,
JJ~ ~ Top Coeff',cu:,nts ... t pOInt
J); li 0 OH I O.,H 10.2H I 03H I 04J/ I 0 511 i 0611 i 07H . 0 &11 I 091/
04
08
1.2
.00474
.0423
.0350
~O"l401·039~ .0.352 .0308 .02604
~O 402 : .. 0 381
.. 0 J~~ i ~O 361
.0358 +0.330 .0297
'02'~1'0
.. 0249 .. 0
.0362 .0358 ... 0 3-43 ~O 309 .. 0256 .. 0186 .0098
ml
.. 0 0~7
' 65
.007 •
1..00 1<C~
'"
1.6 ~O 271 .0303 ... 034\ .0369 ... 0.385 ... 0385 ~O 362 .. 031-4 ... 0.233 .. 012<4
20 .. 0205 .0260 .0321 .0373 .0411 .00434 ... 00419 I .0 369 ! .. 0.280 .. 0151
30 .. 00704
, 0 .0017
.0179 .. 0281
.0253
... 0375 ... 0449 .0506 .05191.0.79 .0.375 .. 0210 i
·0137 +0367 .. 0469 .0 >45 ~0579!.O!.!.3!·OH7 .0 2~6
50 -0008 .. 0 1104 ... 0235 .0356 ... 0469 .0562 ... 0617I·0w61 .. 0~3
.. 0294
60 1.0 Oll .010] ... 0223 ... 0343 .. 0 ~63 .0566 .. 0639: ... 0643' .. 0 >47
.. {J 327
80 ·0015 .0096 .0208 .0324 .04-4] .. 0564 .. 0661 !.06S7i .. 0.621
.0 38[,
100 -0008 .. 0095 .0.200 .0311 .0 -128 .. 0552 .. 06661 .. 07301 .. 008 .0 433
'20 -0002 .. 0097 .01971.0302 .. 0417 ... 0 ~1 .0664 1 .. 0 7!.oO: .. 0 120 .0477
I. 0 0000 .. 0098 .0197[ .. 0.299 .. 0 "08 .. 0531 .0659! .0 761 1 .. 0 7~2 • 0.~1 J
'60 .. 0002 .0100 .0 198 ~0.299 .0403 .. 05'21 .. 0 65.Q 1 .. 071:..4 i ~O 776 ... 0 ~JtI
Design of Liquid Retaining Shells 317
Table (12-4a) Moments and hoop forces in cylindl1cal shell walls with various
loading and boundary conditions
r
0
I
•
'
'
'
,
DI 10 1/1 : 02/1 I 0 J// I 0'/1 I 0.'11 I 0.611 0.711 I 0.8// I 09// i lOH
I'
80 0000' ~ 0001 ,000' .0008 .0016 •. 0028 •• 0038 .0029 -.0022 -.0146
JI: ! Top
I
201·02341.0251 .027]i .. 028~I.028~ ,0274 '02321'0172 ,01041,0031
I
~ i
30.01]4: .. 0203.02671·03221.0357.0362.0330
40
1
Table (I2-4b) Moments and hoop forces in cylindrical shell walls with various
loading and boundary conditions
W
Moments In cylindrical wall. M.
Trapezoldal load
Hinged bas •• free top
M, - - c:oef )I (ILlli +
'Pli') ft Ib. pe' H.
I .:. r 0 I w .:
Negallve S'9" ,nd,cale. lenSlon In Ihe oUl.ode
•
CoemCI."ls .t po,nt BOUom
/I'
Di 0.111 0.2/1 I 03/1 I 0,4/1 UH O.6H 0.711 I 01/1 0.911 1.0/1
0.' •. 0020 ...007'2 ...DU' .. 0230 .. 0301 +.03.' •. 03!t7 •.031'2 •.0197
01 •. 0011 ••0064 •. 0133 •• 0207 ... 0271 .. 031' .. 032'9 ••0192 +.0187
1.1 .,00'6 •. 0051 ••0111 .. 0177 •. 0237 .. 0210 .0:196 •• 0263 •. 017\
, 6
10
.. 0012
.0009
...0044 .009'
•.0033 ••0073
.0145 •. 0195
•.011 • ...OISS
.. 0236 .. 0255
•. 0199 •.02"
•. 0232 +.QU5
••0205 •.0145
'00 0000 .0000 - 0002 _.0001 .. 0002 .. 0011 .. 0025 .. 0043 .004 •
110 .0000 .0000 - 0001 - 0002 .0000 !. 000. • 0017 .. 0032 .. 0039
140 .0000 .0000 - 0001 _ 000' - 0001 .0000 I. 0012 .. 0026 .001l
160 0000 0000 0000 _ 000' - 0002 - OCKM I .. 0008 .0022- .0019
(5) Empty tank with roof under uniformly distributed load of Ps = 500 kg / m 2.
In all cases, we assume that the Poisson's ratio, n, of concrete is equal to zero .
em
I
~
l
,
• o. : I,
-
m
H:6
t-- I :
;
I
Analysis:
In the analysis that folIows, we assume that the conditions at one end of the cylinder wall do
not affect the field at the other end: the wall-base and wall-roof interactions have been
decoupled from one another; combined fields can easily be detelmined by superposition of
these simple cases.
.!!..
aB'
= ~ = 0 75 a---g- .
t _ 0.4 - 0 05
320 Design and Analysis of Shell Structures
also
= 10
Ir 2(0.7357)3
-4077 7
(i06.7 X 10S)a(6x )+
(1000) (8)2(6-x)l
2 x 109 xO.4 'J
.. l.... _ -4077.7
Mx 6 00 i;(Bx) - 0.7357 t; (ßx) = -5542.6 t (ßx)
Variations of internal forces in the wall along the height of the shell are shown in figures
(12-27)
/
6.0 60
so so -
/
c---
40 4.0
/ \
\
JO JO
,
(
\
20 2"
~
l/
'0 ',0
~ r-.. 00
JOOOO 20000 '0000 0 .1000 00 - 1000 Mg-1ft,
Ne N.
(a) (b)
X =M =1000(8)3
2 0
0.05 [0.75-
2(3)1/2 ;m
1 {O.OS)l/2)=4286kg-1Y1/m
2 10 9 x 0.4 { 1
8 2(0.736)2 (106.7x 10s[42861j1Cßx)
7231 }
- O.7368eßx)]+8000e6-x)
"\c-r
1
[28Mo'(Sx)+~~(8x)] -42864!(Bx)- 2x 7231
0.736
t"(Q~)
~ o>A
or
Mx ,. 4286 ['(Sx)-2.2~l;eßx) ]
322 Design and Analysis of Shell Structures
Variations ofthese forces along the cylindrical wall are shown in figure (12-28).
50 f----l--j,.<---j '0
40 f----l-;---+--j i\
\
40
)0
JO
\
10 f----'tt----1 '0
'Ol---l---"-k--l '0
J
V
----
QO ' - - - - ' - _ - - ' - _ - " 00
Kg/m )0000 000 • ·6000 • 4000 • 2000 0.00 - 2000 "-,-""",
Ne (a) (b)
_
Eta _ 2 xlO' (0.4)' ,-....-.ll
k l2(i-v i ) 12 s 1.067 xl0 7 """m rn
I a(l-v)., [ I 8.1-'1
-\t-- (- 2S1K+ Etf ,) ... - - 2(0.736) 3 (1.067}lO"+ 2 xI0'(0.3y=1.039 X 10
1 1 -7
'2 .. 2ß2K" 2 (0.736) 2 (1.067) " 10' - 0.865 x 10
) .. -(~+~ l
22 PI'- Ai· 3kf
Ix.}= - (
2
1
0.736(1.069 x 10 7 )
+ _1_ x l
16000 3(0.45 x 10 7 )
fX.-l
--;.
Xl • -7096.Ji<9I'm
b. 2
i{Xi.
q
2/2974.67. 1.4oa'"
6000
Having obtained the values of edge forces Xl and X2, we can write the expressions for
internal force variations as in the previous cases.
324 Design and Analysis of Shell Structures
Therefore, the internal displacement and force field in the cylinder wall is
0.45 x 10 7 1 ~
(1.06")) (10' ) (0.736) (8)+(0.45)><10 ' . (0.736) (8)· -I
or
and
Ne = ~t w = 6.4 x 10-~ [ ~ - 5.874 X 10- 3 E;;(ßx)1
Figures (12-29) show variations of internal forces in the wall of this problem.
60
V
6.0
1\
\ L
50 5.0
40 4.0 t--
3.0 30
\
\
2.0 20
1,0
\ '.0
00 1\ 00
000 2000 4000 6000 KCJ Im - 20 0.0 20
Ng
".
(a) (b)
We have assumed that the roof slab is subjected to uniformly distJibuted verticalload of Ps
= 500 kg / m2. In this case, we use relations (12-49) and (12-50). The resulting expressions
for displacement and force field in the cylinder wall are
-~
; 3.342 x 10 ';(ßx)
; Et w ; 2 x 10 9 y 0.4
N a 8 w = 10 8 w
5.0 60
~ 1/
so
1/
0
'0
/
V
/
1/
)0
20
'.0 '0
0.0 00
.. 50000 00 5000 10000 ISOOOK9,m '2000 0.0 - 2000 - 4000
(a) (b)
As we have mentioned before, the combined field of fOt·ces and deformations can be easily
determined by simple linear superposition of these individual solutions.
326 Design and Analysis of Shell Structures
A reinforced concrete circular cylindrical liquid retaining tank, figure (12-31) will be
designed in this seetion. Overall dimensions of the tank are,
T
'0
!
1- D .~
·f
This container is assumed to be free at the top and to be monolithic with its base, which is of
a relatively thick slab. The properties of concrete and reinforcing materials are as folIows:
f c =200 kg/m 2
()t = 12 kg/cm 2
Allowable tensile stress of concrete in bending:
()tb = 17 kg/cm 2
fs = 1000 kg/cm 2
Design of Liquid Retaining Shells 327
Design:
We assume a trial thickness value of 12.5 cm for the cylinder wall. Numerical values of
parameters needed for necessary calculations are
-1
eßH = el. 75 xl0 x 400 = 1097
Since the base slab is thick and its connection to the wall is monolithic, the relations for a
fixed base can be used. Calculations to determine the internal force field have been
performed before and will not be repeated here. Without entering into the detailed, but
straightforward calculations, we only present the final results in the form of force diagrams
as shown in figure (12-32). These results have been obtained assuming the coupled
interaction of both ends of the cylinder. Had we assumed that the ends effects were
decoupled, the results would have not been affected too much.
= 551 kg-an/an
Having detennined the internal force field, we are now ready to design the required tensile
reinforcements
On the other hand, according to, for example, IS (Indian Standards), the minimum steel ratio
is
'\un 03_2.50
Pmin ="""'Hl= • '"B""" x .1=0.293%
A
0.293
.=p . • bh=-100 x100x12.5=3.66
cm2/m
s,nun nun
We also need to check the maximum tensile stress in the concrete, so that we are assured
that there is no cracking in the concrete, and therefore no leakage. The tensile hoop stress in
the concrete is
o 119.3 kg/an 2
c 12.5 + (14-1)xO.1l9 = 8.5 <
So we are assured that there will be no vertical cracking of the cylinder wall.
The maximum vertical steel reinforcement in the inner side of the wall is
The minimum area of steel reinforcement, placed in the inner and outer faces of the wall
thickness is
The reinforcing steel bars should be reduced proportionately at the sections with sm aller
hoop force and bending moment.
Figure (12-33) shows the reinforcing detail in a vertieal seetion through tbe container.
" /
In designing of reinforced eoncrete liquid retaining tanks, due consideration must be given
to the detailing of the reinforeing system; improper detailing could lead to loeal and
sometimes global failures; a poor detailing could make an otherwise properly designed
container unserviceable; proper detailing is particularly important at the loeation of water
stops, construction joints, and the eorner points. Figures (12-34) show two types of proper
corner reinforeement details and a third one whieh is considered inappropliate.
correct correct
/L
haunch
(a)
/
/
~ (c) (d)
U' I /
.:-
main reinforcement
temperature steel .
main reinforcement
/
In some cases, and specially when the diameter of cylindrical wall is large and no inner
columns are permitted, the roof would be a dome. Other cylindrical shell structures may
have domed roofs: nuclear power plant containment shells and large circular halls are
examples. The cylindrical wall can be connected to its domed roof with or without a
stiffening ring.
We analyze and design wall-dome shell systems by me ans of the "force method" of shell
analysis. Fundamentals of this method have been discussed in various places of this text. In
the previous sections of the present chapter, this method was applied to the analysis of
cylindrical wall with flat roof. Now we apply the same method to the analysis of wall-dome
system.
Consider the cylindrical wall-dome system shown in figure (12-37a). The force method of
analysis, consisting of membrane analysis, corrective (bending) analysis, and superposition
of fields, will be applied in the following fashion.
The horizontal, H, and vertical, V, components of membrane reactions at the base of the dome
are
V = NCl Sina
The vertical component is carried by the wall in a membrane state and thus would not induce
radial displacement in the wall. The horizontal component produces radial displacement as
weil as rotation in the wall. The values of these deformations are
s 1
0 10 = 2ß 3K NaCos Cl
(12-51)
These membrane deformations are shown in figure (12-37b). If the cylinder is a container
filled with liquid, then the membrane deformation produced by internal pressure must also
be added to the above expressions. For a container filled up to the point of intersection,
these additive terms would be
_ya 2
EI: s (12-52)
332 Design and Analysis of Shell Structures
OS
20 dome-wall forces of interaction
membrane displacements
° s
0 11 = Oll + Oll
2aAsin 2 a 1
- 2 ß 3K (12-53)
~
° 5
0 12 = 0 12 + 0 12
2,l.2 S ina
Etn + 2ß 2 K
1
° s 4 ,l. 3 1
0 22 = 0 22 + °22 = Eat.
D - BK
Design of Liquid Retaining Shells 333
Having determined the influence coefficients of the system, we can now write the
compatibility relations for the structure. These equations are
(12-54)
Substituting the relevant values from relations (12-51) to (12-53) into these equations, and
solving these equations for the unknown forces, Xl and Xz, we complete the corrective
(ben ding) analysis.
(3) Superposition
The results of membrane and bending analyses performed in (1) and (2) should be
superposed to yield the total field in the cylindrical wall-dome system.
In the following section, we present a sampie design example using the results of this
analysis.
_ 3 1/4 11m
ß- 132 x (0.35)2 =0.62
The goal is the design of this structure including the reinforcement design of the wall as
weIl as its domed roof. Allowable concrete and steel stresses are assumed to be the same as
in the previous example. The wall is assumed to be high enough so that the boundary
effects at one end can be decoupled from those at the other. The container is assumed to be
filled with water.
334 Design and Analysis of Shell Structures
2R= 2 6.00
1 30 x O.9
ED 10" 1. 4 7 3 x 10 1 q + x 1 + O. 9 q
2 x 0.62 3 x3 .57 x 10- 3
2357760
Design of Liquid Retaining Shells 335
Influence coefficients:
ED = 2 x 3 0 x 22 • 8 x 0 • 4 3 8 2 1 -20]6
11 0.1
2 xO.62 3 x3.57 x10- 3
2 x 2 2 • 82 x 0 • 4 3 S _ _ _ _-=1_ _ _ _ .4918
ED 12 = 0.1 +
2 xO.62 2 x 3.57 xlO- l
4x22.S 3 1
E)22= 30 x 0.1 - ---=------= 15351
0.62x 3.57 x10-
3
Compatibility relations:
Noting the negative sign of H, and referring to figure (l2-39a), we concJude that the
direction of horizontal force must be reversed.
(a) (b)
This type of analysis was perfonned before. Using those relations, and referring to figure
(12-39b), we see that the shear force and bending moment at the junction of wall and its
base are
35 0.5
(TIöO) { 6.5 1 (~) 0.5l
1000x 13 2 31/4 x x -3T"1/"7"4'- • 1300
13 --2- I
Q = -922t<g/m
o
35
Mo _ Mo TIöO x
ya l -lOOOx 13 3
2[3JO. 5
Kg-nv'm
~ M =6408
o
Now we can use the results of corrective analyses performed in part Ca) and (b) to find the
total internal force field in the system.
Ne = 19691[~(~x) + 4.846(ßx)]
E
..... ----,
gl e I
o
.0 '"
a.
.0
E
I
dome effects base effects . total field
i
E
,g
.c
Öl I functional coefficients QJ ~
E
--- I c:
.!!!
I
x I I Ne No ~x
,- Mx Ne Mx i
I exl
":::'
~ (~x) (ex) 6(Sx) ((ßx)
Kglm (Kg-!!Vrn) Kg,im (Kg-!!Vrn) (Kg/m) (Kg~ I
(m) Kg/m
-- I
---r-~
~
).0 I 0 1.000 1.000 1.000 0 84500 -1106 -184 6408 -797 6224
1.50 0.93 0.552 -0.0792 0.2365 I 0.3158 65000 -5212 -67 1-40031 -li60 19756 -1227
i
2.50 1.55 0.2171 1-0.2070 -0.00495 0.2122 52000 -8606
I
466
1-12460 -1765 30934\-1299
I
3.0 1.86 0.1053 -0.1933 -0.04408 0.1493 45500 -9241 I 964 i -5799 -1546 30460 \-582 I
I I I
3.5 2.17 0.0302' -0.1586 -0.0642 0.0944 39000 -8007 1639 1-611 -1210 30382 429 I
4 2.48 -0.0144 -0.1176 i -0.066
I I 0.0516 325001 -4548 2268 Ii -860 30220
I
i 1603
!
I
2) Reinforcement design
Hoop reinforcement:
If the allowable tensile stress is 1000 kg/cm 2 for reinforcing steel, the hoop reinforcement
required at the top of the wall would be
To control the cracking of concrete due to hoop tension, we first calculate the tensile stress
in the concrete, that is
115160 22.80Kg/cm2
°t = 35 x100 +(14-1) x 119.32
This value is higher than the prescribed allowable stress, so, the concrete may crack. To
reduce this possibility, we can increase the thickness of the shell at this region, or provide
the wall with an stiffening ring at thc top_ The magnitude of hoop reinforcement could be
reduced at lower parts of the cylinder. In all regions, the stress in the concrete must also be
calculated so that the cracking of the concrete can be controlled.
338 Design and Analysis of Shell Structures
In all parts of the shell, a minimum hoop reinforcement of the following magnitude should
be placed in the wall:
250
p. = 0.3 - 350 xO.l = 0.228%
ml.ll 2
As,min = 0.228 x 100 x 1~0 = 98%an Im < As
We must choose an appropriate wall thickness to prevent cracking. The wall thickness
calculation, based on allowable bending stress, is as follows:
We conclude that the assumed wall thickness of 35 cm is not sufficient for the lower region
of the container; there the ben ding moment requires a thicker wall. A possible solution
would be to provide the container with a "haunch" at the lower part, i.e., at the wall base
junction.
Vertical reinforcement:
The next step is to determine the value of required bending reinforcement which is placed
vertically along the shell wall. The allowable bending moment, corresponding to minimum
reinforcement (As min =7.98 cm2), is
In most parts of the wall this bending capacity is higher than the existing ben ding moment.
hence, a minimum vertical reinforcement would suffice in the middle region of the wall. We
place a minimum reinforcement throughout the wall and then provide the lower and upper
parts with additional reinforcements.
In the top and bottom regions, the following values for additional vertical bending
reinforcements could be calculated.
In the lower region:
The minimum vertical reinforcement is placed at both faces of the wall thickness. However,
the additional reinforcements are placed at each side of the wall required for the
corresponding sign of the ben ding moment.
~I'P"··
E ;
'"
E
'" ,'"
~
~'
/
E horizontal section through the wall
vertical section
~~~~~~~~~~
Figure (12-40) shows the reinforced detail of the container. In this figure, the dome and the
base are also provided with appropliate steel reinforcements. The calculations pertaining to
design of these latter reinforcements were not presented in this section. The design of the
dome of this container can proceed along the lines detailed in chapter 8.
340 Design and Analysis of Shell Structures
Problems
P 12.1 - The elevated tank shown in figure (P 12-1) is called an Intze tanle The main feature
of an Intze tank is that the cylindrical wall is connected to a domed bottom by means of a
conical shell; the lateral thrust at the wall bottom intersection is reduced and only a
remaining vertical reaction must be carried to the vertical supports. Therefore, an Intze tank
does not require a thrust ring.
The reinforced concrete Intze tank of figure (P 12-1) is to contain 100 m 3 of water. The
sum of dead and live load on the domed roof of this tank is 4000 N/mm 2. It is assumed that
the maximum water level in the tank (freeboard) is 30 cm below the top of the wall.
(1) Find the membrane and bending forces in the wall and the roof. In your calculations,
assume that the Young's modulus of the wall and the roof are the same, and the Poisson's
ratios are zero. Plot the internal force diagrams.
(2) Design the reinforcement for the wall and the roof assuming allowable stresses for the
reinforcing steel and concrete equal to fs = 100 N/mm2 and f Cl = 2 N/mm 2 (concrete
tension), fce = 15 N/mm 2 ( concrete compression), respectively.
-I- A
\ - 6.0 m----j
~'V_ _ I___~~
~. -45cm I
#(
I
5.1 m
I
Xl~
~X2
~X,+XJ
.;;;:::::;;/x.
lJ
XJ-+W;.-
X.
P 12.2 - On some occasions, the cylindrical wall of a container is connected to its domed
roof by means of a stiffening ring. Consider the container of section 12.11 again. In the
present problem, assume that there is a stiffening ring between the wall and its roof.
Perform the design of the wall with such ring. Choose some appropriate dimensions for the
ring, design reinforcement for it.
(Hint: figure (P.12-2) shows the wall-roof junction with the ring. As you see, there are four
redundant forces. Write four compatibility equations relating these redundant forces. Find
the redundant forces and then determine the force fields in the wall and the root).
Design of Liquid Retaining Shells 341
12.1 - M. Farshad, Shell Structures, Vol. I, 1986, Vol. 11, 1987, Shiraz University
Pub1ications,Shiraz
12.3 - S. Timoshenko and S. Woinowsky-Krieger, Theory oj Plates and Shells, 2nd edition,
McGraw-Hill Book Co., N.Y., 1959
12.4 - Krishna, and Jain, Reinjorced Concrete, Vol. land 11, McGraw-Hill Book Company,
NewDelhi
Chapter 13
Buckling of Shells
13.1 - Introduction
Defonnable bodies may become unstable under certain loading conditions and thus have a
premature failure. The phenomenon of instability is particularly important for thin shells
subjected to compressive forces. In such cases, the loadings which produce instability
modes of failure are several orders of magnitudes smaller than the forces causing material
collapse of the structure. A special mode of shell instability is the buckling of shells which
occurs under certain static or dynamic loading conditions.
The design of thin shells is normally dominated by the stability considerations and not
merely the material strength requirements. Hence, the stability analysis of thin shells
acquires prime importance in various problems related to the design of shells.
In the present chapter, we derive a systematic, but simplified, analysis of shell buckling and
obtain some useful relations between the so-called criticalloads and shell parameters. In
addition, we will present a set of useful relations for the linear buckling analysis of shells,
and some buckling design recommendations.
344 Design and Analysis of Shell Structures
Instability is a universal phenomenon whieh may oeeur in various material bodies. The
fundamental eoneepts of stability and instability are elarified through the following
definitions:
The stale of a system is the collection of values of the system parameters at any instant of
time. For example. the positions of material points in a strueture and the temperature field at
various points constitute the state of that system. The state of a system depends on system
parameters and environmental conditions. For example. in a shell structure, the system
parameters are geometrical and material properties. and the environmental eonditions are the
applied forces and thermal conditions.
Stability - The state of a system, at any instant oftime, is called stable ifthe relatively small
changes in system parameter and / or environmental conditions would bring about
relatively small changes in the existing state of the system
Instability - The state of a system at any instant of time is called ullstable if relatively small
changes in system parameter and / or environmental conditions would cause major
changes in the existing state.
Stability and Illstabilily oj Equilibrium - The equilibrium state of a system is called stable
if small perturbations in that state. caused by load changes or other effects. would be
eonfined to a vieinity ofthe existing equilibrium state. The equilibrium state of a system is
ealled unstable if slight changes in eonditions related to that state would force the system
away from that equilibrium state; an unstable system would find other equilibrium state(s);
the new equilibrium state(s) may be in the vieinity of the initial state or may be far away
from the initial equilibrium configuration.
The concepts of stable and unstable equilibrium are illustrated in figure (13-1). This figure
shows a small balllying on a smooth surface. According to thc forcgoing definitions. the
equilibrium state 1 is stable while state 2 is unstable. The relativity of the foregoing
definitions is clearly demonstrated in this figure; the state 1 may be stable in a certain
limited region. but be unstable in a larger domain.
The type of applied loading affects the modes of elastic instability. Loading systems are
classified as conservative or llOllcollservative. Dead loadings, such as the weight of
structures, are conservative forces; time dependent loadings, and the forces which depend on
the state of the system are generally nonconservative. Conservative loadings are derivable
from a potential function whereas nonconservative forces have no generating potential.
From this viewpoint, frictional forces are nonconservative.
Elastic bodies subjected to conservative forces may Ioose their current equilibrium state and
find other equilibrated configurations; this mode of elastic instability is normally of the
buckling type. The equilibrium of the same elastic bodies subjected to nonconservative
forces may become dynamically unstable; the system could undergo oscillations with
increasing amplitude. This mode of elastic instability is calledflutter. Thin panels or shells
in contact with flowing fluids could develop a flutter mode of elastic instability.
Ideal structures could Ioose their stability in one of the following types:
One of the salient feature of static elastic instability, i.e., buckling, is the biJurcatioll of
equilibrium state. At a certain stage of loading, the state of equilibrium of a structure may
reach a point of bifurcation in which there are two possible paths (states) of equilibrium.
The intersection of these two paths corresponds to the so called "bifurcation" of equilibrium
because at such point two states of equilibrium can exist for the same load. Beyond the
bifurcation point, the system can have one of the two choices of behavior. It can stay in its
initial equilibrium regime or it could diverge from the primary path and follow a new path,
the secondary path, of deformation. From the physical point of view, the structure chooses
the path corresponding to a minimum of total energy of the system.
The bifurcation point of an equilibrium state marks the criticd state of behavior of an elastic
system. The primary path, i.e., the initial state of equilibrium, beyond the bifurcation point is
an unstable path while the secondary equilibrium path is stable. The loading condition
corresponding to a bifurcation point is normally called the criticalload.
this column would first undergo axial shortening; the amount ofaxial shortening is linearly
proportional to the applied force. This trend of behavior, i.e., axial shortening of an
otherwise straight column is called the pril1U1ry path of equilibrium of this column. With
increase of applied force the column is further compressed but still retains its straight
configuration until the so-called bifurcation point is reached. A certain value ofaxially
applied force would mark the bifurcation point.
With further increase in the applied force, beyond the bifurcation point, the column can
follow one of the two following paths: (1) it can remain straight and to undergo further pure
axial contraction, or (2) acquire a bent form and undergo lateral as weIl as axial deformation.
This bent equilibrium state is certainly a new trend of behavior that is different from its
straight configuration. This bent form of the compressed column is called buckling. The
straight form of equilibrium, beyond the bifurcation point, is unstable in the sense that a
small perturbation could alter that straight configuration and bring the column to a bent
position; the bent configuration of the column is stable figure (13-2).
Load C
Secondary path
(stable)
Per
o Axial shortening
The buckling of such elements as concentrically loaded straight bars and rings under
uniform external pressure, as normally demonstrated by figure(13-2), is marked by
bifurcation of equilibrium. In these elements, with the passage of applied load through
critical state(s), the structure passes from an unbuckled shape to adjacent buckled
configuration(s).
The existence of two equilibrium paths, emanating from a bifurcation point, and the fact that
the system chooses the buckled shape beyond that point, is charactetistic of a buckling type
which is usually referred to as classical bucklillg. In figure (13-2), a classical buckling load
(Per) can be determined by linear theories of stability analysis. For example, the classical
linear theory of slender compressive members yields the weIl-known Euler buckling load.
The Euler buckling load is directly proportional to bending rigidity of the member and is
versely proportional to the square of its length (third power of radius for rings and arches).
Buckling of Shells 347
According to the shape of the post eritieal eurve, two types of bifurcation of equilibrium ean
be distinguished:
(13-1)
Figure (13-3b) shows the load-deformation eurve, pas a funetion of A, plotted on the basis
of the above equation. Figure (13-3e) shows the load-deformation plot of the same relation,
this time as funetion of o.It ean be observed that both of these posteritieal plots are
aseending eurves showing that the post eritieal state of this system is stable, sinee the
inerease of deformation requires an inerease in the applied loading.
Another type of divergenee of equilibrium path is ealled the "labile bifureation". Labile
bifureation may be symmetrie or nonsymmetrie depending on the type of loading and the
strueture.
Figure (13-3d) shows a simple model in whieh symmetrie labile bifureation eould be
realized. The nonlinear load-deformation relation of this system is
Figures (13-3e) and (13-3f) show that, beyond the bifureation point, both eurves have a
descending trend. This me ans that for an equilibrium state to be preserved, the applied
loading must be decreased. This is the ease for some types of shell struetures.
A model of labile unsymmetrieal bifureation is shown in figure (13-3g). The nonlinear load-
deformation relation for this ease is,
(13-3)
348 Design and Analysis of Shell Structures
Figures (13-3h) and (l3-3i) show plots of this load-deformation relation. This trend is
characteristic of the structures in whieh the postbuckling deformation can only oeeur in a
preferred direetion; this phenomenon ean oeeur in some imperfect stlUctures.
h' tP
lt.
I
I
I
I
&
li
b) CI
~
r,
I
p
, ... , ... ,
,, ",
,,
/
,
i
,
'@ Po,
0
6. &
d) e) II
J=i--j
h _~
1-
( I 6
~)~tts _L
0 Bifurcation point
Defelction
Figure (13-4) Snap-through buckling of a shallow arch with constrained
ends
The equilibrium of thin elastic shells subjected to certain force fields may become unstable
and the shell may undergo prebuckling, buckling, and postbuckling deformation. The
occurrence of buckling in thin shells is quite probable due to the fact that the thickness to
span ratio of shells is usually much lower than other structural elements.
The response of thin shells to compressive forces is essentially very different from the
behavior of other structural elements such as struts, columns, and plates; so me types of thin
shells are extremely sensitive to geometrical and loading imperfections.
350 Design and Analysis of Shell Structures
Geometrical impeIfections include all deviations in the shape of the structural member from
an ideally assumed geometrical configuration. Thus, a slightly crooked column,in
comparison with a perfectly straight bar is considered imperfect. In the case of shells, the
geometrical imperfection is marked by deviation of middle surface geometry from a
conceived ideal shape.
Buckling experiments cani.ed out on shells have shown that some shells are very sensitive to
geometrical and loading imperfections. Thus the buckling load of laboratory shell sam pIes
are normally smaller than the criticalloads that a peIfect system could sustain. This is, on
one hand due the fact that the actual shells are, by production, never geometrically perfect
and also that an ideally perfect conceived loading can never be produced and, on the other
hand.due to imperfection sensitivity of real shells.
The imperfection sensitivity of shells has important analysis and design implications; to
obtain a realistic estimate of buckling strength of shells, geometrical and loading
imperfections must be taken into account. Prebuckling deformation is important and must
be considered in the analysis scheme.
The comparative behavior of bars, plates, and shell fOlms is schematically depicted in figure
(13-5). In these curves, graphs of variations in buckling load with some index of
impeIfection parameter are plotted.
CI)
...
-
(J 1.0
o one dimensional column
domes
pe rfect qodv
imperfect body
o
o 1.0
Imperfection parameter
In figure (13-5), the solid lines represent the perfect system while the dotted curves
demonstrate the behavior of related system in an imperfect condition. As we see, bar and
plate-like elements are virtually insensitive to imperfections while domes are highly sensitive
to imperfections.
It is also noted from figure (13-5) that for bar and plate problems there is no buckled form
of equilibrium at loading sm aller than the criticalload, i.e., post buckling branch never drops
below the primary prebuckling path. This means that the loss of stiffness after buckling in
not great enough to cause the buckled shape to be maintained at lower load levels. This is
not so for shell structures. As we see in figure (13-5), the lass 0/ stijfness after buckling is
so great that the buckled shapes can be maintained in static equilibrium only by the return of
the system to an earlier state of loading, which may be by several orders of magnitude
smaller than the buckling load.
Theories of shell buckling which are linear and based on perfect system, predict behavior
which is not materialized. Linear stability theories are capable of predicting the buckling
"threshold" and in particular the bifurcation point. On the other hand. it so happens that in
some systems and. for example. in the imperfect systems. the bifurcation point does not
essentially exist. Now. in the cases of for example. struts and columns the buckIing load of
imperfect system couId be predicted by linear theOlies through the proper interpretation of
the load deformation relation. For exampIe. in some cases a drop in the value of appIied
force required to sustain astate of equilibrium can be interpreted as the threshold of elastic
instability. However. in shell structures, with sufficiently Iarge imperfection, the stmcture
can pass from an unbuckled equilibrium state to a nonadjacent buckled equilibrium state
even be/are the classical buckling load predicted by the linear theOlies is reached.
Figure (13-6) shows an axially compressed cyIindrical sheII. In this figure. OB represents
the prebuckling branch and the curve Be signifies the classicaI postbuckling behavior of the
shelI. The arrows in figure (13-4) indicate possibIe OCCUlTences of buckling far below the
classical buckling stage. In real situations. depending on the degree of imperfections. the
tme buckling load couId still be even sm aller. It is in such situations that a behavioral trend
having no distinct bifurcation point is manifested by the shell This trend is sketched as
dashed curve in figure (13-5).
(1)
p
~
o
ll.
Shortening
To sum up, the prebuekling and post buckling behavior of shells form two inseparable parts
of a single proeess that is the stability behavior of the shells. Linear theory of shells is not
sufficient to prediet the eomplete behavior and we must use nonlinear theories whieh are
mathematieally elaborate and analytieally difficult. Nevertheless, the stability analysis of
shells by means of linear theories yields some useful information about the buekling
behavior of shells. Partieularly, elosed-form analytieal expressions demonstrating the
interplay of various parameters affeeting the stability with appropriate eorreetion and
statistical eoefficients eould be implemented in those relations so that reliable design
relations would be obtained. With such justifieation, we will now pelform linear stability
analyses of some shell forms in the following sections.
The goal of linear stability analysis of shells is the determination of eritieal buekling loads
and the eorresponding buekled configurations. In linear stability analysis we assume the
existenee of a bifurcation point, and an adjaeent equilibrium state. Moreover, in the analyses
that follow, we assume that the shells have a perfeet geometry and a perfect loading system.
The general methodology of linear buckling instability analysis of shells ean be summarized
as follows:
(1) Consider a deformed equilibrated state infinitesimally elose to the initial unbuekled
configuration. The unbuekled equilibrium state is assumed to eoineide with öe initial
geometry of the shell. These are the assumptions that are also normally made for the
stability analysis of other types of struetures.
(2) Derive the kinematic, constitutive, and equilibrium equations for a shell element
in the buekled equilibrium state. The adjaeent buekled state is an equilibrium state, so the
equations of equilibrium ean be written for such equilibrium eonfiguration. We assume a
linear elastic and isotropie behavior.
(3) On eonstrueting these equations we derive a set of simultaneous homogeneous partial
differential equations in whieh the critical buckling load appears as an unknown
parameter.
Consider a circular cylindrical shell subjected to distributed loading. Figure (13-7) shows a
complete circular cylinder and figure (13-7b) shows a sector of cylindrical shell. In both
figures components (u,v,w) of the buckling displacement field are shown and their assumed
positive directions are indicated.
I
t--.- l -I
I
f w/ i\~
r- JI
z\( / \.
(a) (b)
Consider an element OABC in its deformed (buckled) configuration. Figure (13-8a) shows
the forces, (13-8b) the moments, with their respective derivatives. The buckled shape being
infinitesimally elose to prebuckled configuration, the displacement components,
representing the buckled state, are infinitesimally small.
(a) (b)
To derive the equilibrium equation for the deformed shell element of figure (13-8), we
obtain the displacement gradients necessary for projecting the forces in the various
directions; these are shown in table (13-1).
Table (13-1)
a2. w
_ ( a 2. w + a v) d S
asax Clx
- --dx
ax2.
Abouty
axis
a2. v About z
(~_ aW)dS --dx
asax ax ax2. axis
I
Other two force equations of equilibrium can also be similarly derived. The three
equilibrium equations for the forces along x, S, z axes are
aN x aNs x a 2. w a 2. v a v a 2. W
a - + - - - a Q --a~ --Qe(a+~)
ax as xax2 Yax 2 x x
a2 v 3w
- Ne ('ä"ääX - ax) =0 (13-4a)
_ Q (1 + av + a2 W ) =0
e aae aae2
Buckling of Shells 355
ao x aO e av a2 w a2 w
aa;< +ae +N xe (ax +~ ) + aN x ax2 (l3-4b)
av a2 w av a2w
+ Ne(1+aae+a~e2)+Nex(ax+axae )+qa= 0
In these equations, q is the applied radial force, The equations (B-4)include the non linear
effects arising from projections of interna! forces having sm aller (higher order) magnitudes.
If we neglect these nonlinear terms we obtain the equations:
oN x aN ex (l3-5a)
a--+--=O
ox ae
o
(l3-5b)
In addition to the force equations of equilibrium, there are three moment equations; they are
o
(13-6a)
356 Design and Analysis of Shell Structures
dM ex dM x a2v a 2 vaw
~e + a--;:;-x + aM xe - - - M (---~)- Q a = 0
o 0 dX 2 Y aeax 0 x x (13-6b)
(13-6c)
aM xe aM
BQe ---a-
ae
e
ax
aM ax 3M x
aQx ae+ a--
ax (13-7)
Nxe = N ex
Now we combine the two set of equilibrium equations (13-5) and (13-7); we eliminate ae
and a x and we neglect the effect of shear force ae in equation (13-5b). We obtain the
following condensed equations of equilibrium.
9N X
a-- +
dX
(13-8)
a2w av a2 w
- aNe + a2 Nx ox 2 + aN xe (ox + oeax )
ova 2 w av a 2w
+ Ne (aG"+ oe 2 )+aN ex (ax + OXdS)= _qa 2
Buckling of Shells 357
We derived he constitutive and kinematic relations in chapter 4; we repeat them here. The
relations (4-14), (4-15), and (4-16) (with coupling tenns dropped) are rewlitten as follows:
Et
D
12(1 - v 2 )
I-v
Nxe = D~Yxe
(13-9)
Et 3
K =
12 (1 - V 2)
ae x a2 w
\jJx = --=-
ax ax 2 (13-10)
1 aße 1 aw
ljIe a ae a 2 ä8
1 1 ae x aee
\jJxc 2" (a- -ae- + äX)
358 Design and Analysis of Shell Structures
If we combine the governing equations (13-8), (13-9), and (13-10), we arrive at a set of
equations in terms of displacement components. However, we only substitute part of the
constitutive and kinematic relations into the equilibrium equations to get the following
relations:
aN aN ex
a~+
dX ae- = 0 (13-11a)
aN a Ne
""""ä'8 =
a--.!.i + 0
"ZIx (l3-1lb)
(13-11c)
in which
Equations (13-11) are a set of nonlinear governing relations suitable for stability analysis of
quasi-shallow cylindrical shells.
To obtain the linear stability equations, we drop the nonlinear terms shown in parenthesis in
equation (13-11c) we obtain
aN
x aN xe (13-13a)
a--
dX + ae @
ZlN xe aNe
a-- + 0 (13-13b)
Clx """"äe
K9~w + ...l.N =q
a e (13-13c)
Now, we use the remaining constitutive and kinematic relations to express the forces Nx, Ne,
Nxe in terms of the displacement components. If we calTY out this substitution, we find the
following set of simultaneous differential equations in terms of the displacement
components.
Buckling of Shells 359
(13-14)
a 3W
'ii'~u
v a 3w 1
----
a - -3 + a 2 axaa 2
ax
(13-15a)
2 +v a a 3W
'ii'~" -------
a 2 ax aa
3W
2 a~
1
aa 3
(13-15b)
I-v 2
K'ii'8 w + ---0--a~w
'ii'~q
a2 ax~
(13-15c)
in which
s
'ii'w ='ii'~('i7~w)
Relations (13-10) to (13-13) constitute another form of the goveming stability equations of
quasi-shallow cylindrical shells. The membrane shell equations can be easily derived from
these relations by setting the bending stiffness, K, equal to zero. If we do so, we obtain the
equations
aN x ClN xa
a Tx +ae- o
o
(13-16)
360 Design and Analysis of Shell Structures
These equations express the governing relations of the cylindrical shell in its unbuckled
state. As we recall, we assumed that there is no bending deformation in the prebuckled state.
This assumption is compatible with the premises of linear stability theory of shells.
By solving the equations (13-15) we can determine the membrane force field in the
prebuckled state.
Having obtained the force field in the prebuckled state of the shell, we can perform linear
stability analysis. At this stage, the prebuckled force field Nx • Ne. Nxe are known, so the
terms in parenthesis of equation (13-11) are no longer nonlinear. We must distinguish
between the unknown hoop force Ne (in the buckled state) appearing outside the parenthesis
and the known parenthesized prebuckled hoop force, denoted by the same symbol.
From the linearized version of the stability equations (13-11) we can obtain the stability
equations for flat plates subjected to in-plane as well as out of plane loading, by letting
a~oo. As the result the second term in equation (13-11c) would be eliminated.
Ring buckling consists of an axisymmetric deformation with longitudinal waves along the
length ofthe cylinder, figure (13-9a); there are no axially oriented nodes.
The so-called "chess-board" mode of buckling has longitudinal and circumferential waves
of deformation; the deformed shape of the cylinder will consist of a regular pattern of
"inward" and "outward" deformations resembling a chess-board, figure (13-9b).
The so-called "diamond shape" buckling of cylinders occurs in the post buckling stage of
loading. Under certain conditions, the critically loaded shell "snaps" from one equilibrium
shape to another which requires less energy to be maintained. The second deformed shape
has a pattern of "diamond type" inward and outward deformations, figure (13-9c).
The length of the cylinder is one of the main parameters influencing the general bucking
mode; ring modes occur in very short cylinders, figure (l3-lOa); diamond modes occur
in moderately long cylinders, figure (13-lOb).
Buckling of Shells 361
I tt I t t
ring buckling chess-board buckling diamond shape buckling
If the eylinder is very long, another nonsymmetrie "overall" buekling may oceur, figure(13-
lOe). This mode of buekling oecurs mostly in tubular members and the piping systems
under axialloading.
(c) Long
cylinders
(b) Moderately
long cylinders
Figure (13-10) Effect of eylinder length the buekling modes, (a) very
short cylinder, (b) moderately long cylinder, (e) very long cylinder
362 Design and Analysis of Shell Structures
o (13-17a)
(13-17b)
aN a 2w + N +
X ax 2 s
(l3-17c)
Using the constitutive and kinematic relations we can express these equations in terms of
displacement components. We defme the following parameters:
C1 = N = (13-18)
Et
After substituting expressions (13-9) and (13-10) into equations (13-17), we obtain
(13-19a)
a 2 U 1 + v a 2V v aw + 1 - v ~ =0
ax 2 +2ä"'""" axas--a ax 2 a2ae2
(13-19c)
Buckling of Shells 363
w = 0 =0
To solve the homogeneous simultaneous partial differential equations (13-19) for the shell
with simple end conditions, we proceed as folIows. Consider aseries solution of the form:
u = Lr ASinn8 cosm~x
nrn
w rr CSinn8 Sinm~x
nrn
This assumed solution satisfies the prescribed end conditions. If we substitute the
expressions (13-20) into equations (13-19), and perform the necessary algebraic
manipulations assuming Ä = (n m a J I), for one term of series expansion,we obtain
+ C [n + an(n 2 +).2)] = 0
A\I). + Bn{l+ a[n + (1- \I)A2])
2
+C[l-).2N+a().2 +n 2 )2]_O
,,1
"A
n+an(nJ+),J)
l( A}
l-.IN+a(l.l+nJ)1
B-O
C
(13-21)
364 Design and Analysis of Shell Structures
N (1-\12)
N __x~:-:-_ _ R
(13-22)
er Et S
Where
(13-23)
This analysis inc1udes the ring and chessboard buckling modes as special cases. For
chessboard buckling modes, we take the length parameter, I, very large in (13-22); we find
(13-25)
N cr is a function of A.; the value of A. which would yield the minimum value of criticalload,
N cr, is given by
(13-26)
(13-27)
Buckling of Shells 365
and the minimum axial stress for chessboard buckling in the cylinder is
E t
"cr a (13-28)
w . m1TX S'l.nn
C S l.n-,t- e (13-29)
Note that in this linear stability analysis, as in any other, the amplitude of buckled shape
remains undetermined; this can be found by using a nonlinear postbuckling theory.
The ring mode of buckling can be studied in a similar fashion. It so happens that, for long
cylindrical shells, the critical stress values for the chessboard and ring bucklings are equal.
To render the buckling formula (13-28) more useful for practical shell design problems, we
must apply some correction factors. For example, for reinforced concrete cylindrical shells
the following corrective coefficient has been proposed:
.c=1-0.9(1-e-16
Ilft ) (13-30)
The buckling formula (13-28) can be used for estimation of cylindrical shells subjected to
bending. In that case, the following correction factor has been proposed:
(13-31)
Design buckling strength can be estimated by multiplying the theoretical values by these
correction factors.
366 Design and Analysis of Shell Structures
The buckling analysis of circular cylindrical shells subjected to unifonn external pressure is
similar to the previous section. In this case, however, the effect of hoop forces are more
important, so hoop effects must be retained in the governing equations. In this section the
details of stability analysis will be omitted and only the results will be presented.
per = __ 1_ E (1.)3
4(I-v 2) a (13-32)
(2) Short cylinder with ends held circular, but otherwise unconstrained:
(3) Closed-end cylinders subjected to both axial and lateral pressure (axisymmetric
buckling):
per = 2E (1.)2
"j 3(l-v2) a (13-35)
(13-36)
Buckling of Shells 367
The mode of buckling of concrete cylindrical roofs depends on the applied loading as weIl
as the span to length ratio of the cylinder:
(1) For long cylinders (11/12 > 4) the beam action of the shell is strong and hence the
longitudinal bending stresses are predominant. In these types of shells, the instability mode
consists of buckling of longitudinal strips in the mid-length of the shell at its crown region,
figure (13-11a). The first mode of buckling consists of a single localized wave in the
longitudinal and transverse direction.
(2) For short cylinders (11/12 < 1), the arch action becomes important: the applied loads
are carried predominantly in the transverse direction. The mode of instability consists of
buckling waves appearing in the transverse direction, figure (13-11b). The action of such
shells resembles that of cylinders under lateral pressure.
(3) For cylinders with intermediate lengths (1 < 11/12 < 4), the applied loads are
transferred both in the longitudinal and transverse directions; the buckling modes of these
types of shells are influenced by both the beam and the arch actions, figure (13-11c).
Figure (13-11) Buckling modes of roof cylinders, (a) long sheIl, (b) short
shell, (c) Intermediate shell
One of the phenomena that can occur in postbuckling of open cylinders, as weIl as c10sed
tubes, is the jlattening of such shells. In open cylinders, due to longitudinal bending, the
longitudinal edges are pushed outwards, causing an increase of the shell radius. For closed
tubes this phenomenon is called the Brazier effect. The longitudinal stresses are responsible
for this particular phenomenon.
368 Design and Analysis of Shell Structures
Shells of revolution subjected to applied loading may buckle if the applied load is increased
beyond a certain limit. To perform a stability analysis of the shells of revolution, the
goveming equations of a shell element in its deformed (buckled) state must be derived. In
this section, detailed stability analysis of such shells will not be carried out, but merely the
results of linear stability analysis of shells of revolution will be presented.
per = 2E (.t)2
,.j3(I-v2) a (13-37)
(13-38)
(2) Long conical shell with semi-vortex angle a under axial vortex load P
(axisymmetric buckling):
p _ 21tE Cos2a
er - ,.j 3(1- v 2)
(13-39)
The spherical dome is part of a complete spherical shell. Thus we can use the formulas
presented for complete spherical shells to find the buckling loads of domes. However,
there are some differences in the buckling behaviors of domes and complete spheres:
(1) In complete spherical shells, the instability of the shell could inflict the whole shell,
figure (13-12a), or may be of snap-through type confined a limited region, figure (13-
12b). The occurrence of one of these two instability modes depends mainly on the
thickness, shell radius, and the central angle (or equivalently the base radius, ro) of the
dome. As an approximate formula, we may say that if ro > 3.8 VaT then the buckling
would be local, otherwise it would extend to the whole shell.
Buckling of Shells 369
(2) For spherical domes, the loss of stability can appear by symmetlical, figure (13-12a),
or unsymmetrical deformations, figure (13-12b).Both of these instability modes can
extend to the whole shell surface.
(3) The postcritical behavior of spherical domes differs from that of complete spheres. A
dome can pass to a postbuckling equilibrium position produced by deformations much
larger than those of the corresponding complete sphere.
~~
Buckling 01 the whole shell Symmetrical buclding
(a) (b)
Domes are doubly curved, synclastic, and nondevelopable smfaces. Therefore, they are
generally very strong and highly stable. The critical stability loads of concrete domes are
usually much higher than those of concrete shells with single curvature. Nevertheless,
thin concrete domes with large spans are susceptible to buckling; indeed the buckling
considerations are one of the main design criteria of such shells.
In the previous section, we presented so me theoretical buckling loads for spherical
shells. In this section, we complement those relations by additional approximate
formulas suitable for designing concrete domes.
The theoretical buckling load for a doubly curved elastic shell having the values of
principal curvature 11 RI and 11 R2, under the dead load, is
1 1 (13-40)
qcr = -;==:;~ R 1 R 2
13 (l-v L )
In this relation, qcr is the intensity of buckling dead load, E and v are the Young's
modulus and Poisson's ratio, respectively.
2
Cl = ---;:::::::::::.:;;:: (13.41)
h (l-v 2 )
370 Design and Analysis of Shell Structures
Experimental investigations yield buckling loads much less than the theoretical ones
given here. This is due to imperfection sensitivity of shells which dramatically affects the
stability behavior of shells. On this basis, some building codes recommend values of a
reduction parameter for design purposes which lies in the region 0.05 and 0.1.
Based on these considerations, a more practical formula for the buckling strength of
shells with double curvature would be
(13-42)
50 (--L) 2
(13-43)
a
The buckling load of the shell is quite high, several times higher than its service load.
A compressed strip of the sheIl, cut along the compressed direction, could be conceived
to act as an arch on an elastic foundation, figure (l3-13b). This foundation effect comes
from the restoring action of the stretched strips in the opposite direction, figure (13-
13c).
Buckling of Shells 371
(a) (c)
(b)
Based on linear stability analysis of HP shell roofs, the details of which are not given
here, some approximate buckling formulas for the HP shell can be derived. These
formulas will be presented in the following.
(13-44)
A= 2E Yab~fI
2 f3(1-v2) Vf (13-45)
Problems
P 13.1 - A circular ring of radius Rand cross-sectional bending rigidity EI, as shown in
figure (P13-I), is subjected to constant external dead pressure loading , p. The dotted curve
show a possible mode of buckling of this ring. Show that the lowest critical pressure of this
ring can be obtained by the following formula:
Per -1ill
- R3
Find the expression for the corresponding buckling modes of this ring.
Figure (P 13-1)
P 13.2 - A segment of a circular ring, of radius Rand central angle 2a and bending
stiffness EI, forms an arch with hinged ends as shown in figure (P13-2). This arch is
subjected to a uniform dead pressure loading with intensity p. A possible mode of
buckling of this are is shown by the dotted line in this figure.Show that the intensity of
lowest critical pressure is obtainable from the following formula:
\
~_..-_r-_I
I
/
, /
'/
/ "
Figure (P 13-2)
Buckling of Shells 373
rp
l~~
F
Figure (P 13-3)
374 Design and Analysis of Shell Structures
13.3 - l. G. A. Croll and A.C. Walker, Elements oj Structural Stability, The Macmillan
Press Ltd, London, 1972
13.4 - 1. Kollar and E. Dulacska, Buckling oj Shells jor Engineers, lohn Wiley & Sons,
Toronto, 1984
13.5 - R. Narayanan (ed), Shell Structures-Stability and Strength, Elsevier Applied Science
Publishers, London, 1985
13.6 - W. Flügge (ed), Handbook oj Engineering Mechanics, McGraw-Hill ,New York,
1962
13.8 - 0.0. Brush, and Bo. O. Almroth, Buckling oj Bars, Plates, ans Shells, McGraw-Hill,
Book Company,New York, 1975
13.8 - V. Gioncu, Thin Reinforced Concrete Shells, lohn Wiley & Sons, Chichester, 1979
13.9 - L. H. Donnel, Beams, Plates and Shells, McGraw-Hill Book Company, New York,
+976
W. T. Koiter, (editor), The Theory oj The Elastic Shells, Proc. IUTAM Symp., North-
Holland, Amesterdam, 1960
l. M. Hutehinson, and W. T. Koiter, Post-Buckling Theory, Applied Mechanics Review,
Vol. 12, PP 1353-1366, 1970
Formulas for the Membrane Field in Shells 375
Appendix A
Notation:
p=yH Hydrostatic pressure (normal to the shell middle surface); H is the fluid
height and y is the specific weight of the fluid
Pw Intensity of wind loading, per unit area of the projection normal to the
wind direction
Young modulus
G Shear modulus
v Poisson's ratio
Ne Hoop force
Unless otherwise specified, it is assumed that the shells are supported continuously at their
lower edges.
Co)
.....
cn
~ ..
~~, ~" Spherical Shells
:: --=-- - -t
- ~ r •
'~Q ,~,
System Loading N. "e T
.0.' 0
Gi·
. without opening o.....
I Ul
PI • PE CO!lo. -PI; r 1.C05. 0
PE r(I.!O'•• cO'.l
~ ~
Ul
r" 51n 2 • 0 .i. 2• 2
Px • Ps lint cast ·psiO -~) Ps ~I • ~. 2eo ••1 0 ä
si• •
ijQ b
• 0 without opening
·0 ~
r r
~ p• • Ps co. 2• -Ps "2 ·ps f co.2+
---- --
2
1 h ~ Ifi 1'0 h .1. 2•
0
-yr IIi-II---;-, -yr1j( I • -r-) . IJ
~III $ I1n •
("".;3~ _ ~() ... 14t cos S• _ cos14
cI P z • y(h - reo!!..> - I ,) I • ~ O 2 -leos"»)
3 S1111 .. l 51n 0
2 h 1 cos 24 2 h I cos 2
-yr (i-r- - j (I • J.cos.) J
~l -yr ' h .. C05 • • 'J (J • 1.(o:.J I
"
S pherical Shells
I
System Loading N. T
"a
for points above water level o o
I
.L. for points below water level
I C2 I Pz • y(r - reD •• - hl
';-Y-i-- _y~!!.I_J_ ~3-!!.1 _ 3) •
6 r sln2. r 2
- y ,2 U _ co" _ !!.I - N o
2eo •.).) r •
l+cos.
'T1
for pointe; -above water level ~
J h2 h2 h J 3
- y - (3 - -) ---.--
.ln'/6 (3 - ~) 6 r .In'. c
I»
Ci>
I
e 3 Pz • - y(r - reD •• ' - h) for points below water level
-
0'
"1
tu !!. _ I • ~cos2tj
~ I y 6 r +cos,...- y ,2 (~ _ J • cO'4'I _ ".
~
I I -+--- E:
(1)
-----r I1n 2• 0 2
51n • 0
-P2(J-~) -p ! (I • ----y-l g.
2 sln • "1
~
(1)
. .
~ '.' "without operung II • ~
~
1:L~ \'1"-' \.; 1-'; .~__ ~ 0..
S'
CJ'l
::r
~
(j)
w
....,
....,
w
.....
(X)
o
(t)
~.
Spherical Shclls
:l
~
:l
Na T p..
System Loading N~
:>
:l
Sln~ 0 51""0
f PL -P L -y PL ----r 0 ~
edgeload s In ~ Sln 41
~.
[J)
~o • 0 load at the apex o......
1 1
g -P L --~2- PL --~2- 0 Cf)
PL line load 211r 51n $ 2"nr 51n •
~
~
Cf)
::rt::
n
Spherical Shells
2"....
(t)
[J)
N~ Na T
System Loading
_p .!:. cosO cos. x Pw J cos~ x (cos.(3 x (.os.o -Pw j Sln~ x 13(c05.0 - cos.)
"".l 5103. 51n cJI 51n •
1.(3(C05. 0 cos.) C05 3 $o) 351n 2• - 2 cos 4 $1 cos 34>o t cos 3-P1
((05 3• 0 co.,.Jrp)1
L - -___ _
-~
Spherical Shells
System Loading NO
".
. - 1 ] {eh - nos41)(R. rSlno$l) J( Sln,*.
(~ -. -r:-r-n:t-)--~'lil..~ [-Rh( Sln4l 0 - ~ln4>r
S In 1$
11 .x. ln<,~o <;In} co~4> ~ .. $0) r05 2$) - ~x (51n1l0 co<;1t o - <;1M> x
r2 J '\
- (CO" \1' l 0'" 'If» I cn"'ll-,p·~o)· 3"(co,> ~o-ros 11>1)
!~~ , "
Pz ~ )(h - r (00;4')
~ 1'() ~I
System Loading N. Ne T
I p . ' PE UM 2 J
~ ro 1 - cos 3• _ ~ 2 - 1 co • • • co, • 0 a>
• -PE 1 -~- PE J si.'. !;j'
I 51n • cos ~
P, .. PE cost o.....
Parabola _____ -----~----------_t_- (fJ
I f I 11 PI • Ps 51n. c o . . , •
[
(fJ
I COI •
~ b -Ps r ck. -Ps r 0
PI • Ps 1:0,2. 2
Parabola g
_y;'lh(2tlz• • I) • ~
~ '2'0 ro 2 I
~~ ~ c p.' y(h • f- t, .1 -Yph • .- t, tl co.. • '0 tl. (tl• . ~lJ co.. 0
I r 2
~I _ d PI.
" P r~"
-I' 2 .. J_
co~41 -p r ! :D:~n • 0
_
Parabola
_____ L
______.-L_________L _ _ _ _ _ _ _ _ _ _--'--_
(1) Axis of revolution
Shells of revolution co intersects the merdional curve
~
System Loading Ne T
N.
_p r C05$o - coscfl - ($ - .0) 51n. o
Px '" PE 51n$ F. ~ - 51n$0) Slnot
-P E -T 1(+-tl51n. -
Sln • 0 0
~
~
ii>
Px " Ps SIR. cos" -Ps }(cOS2+. 2S1n. sln. o - '"
sin. sin 2•
Ö'
"'t
b -Ps f~l - 51n.O) -~) 0
Sln • :;-
~ PI. .. Ps cos 1• tIl
I
"'t
Px .. PE sinofl
- PE rleos. - ~.
SIR.
i
~
I 1.:05 • • ofio; In.
tIl
Pt ~ p[ cast
·P f r <;,i-r;.rr<;-~~n-i) • sm$o(cotg+ - ~} I 0
ffi . 0 51n •
~
I( '" lp n( r!f. ,;
t. F "
~
2 'Po 0-
x Sln. o - 2Sln 2)
Shells of revolution Cl
ro
System Loading N~ N. T
!!l.
::l
~
tt • t • f •t 1\ .. Ps 51n4> CO~$ ~ln.p • :.':~.ln.p
0-
I~ b P '" p . C05 2 4> -po ' 0
-- - - .---- -p -, (cos24> - 2s1n4> sln4t ) 0
l!> ~2'>ln.p.'>ln.p 52 0 >
2 2 0
::l
t• R " Ps 1fr Sln 4>0
11>
1-<
'"00·
o......
(fJ
~
(fJ
()
ä....
,,--
1I~. (2) Axis of revolution does not ~
-f (1l)'>'~O - LO'>$))
p)( Pi '> In,~
'I' ~ -4>
1\ - P,_ lO"~
" ,
~0 _ J
,
1t.!1'· ~1.!. _~51_'>11 ~p I(H. 1,>10$)
", "::O'>~- H~ - 1(1 \u""~11
'I (H • t '> I n.p) '>I no$
6 , 1 1 -- co<, \p
-Pi h\l ~~-~-H~t.--diJ2;;)\5t.l~ -p[ rOll ~nz40co~ -- tgofl-
I p)(.=PS Sln~..:os41
2,,2.[1: (I
[
~ la 2 tl.+b 2 ( 2_a2b2~ ( (b 2 • •
Px " PE slnifl PE x Z II
2 a2 sln~tg40 a b 2 • alta2. Ö'
2tg2.~ t. '"1
Ellipsoid I' I ~"n~~-"-z..~) • ln (l.t)l;;,r:--"-I~.!.._
~ I P, • PE <ost
( b(t.;;. ,.'tl b(t.~b2
~
I . ;. . ,,2. 2 rr-'r7
• (.~--
/77 - )-a /Ib- •• -tg.-J
, ____ ~_u _____ + b2.,2,it --1--- f
. Pll~COs~
r:r
'"1
-Ps . ' / I ~-,2J_ -Ps .2 b 2 __ alt,l. PJ
~
T~~~ r1l
Pt ". Ps cos 24'
2/~ a'tg';
~. :r I 'TI
---\---
m"
0:
I <1
2 b 2 __ (a2_~).:?~,t 5"
~~ldll\'P ur," rcl -r----, Cf)
( ,'12 h ) sln'"!!; -~. ftD 2 +( ;12 __ b;?1~;
" h
1[
rn
(.0)
(X)
(.0)
384 Design and Analysis of Shell Structures
..
..e ..=
u
0
.
.... .., ~ 0
...
~
.
';
. ;-
~
;-
.. ...
..
~
~:
...
~
~>
'"
..
...
·
z<>
:<>
~
.
'"=
~
8
~
~ori
;-'" ;-'" ;-> ;-"'
..
c
"'"
~
~
C
=
4 ~
"'
-;; . -; :::-
z
~ ::: ::. :::
I "',;:
..
.1'~
~!;; ~!;: 0
..."' <!
I ~ ~
...
..e ., ..
I
0
u
... II i
i
bl)
c::
:g ~ a N
= !
e II ~
~
.. ~
.."' '" ..=
....·
"' .? ! w
.3 .. ... ... ..
I
. .
"
~ ~ ~
e. e.
·
1----
rr
"
e
~
'"
;;..,
cn
{ 0
~
0 0
"H
0
(
111
"2
0
0
{
~
0
..0
o:l
@
0..
Cylindrical Vaults
T
System Loading "s
".
P" 2:0 (1 - s) x cos.
=~= . P, - P., cos• -,.. r o Sin 2• P"" 'J l.~
- •) .........
Catenary x (2 • 51n 2.) coS+
-- ---------- -
I I 11 "' Pr .. - Ps unq.c:os.
I~ r
b 0 . 0
-Ps 51n.
Parabola PI .. Ps Sln ' •
tI>
....
tI>
p (l-5) sln6cos4 x
. -i-r. C05. (! _ S) I. 2e •• 2.
o
.....
PI .. P., cos.
---~-- · x(3 .. 2 sln 2 if1) -Pw r o 51n 3• p.. 2 unt (f)
Parabola i(
::::
(f)
Pr .. - PE cos. g
~ -pE ~ (1 - S) "PE r o 51n 2 • -PE(~ - 5) 3eo ••
g
2r o
Cycloid
· Pz • PE sin. m
Pr .. - Ps s1n+Cos.
b ~ (1 • 5) I· ~S1n2.
n
Cycloid Ps r o Sln. ·ps r o un 3 • -Ps(~ - S) 451"eo ••
P z .. Ps I1n 2 •
---------
S
rp~ i- (I - ~) ..
u
Cycloid PI· P" COS. -Pw r 0 51n4l<:05. -p l! - S) .!....:.2~~~,
· ); (1 - (054) ~2~t_· w 2 sin.
:.an •
-~-
- -
Cylindrical Vaults
NS ,T
System Loading I ".
Py • - PE cos. -P L i- (R. - ... ) .. -p[ (~ ..... ) x
Py • - Ps sb,,<o" _p 35(1-5)
S 2a2bln~ x ps(~ - ,) •
Ellipse b x (b 2 (a 2cos]41 . . b251n2~) • _p a 2b 2 ~ x lSll~CO<{I (b Z . . lfl)
S 03 2
P, • Ps 5in2~ • 2n2(sln2~, ... cO'lj2cf1)1
~. f
t.:i ~
f.--~ .... _- - - - -
-- . -- _p ~ (I - S) <0-;;-2
w2N~IIl. I b2 2
~
pw(I- S)l::..._~~ ...
Ellipse • P z • P", cos. • '1\(,,2 .... 2) (I ... 3'ln 2tJ . . _P!II .2b Z ~
Ö'
"1
" (I • cos 2CI) '\ln+
0
... 6(b 1 ... a 2 )2 5 ln 2cf1co..,2. I
~} g-
~ ~
----- --- --~
Py • - PE cost
r
0
~
Catenary 0 . PE 51:.
•
Pz • PE slnt
~
'Tl
~ -- [
Py .... Ps 5lnfcosf Ps F (1 . . <;) sln2~ x
0-
0
Catenary b ... Ps r o . . PSc} . . r) 'iH1,'CO'i' S"
2<;'111 2 )
• (I en
PI = Po; "'ln2~ ::r
~ I ~
tn
Co)
00
.....
(..l
(XI
(XI
CJ
ro
Circular cylindrical shells !!l.
"5 N. T
System Loading
Py • - PE co •• !!l.
<n
. o..,.,
p • • PE 51n. U'l
12
"E[6i - vr • rs (1 - s)) sln+ -PE r si'" -PEIl - 2') CDS.
~
U'l
a ::r
~
Q.
~
x sln~ - vrlr)
taI
Formulas for the Membrane Field in Shells 389
.... 0 0
~
.
oe!
0
z
I §
L"
"-
I ~
-oe!
I
!
i
I
z" ~
I 0
! c
w
" : N",~
~ i c.'
I
bJ)
s::
:.a
..e
.3'" ..
~
oe!
~
c.
,
0.
.
"- 0." 0." Co"
'" I u ~
B
Iltltt
0 6.
E
B
'";;...
CZl
..... 1
w
CQ
o
~
j(i'
T ~'l
r~u;."- ., \X'I ::s
•• ~ .J
~
Conical shells ~ .. ~ ::s
p..
r--
System Loading N Ne T
s ~
~
2 1-<
'0 1 cos2~
l\ " p[ 51n41
~,
-P f 2s -PE S »il+ Cf)
~
o
'0 • 0 complete cone '""
U'l
P, • Pe
I cu,~ --1C~ ~
"2
m I' ___ ~_-P' ,1 ,1",,_________ -Pr ' . n. U'l
, 2 2
s· - s 3 n
~
Px • Ps 51n~ cos~ -Ps ---rs--o cotg~ -Ps s sl1i+
r- I I
I b
2'
@
Cf)
I • • 0 complete cone
ILjI o
cotg.
cos'.
PI - Ps cos 2• -Ps';' -Ps 5 srnr
2 2
II s s s - s
y 1: (h --2--~ cotg~ --3-° cos~) - ys(h cotg. + 5 cos4')
I ..
~~ C
2 P z • Y('SlM - h)
Points below water level
l. coo;t h 1 + o;2(1.,c:.l'>41. Jh cota.» y·(·co,. - h cot") 0
• bS( slnJo$ I -
-- -----
- Points aoove water level
)'h l cos4l_ 0
0
~ 5ln l •
Cl p '" y(h - 55111.)
z
P , t below water level
OIn S Y'(h cot" _ sco.t)
W r-<3h cota. - 2scos.) 0
f
~
2 2
-P~cotd -p 5 eot,t 0 ~
d P , • 0 Ö'
z • P 0 Complete cone >;
I
~ -p i cotd -p 5 COle. 0
w
CO
......
Co)
CO
N
Ci
(1)
)4 !!l,
Conical shells with free edges
(supported at the apex) ~
g
System NS Ne T P-
Loading
rn
12 _ 52
d PI • P P -y.- cotl. Ps eotl. 0
~
I
j,-
I
(3 _s 3 I
5Jn 2411 cos6 5 COS+
. Pt .. Pw sln. cos 0
1\
,Li-
3~
_~
2 28 COS.
-P" cos6
Pv7 !
unb I
--~--
- I
Equation of surface
Po -L.' , z • !L !!~
,
Boundary conditions
, • 0
'7;";" NJII " 0
System Loading N, N
Y
!.l~~T--'
Jt t l'"
-I' .tin~~~x
I 2 ~r
-P f 2 n ~
1. 2 .l2 ... n 2 x
• • n
Pr I 2 2 2
f
E..
P z " - PI "2 1.'" Y ·n
tt~ (,(l';~, cost
e:
x co~{I x cos'" ö'
"1
~
~-~----
--f---.-~-- 1 - - - - - - - - -1---------
[
"1
2
'~ .
Pz = - Ps eosy Ps 2C ~
~
/1)
-- ---- ---_._--------
0:
S'
CF>
~
rJl
w
CO
W
Co)
CQ
Hyperbolic Paraboloid shells ~
System ~
g).
I Loading I N, Ny T
y
OC--
y ...... xl I y 4 2
I ;;2 ''4- • z {S)''' • n 2 } • ~_ I~ • ~ (Sx 2 • n2 ) i!. (h - ~) x
~I Pn·Y(h-~)
,CJ .. tj.o cosljl
41111,,) I (,"0;'4- • 4nhxy) cos~ Je (A 2 • ,2 t n 2)
g).
(/l
o
....,
_ _ _ _ -<1 _ _ _ _ _ _ _ _ . _
-------
~
:=
w
s::
Pn .. p 1,,:t co ..", !!l. cost p~.n2
Jl 11 - cos~ ~ P n "Cöif
'~ 2n
i
r-- ----------1-----------.-- -------- -~---+--------j
~ll", ~- (x
___..1:_ x -Pw~()'-~X
2 '-2--2
Ir • n 2
Pn .. Pv z-1-r--z
II •
Y • Tl P" ~
..... n" tx --L-) cos~
§~ t~· --~-)~.,. rr---r cos~
fl-' cO!o'~ Ix· • n'"
•r • n
Tables for Analysis of Circular Cylindrical Shell Roofs 395
Appendix B
Notation:
Source of tables:
"Design of Cylindrical Shell Roofs", Manual 31, American Society of Civil Engineers,
New York, 1952
396 Design and Analysis of Shell Structu res
Hr
Hor:zootal Duplaeemen t .l.H-
... p.r r~4E{ x Col. (51 + [1 + ~ Ur)' + p. . ";'E [(1)' x CoL (Ul J
+1~__". (-'1')'] X Co!. (6)]
1 X ~in
" T XSillT
.'-0 I T. s IlH T. s
LoO&ILUdIOal Foree T..-- LoncJ'udlb.tJ Force Tr- LoDlltulhnal .'oru T.- Lob&ILUdlhal For~e T.--
~
11< [ I
Vz.[ (!)' XCoI. (1)]1'111 F/ U) X CoI. (6)] " .... 8L [(D' X CoJ. (In] IIßT &1111
rAfL[(')'
r x Co), (3) ] T
.. g.
Shea.rlDI Force S- SbearibS F orelt S- Shearml roret S- ShcariA, ForeII 8-
VL ['; X CoI. (2) ] .o. T
.. 11< [ ;' X Co,. (6) ] ... UT
ff
SL ; X Co! (JO) ...
[' ] "T ~[! XCoI.
• • ö'
CI.)].",,:..I ! ....
TraM,..I'1I' Force T ..- Ttausverae Forte T.- T,aNveme Force T f - TtAhaVerle Foree r.-
VL X Col. (3) X em T /lL X Col. (7) X .i. T SL X Col. (111 X .,. T A~L X Col. 0&) X ai,. T ~
Trau\·era. Momett& AI.- Tnnevcl'M l\lomeDL AI.- TranA\'eral!' Moment JI.- TrAmvel'l' Momen& AI.- ~
VL I. X CoI. (f)] 111
.i•.,. /iL Ir X Col. (811 ". T SI, Ir X Col. (12)) .,D T ~IL X Co!. (16) X .in
(ji'
T o......
SL
f 0.
....
'~
1 ~ f ;
f, SL~~ A '
~ I • fA ."
~ .'
E.
r. oS r, M. T. oS
e:
T, AI. T. oS r. ()
.". r. oS T, M.
• (I) (2) (3) (4) (5) (6) (7) (8) (P) (IO) (11) (12) (Il)
I I I I I I (14) (15) (10) ::::
I I I I I I :::s
(6) r/I-loo.n drll-O.\
§:
•• -30:
30 - 6.382 0 -3.47\ -0.3278 - 0.0013 0 @.
20 +0.9997 +0.\310 -0.0\104 0 -0.0207 -0.0010 - 0.0038
- 3.575 -2.619 -2.656 -0.2779 + 0.0017 -00001 +0.9848 +0.1162 -0.03118 -00403 0 + 00008 + 09738
10 + 1.809 -3.266 -0.776 -0.1612 + 0.0034 +0.0017 -00081 -00007 + 0.0172 + 00021 + 00022 + OP767
0 +1069 0 +0.500 0 - 0.015\ 0
+O.9399 +0.0722 +0.1118 -0.0250 +00142 +0.0002
+ 00264 + 00173 + 0.001< + 09SS4
+011660 0 +0.3600 +0.1000 0 0 - 0.1370 ~
0 0 + 1.000
~
-~
o
in'
Co)
CO
.....
w
CO
IX)
~-35
U - H~O 0 -29511 -03703 - 0.0001 0 +O.U!lRG +01751 -0_0774 o -00243 -0.0015 00310 0 - 00051 +0.90U
W - :I :t. .'O -0 U:!I -2820 -U:15!11 - I) (1)1:1 -0.0008 iOUO!'•• iO.1717 -ollnU·1 -OIl:!05 -00213 -011014 011211 - 00076 - 0.00"7 +0 1ltl50
~ - t 5:iH -2 :1U2 -I HI:! -u 2781 + U UUIiU -001:10 -Iumm:, ·to ,..25 -DO()S' -U01:111 -nexus -11 11008
-(J:UO
+ 01l37~ - 00013 + 00040 +0.9710 l!!.
W ... 1812 -:!.11Il -OII·H + 0 +0!"171 +00818 +0.1212 -00149 +00185 -00002 + uUlfm + Ofl:!SI + OOCH5 +0 ..,825
0 tU 571
"I." +0 11041 !II
o + 6815 U - 00316 0 +0.8192 o +0.3070 +O.IUOO o o - O.~~iI 0 0 +1.000
~.-~ + o
w - 2210 0 -2!,:1' -0401S - 001(';5 0 -, 0 OHM -1022-11 -001l75 o -00279 -00022 - 00712 0 - 00111 +09530 .....
w - 181).; -1 1:18 -2 IhR -0:1118:1 - 0 (M).IU -OlK)I'i8 +OI'K:lti -102011 -onuil -OU:t11 -001711 -OIMIIH - 001110 - 0 U293 - () n07 .. -10.9570
~ -O:",7K-l.H:tU -12m, -o~mn -ClIlU:':.! fJ)
+ 00l71i -'Ot"'211 -I n.II,72 -IOIIIHI -IJUI:!:J +0 IN)!', I -UUUO!I -I- IJUHH, - OUIIM + U01l2 -10 Htl.s.l
w + 1 5111 -14011 -IIUt!'. -0.1:1::'2 -I OOI!"li +0001:51 -t n H!iH!', -1011111 -10.12:10 -0IKI[17 -10 U221 -ouuo:! + OU72.. + OOl.lU8 -I- UUlOO +OU798 :r
o "".J. SJI 0 -fOtHl o - UUlJ:U 0 +07tiW o +0.2082 +O.IUOO o o 0.~9Y 0 0 +1.000
~-45 ~
~ - 1(,02 0 -2.231 -0 4~.s5 - OOJ!",U 0 +0.9873 +0277D -00.")96 o -00.115 -0.0031 - 0.1280 0 - 0 0~37 09392
W - 1 bll; -0":14 -2.10.1 -O"~7-I - 00:10\1 -00094 +O.USIU +0.2743 -0 (1S5U -OOlM -002U2 -000:10 - 0.1102 - 00325 - 0 038.~ +0.9~02 fJ)
W -IOUS-IIRU -161ß -O.:JIiM + 0 00:1:1 -0.11181 +"Uß1~ +02461 -00258 -003U8 -0.0123 -00022 + 0.0I2B - 00642 - 0 ..147 -004611
W - 0 llih -I :":-,,, -()7(i:J -02562 + 00-112 -(JuoftO +1I91:1S +018!1l t +0.0:138 -O.O:18!' tO 0120 -011010 + 11.152:1 - 11 11158 + 0 1I~7 +0.9595
W + 1:110 -1263 .. n,17!' -0.1255 -I- 002'hf) -fU.1I75 +08251 -10.1072 +0.1222 +0 on:!!', +O.u:!50 -OUOO2 + om'BI + 00680 + 011212 +09771 2
U + 3.407 0 tU 707 o - O.llHf, 0 +0.7071 o +0.2:178 +O.IUOO o o - 0 ·\013 0 0 +1.000 n
~-OO
~ - 1051 0 -1 1 131 -015-1' - 001)'):1 0 +0.9696 +0.3336 -005~1 o -00350 -00042 - 0.21148 0 - 001lllS +0.8218
c
W - 0 !'18 -0558 -1741 -0 42115 - o(r.!IJ -00:126 +0 unrJO +031!18 -00123 -00271 -o.n:!us -0011"7 - O.IIt12 - oour,.. - 00573 +09258
~ - Otil~' -omm -) 208 -0:1 .. 112 + o (J'..!HI-OO:U;«j +004:10 +0.278ß -(tOOOB -00121 -00057 -00025 + 0081!J - 0.1086 + 00111 +09371 m
~ - DUCH -1.181 -04!H -O:.!:I7!» I- u OS:"''O -0 OU3u +0"851 +021<11 +0.0440 -00.140 +0.0181 -UOUlD + 0 :m03 - 0 ()O~19 00.576 +0.9537
+
W I· Otl'IR -on:w . HJ.H7 -0.114' + 0 n:17-1 -f U 0:157 -I 0 77:1S +0.1107 +0.11118 +OIMI77 +0 n275 -00002 + 0.117'; + 0.1119 + 00373 +09745
U .. 2 -lag U -10700 o - O.~U"7 0 +U 0428 o -10.21.18 tu 1000 o o - 0 G6U.5 0 0 +1.000
...... 30
30 1-"1"'1 0 1-:lI~"I-o:ln281-0IlH21 n 1-111081"1.11112191-1110.161 0 1-004141-000051- OMDI 0 1-008131+011009
:•.'0 - 3·... 0 -2631 -21':12 -11~",1i5 - U1II2:1 -Olr.ma 11I!1757 -t U. 1081 -1)07[... -U0776 -00145 U - 0.073 - 0.2142 - 0014.1 -+OU124
10 - I 71j:1 -3 2:!tI -0787 -0 1:196 + 007nn -0 UU87 +U.\I·127 -I U 0078 tu 2208 -0 'HIiU +00281 +0 UUOO ... 0.557 - 0.0"29 + 00.\418 +0 GUß
o +10.\12 0 +o~uo 0 - 0.1140 0 -1·U.8CitiO 0 1-07119 -102000 0 0 - 1.067 0 0 +1.000
(ASCE Tab" 2A) (c"",,n/ml)
---------------
Vr:RnCAL Epo~ l.OAD HORIZONTAL 1: ...0'; LO.4oII SUUR t:oor. LOAD LOGE ;\IOInNT LOA.D
.. T. s T. .1(. T. s T. M. T. s T. M •• T. s T. .\{.
(I) (2) (3) (') (5) (6) (7) (8) (0) (10) (11) (12) (13) (14) (151 (161
I I I I I I I I I I I I >-l
(c) (Con1inucd) PJ
0-
•• -35:
35 -30870 -2.871 -03347 - 0.1[,52 0 +00r.03 +0.1591 -0.1467 o -00·1:'8 -000241- 0012 o 0 1014 +0 S637
- ff
30 - 2 93Q -0833 -2744 -0.32-18 - 0 12GO -003P9 +OO(j:!2 +015.,9 -0 1334 -0 ()~Q4 -00101 -Ooo:!2 - 07'012 - 0 2:\4~ - 0 I!Jf\') .... 08MP
20 - 1 6'39 -2 I.3G -1807 -0.2514 + 00;,111 -00n:'4 +OUf-!" +01301 -00106 -00815 -00018 -0 (K1l2 + 0327 - 03i47 + 0 O:U;. +08 Q I4
10 + I.S01 - 2.296 -0.365 -0.1302 + 0 152'; +0 0014 +0.0224 +0.0781 +0.2.178 -omso +0.030U -0.0001 + 0.928 + 00466 + CI 1~12 +0 P362 ~
o + 7.434 0 +0.574 o - 0.3244 0 +0.81112 o +0.6159 +0 2000 o o - ~ ()45 o 0 +1.000
•• -40. >
40 - 1.764 0 -2398 -0.3556 - 030f)G 0 +0.9087 +0 1!'62 -0.1273 o -OO.S25 -0.0035 - I 421 o - 0 3~S,) +08156 ::s
30 - 1.640 -0.047 -2.105 -03239 - 0.1336 -\...1:\89 +(UI3:i1 +0 1815 -00880 -0 Dfi:!(j -00.1:15 -00029 - 0 [011 - 0 r.-t(il - 0 ~(1()7 +0810G
20 - OOG,) -16H7 -1.251 -02374 + 0.2243 -0.1133 +O!lGSO +01487 +OO:i23 -00817 +0 01O~ -0.0(113 + 1015 - o 53!1I3 + 0 1.~Ji4 -'-08707 a
10 + 1111 -}7S3 -0109 -0.116S + 0 Zn1 +0 0557 +O~J101 +00870 +02108 -HOl 11 +0 0434 -00007 + I :UG + 0 252:i + U 2:!:,y +0 ~27I
o + 5.8J6 0 +0.613 o - 0.7210 0 +0.7&60 o +0.5460 +0.2000 o o 3.452 o 0 +1000 ~,
..... 45. Vl
45 - 0.775 0 -1.034 -0.3619 - O....,G70 0 +0.7041 +O.2!!fI3 -0 lOG7 o -0 0.~77 -00050 - 20i4
40 -0820-0217 -1.897 -03:;5(; - 0 4!'87 -0 1493 +08118 +O:!2(j~ -0.0988 -00287 -0.0535 -0 D048 - 1.82S - O.r,.f,(J +0 7S09
o - 00 (.:)s~ +07S!,8 ;"~'\I o
30 - lOfi2 -0731 -l.584 -0301>8 - 00210 -0.3078 +00229 +02071 -0 O~06 -00i2fi -00228 -0 D03l - OI1G - 1 138 - 0 HII) +0 j!l~J()
20 - 0 ['34 -1317 -0910 -02175 +1.008 +O.IGH +0 0570 -00737 +OOZ:.!4 -00014 + 1 11:,8 - 05!IS7 0 .~f':'f, +08510 '""
+ 0.[,:181 -0.1652
1-+
10 + O.7DO -1478 +0039 -0.1073 + 04215 +0.1611 +O.!lIM +0.09-16 +0 23:;3 +00028 +00482 o + I MI + 0 :;811 + 03778 _L{) HIH6 Q
o + 5.344 0 +0.707 o - 1.392 0 +0.7071 o +0.4995 +0.2000 o o - 52~9 o 0 +1.000 '"1
•• -50. n
SO + 0052 0 -1.437 -0.3493 - O.P469 0 +0.5813 +02!)14 -0 06~S o -oorm -ODOC>5 - 2814 o - 12~.~ l+or.. 13 ~
40 - 0 2!11 -0.028 -1.413 -0.3315 - 0.5827 -045M +0.7040 +0.2452 -00i14 -00H9 -0011)8 -000.')(; - 1.7'62 - 1381 - 0 8 n" +0 (.8119
30 - 1016 -0381 -I 246 -0.278G + 0 :.?88j-0 [,!,26 +O.!HiMj +02228 -OO!;.S -0 Qi.?3 -00113 -00035 + 07Vi - 16'1() + (I U5-,~ ..J..Q ;:,Z:1 iii'
20 - 1214 -10[,9 -0715 -0 1!l~3 + 0.1015 -0 W20 +1112 +Oli04 +oonnl -0 ()r.1~1 +O(nI2 -00013 + 3038 - 05';(; + 071<17 +0 8138 '"1
10 + 0381 -1127 +0110 -OOP43 + 0 56~·1 +03610 +O.9Iil +01015 +0 n.)7 +0 {)12~, +O.O.:jJ3 o + 1811 + 1 ('83 + 0 :.;18 ..,..0 fit!>"
o +.} 670 0 +0766 o - 23~18 0 +OGJ:!8 o +047018 +0 :!OOO o o - 7 473 o o +1000 n
(d) r/I & 100 .nJ r/I - 03
....
5'
0.
.~-30.
'"1
30 - 4.;09 0 -3.362 -0 27691- 0 2662 0 +0.Q.;71 +0.100~ -02018 0 -005fiij -00021 - 2101 0
I n'
20 - 3 S6S -2381 -2 G~2 -0.2347 - 0 DW-" -OO!U3J +0.n752 +00982 -01182 -0 lIa -OCl:.!31 -()IKtl(i - o lOS - 0 7(i(j~ : g~~t.~l:~ ~n~
10 + 10[.3 -3.2\)0 -08.;7 -O.l2iS + 0.2!142 +00035 +OSl6:lP +00622 +032:.1 -OO7!,0 +0 h411 -u nOO3 + 2372 + 00141 a
0 +l2A3 0 +0.500 0 +0 86GO 0 +1.102 +03000 0 0 5.456 0 + O'b''':'1 ~~ ~~5
o 1- 06606 -
gJ
~
~
o
;n-
w
CO
CO
~
o
o
Vt 111 ICAI. J.:llt-'t. LUAU Iljllll/nNIAf. EHt:~ I.HAII SIIJ:All EIH)f: J,OAU ErHU' Mo\u NT LOAD lo'Q"
::3
,. 'I'. .~ 1'. M. 7'. s 1'~ M,; 1', s 1'+ M. 7'. oS 1'~ M. D>
::3
(I) (2) (l) (4) (5) (0) (7) (S) (9) (10) (11 ) (12) (Il) (B) (15) ('6) 0..
--- I I I I I I I I I I I I
(11) (CUllIIIIllCd) >-
::3
~
:,:, - ~ I!H, Cl -:.!fiR:.! -02I M):I - o.m.,,,/ 0 1-0 R7{)fj HI.l:Ui2 -0.2110 o -onfiS')
-:.! ',!II -() OO~II /- ~A9~ 10.7081
:10 -:.! .!O.I -IJ :""~'-I -11 :!H.!I - ·IS:.!:I
U -0.1:,11 -tn M~'I:! -I-li ':I~K -U.IH:t7 -0 U5f1S -UUI,U:l -tJ tIlll:t - 2 8118 - ORIH5
o /-- (1';;1111
07'1\:1/ ;t1711i1 ~"
~o -:! 111,. -I 817 -I W,II -{I ~.!Ol -I- U.2WK -().~:Ion -HI HR:,~I 1111':'7 -nlJ~11 -O,I:l18 -()UO-lti -11 ()Uld + I r,flti - I :1!iR +0 IOllM HI772ß Ul
10 t- () .'",I;~ -:! 1:.li -11 IX:.! -1-UnUtlti +0 :I'lr,(i -011171 -0 n002 :1 7tll 0.-1372 +U.Sfi8')
"'' :' 1
o +IOU.' I
I 11 +0;,.1
-0 "'8 --I-
o -
n fi.!-I:.! ·t 0 U1a5
151:1 0
+n ~IK(}()
-1-(181\12 o -IO.tl(j71i +u 30UU
-+ U ur,;l!j
o o
+
- ! I tU:l
+ OA2.h) +
o 0 +'UIlO o....,
• • := 10
411 - 0 :.!Ih n -I !177 -028'):1 +0 ßn:, -1-0 1!i5~ -().l(jfiR o -0073ß -0 110'8/- 5.19~ o - 1.-IIiH +0 M71 (j)
:H1 - IO.!u -CI :.!7:) -I H1I2 -0.!1i.t1 -- 1.17ß
(J 181ir, / 0
-051·lrt +1) 82ng -10.'.'88 -O_12aM -nIlH:l7 -0018., -(I no:m - 2 :.!~:! - :.!.28R - 07t7fl +01il-l4
211 - :.!:.!1II -1.~O:1 -I·I.IU -!lI!IK2 + U!l111 -(l:N.",S 1-1 Uli!1 +tI.I:!:l2 -f (I O~h.t -IUItH -I-U ()1~5 -0.01118 + ;1 Oll.' - I.SI)!) + () ;,71" +U ];INJ ::r'
lfI - CI .!'G -:! 1·11 -11 ,I.U~ -1I,IIH8 ·t 1.07 J +0 2H7U +1 f11H I-U 076:! +U.;I:HI7 -(J n2:,~ +001111 -0.0001 + [, (1(11 + 1.1811 + 0 7!lfiG +0.8:,72
o , IU 0' 0 +U 013 o - :1 ()oH 0 +O.7tiOO o +0.8~:I' +():looo o o - 14.38 o o +I.IIOU ~
'1/1,=1;) (j)
,.; + 1 II)(} 11 -, '7R -O:.!.'",I1!', +0 :m:"i +n 1.')91 -011'12 o -1-0.07'9 -O.OO5Q 6.811 o - 2 -t(,3 +047-10
+ I u:-n 10 :1:,' -l.:!H -() :!!',:t7 -- 17:!H
'!tm 1-{I 0:")178 +O:t7tU. 1-11.1 [,8!1 -0.11 ,~ -011:110 -()1I67-1 -OOlI.",n - ti.();t! - t K02 - 2202 ; 0."871 q
:io - 1 1110 I (I I:~.! -I 1:11 -O:.!:!IlIi - HOI7:! -I onl 1-11 K:lgr, +0. I f,.",1) -00776 -(I (IW,I -()U:t.!K -0 olim - 0:110 - 371:' - (J 17711 -f U .,)768 ~
:.'0 - :1 110 -07S!i -I,:.!1i -u 171.1 I- 1.'111 -o ..n!"d +1.:!7\1 ;-U!:UI:J tO(l.Ht! -0.101:1 +IIU:!·II -0 lKJlU + ti.770 - I.Sli7 + 1312 +0712-1 n
- 0\111, -:!'!II -():UII -0 (K)OI
'..,"
-OuHml t I rll:I -Hlli:I:!8 +1 Ir,8 +01l8:.!1 f--U:':!C1'1 -U()128 +nOfil5 + !jliHIJ -{- 2.230 + 12:19 +08';10
o -1-11 :.! I U 107117 o - :l :!:,7 n l-tl1U71 1<J8H' +u:\ouu () o - I!II;I o o +1.000 8"
<I>_:...r.o " ....
~,II I 2 Cour, tI -{l.I:!li -02'100 - :! N11 () -fI ~1C17 IU 14( 1) -0 o:,nf~ () -(lWill o I't>
_nnnr..11- 7.871 Ul
-11 li71; -flur,I,1 -(I -
·IU I I :.!7h 11 180 -II:!II:I:! - I 711 -1.;IHi I () 1:,77 1() 1171 -HIUird -011:\17 uur,1i - "'1:11 :J 782
-- 3~!"31;-OJU3
2 :1\11 -tU :17:H
I .~ 17 IIOI.:.! -11I.! - (J I'I:!H 1 11 K7 I.! -lli:.!1 1101.1 Itllf,lil ··H tltil.1 -u U7un -11 U:!IJH -(l'M!.IH I- :!:.!07 - -I Ilhli -t- tI :!:.'Ol -f U .;'1:,7
.!tI
'" _ 111'1'1 t) hqH -I ,li.! -11 1;,:.!7 1- :ll.!'1 -u IUili +11i.!.1 -lU I-I:!:! ·IHU:.!I-I -UIIK7'1 I CI 1I:!11i -11 IM)!"I 1- H ll17 -- I·I:!'.! I· 2.171 ·1 n li~ltl5
I IS7 .! pl7 -u .17~1 -noH(JI 1- 17Hfl +1.1:':. -1-1:11:". I CI UHHr, +11 :~otlr, -{JUli:' I-UUh.I!1 I-(JIHKJ:! -I- [, r.1iS + :'.117 + 1.li:!!', +11 8.",08
'" tI t I:! loh tI I-H 71iO 11 - 7.77:1 o -I I) lil:!8 o -I-U 8!ltI:! I-u:moo o o - 2.176 o o +' 000
-------------
(e) ,It .,. 100 ,~IIII , If = U.·I
---;----,
.,.. -='IU
:w ["IM (I
-:1 17:, -CI 2101 ,'I-n.8RI -IOn!'''7 -U;);UK --uoo:!" 5:1:1
0 (I
I.!I:'-o - 087!!
0 W,j 0.r.,"8
+O.li!lur.
I (I.UH:"n 1.';lj8 -0 o1 !'Hl3 -0 ()772I
:!t1 -I_!II:!
-- :t:1 ['{lIl I -2.f,!I~ -11 :!tJIS
7:lli' -0.2(~2
-- ()U U:18 +1I.!I[,8 -n 1-0 u:m -O,IH)I!I - O. '" 1- I'
111 - 021-1 -3 :.!81 -11 UI7 -U.II:l1 '~-U()r,H +0.·11.'):1 -OOfl!I8 +0 0545 -UHUvJ "27 + 0.187 + O_5f):J -f 0.816\1
+ 0 80' -0.0008 +I 1108
1-+
o +1523 U -I-U SOO o - 1.77~ 0 +0.8I)(i o +1.508 -+0.4000 0 o - '5.27 o U +1.000
(Ase..: Tahle 2A) «((1111111111',1)
VF.RTlCAL J·~OOr. LOAD JJOlt170NTAL EI/UFo I.OAD f;lIt.4" E,uH: "(lAU 1':1'011: ~lm,tI':NT 1.0AO
• T. I s T.
I I M. T. I ,~ I T. I M. 1·. I ,~ 11"' I M. 1". I . I T. I AI.
(1) (2) (J) (4) (5) _ (6)_ _(7) _ (_8) __ _(9)_ (10) (11) (12) (lJ) CU) (15) (16)
..;
(e) (ConLlllucd) ~
0-
.... 3.5. n>
V>
35 - 0217 0 -2.201 -02:'1[,1 - lori 111 In.n7o ~ 0.10111 -() 2:;IH -00R12 -00010 - R IR 0 - I R<lO 10 511i"
30 - o,,;r.r. -0.102 -2.2·11 -022!13 - 1.2101 -0 :lfI,7R
° ~072:t ~ 011178 -u 2;11 Ii -0 Ufi72
° -(lu7In -I) on.t7 - fl No!' - 217:1 - 1:;74 f f) ,',:-""1
20 - 2."82 -1.0811 -I.!III -0.18-13 + O,fi8CI -0 f,777 +1.01~ ·I·n onr,2 -U"1R11 -0.1.".,1(' -0 um:t -()IHIHI ~ :1(.1, - a .IIHI ... 021" IOfHr.:I ö'
10 - 1512 -2.715 -0.712 -0.OU8l + 1.!i!Ui +011182 +1.0112 +0.11612 +UA:W8 -u(lIi71 +OOIi7:1 -(1(1001 + U :17 + I.IOG + 1.0('" tO 8nuu '"i
0 +15.11 0 +0.:-'14 0 - 3.900 0 +0.810 0 ·H 373 +0.·11100 0 0 - :lJ.fm 0 0 +1.000
•• -40 :>
40 + 2.485 0 -1.221 -02044 - 2.fiM 0 ~ 0.2'.0 +O.IOSR -0.1512 0 -00818 -OOOIß - II I!; 0 - 3.211 +oall<1fl ~
30 + 0 117 +O.PI' -1.!'d2 -O.!P·IS - 1.12' -1.171 ., O.rlClfi +O.1nU:1 -01:1Ir, -nOS:;1 -0 ()!',fi7 -o,xu!'! - .. 83 ~'2"
-_ ..3 flfJ' - 0 IM +04HI
20 - 4.2fHi -U.278 -1.110 -01580 + 2058 -OOI;]fl +1.2:I(j +0.11)1' +n IKl2:' -013:01 ., o ()(Jt}fJ -O.fXH7 + 87' + 12;.n l-O f.nOR e.
10 - 2814 -274f' -0.7·1!'! -00862 + 2AO;' +0.ro036 +1272 +00665 +0.'1228 -flOHI +0 (1728 0 + 11:10 I· 2733 + 1.7fil l-O 7'118
1
0 +17.20 0 +0.613 0 - 7.ln 0 '-O.iGG 0 +1.:Ui8 +011100 0 0 - JJ.3t1 0 0 +1.000 6,
.,-45: V>
45 + 4.329 0 -0.110 -0.1.")13 - :I 817 0 -0.:172 +0.08PP -00fi!',:l I) -00n;'7 -ooorol - 1250 0 - 4 rol7 1027ln
40 + 3fifi;t +1.126 -0.277 -0.1526 - 3.3rll -11KI.", -02Jn +001'23 -0072:1 -0 nu~r. -OOfi:lO -00019 - I1 (}{) - 3209 - .. 1',8 tO zu:!
- ()nl~ +2020 -1.2'19 -O,I.",IH - O.lr.o -207\) +0.70' ·1 n.lOfiO -0 w:m -u UG7:1 -(1(1;'81 -nno,t .... o !12 - (i!I;n - () P71 "'0 :\Rlfi
o......
30 -
20 - GO:\1 -O.OOS -1.88n -0 L177 + :I 7(tß - I UI;' +IJi:l" +n 108:1 -OO:'H -() 1112 -I n 01!',8 -«Iom:J + 12 f,r, - :t r.17 + 2111 IO.",1j1l
10 - 3,35.~ -3.104 -0_";'1 -00780 + J.104 +1.200 +1,4!17 +0.1l73 I +0:1".",1 -fI 01 t!! ·tOO7Hj .0OOU.l + 11 !'II + .. 262 + 2 Jli3 +07'l1ft n
0 +10.64 0 +0707 0 -10.78 0 +0.707 0 +1388 +o·t!x»n 0 0 - :t\I ..",U 0 0 +I OIMI ::;'
•• -50
r.o + .073 0 +0.78' -00881 - 4.-177 0 -10'1:1 +OO[,.1r. ·IUO'.Ur. 0 -OO:lI)!" -0 (XII.". - 11 " ..., 0 - [.4P1 10 or•• r.
40 + 2.829 +2 '28 +Ofl'!I -01028 - 2.8:18 -2,155 -O,JM I flllilli -I» 02711 .f Onn2.", -nllltlA -0 UIIII - 7 r.!; - .. 1M2 - :\ ':7:1 -1111,"',1'1 8
30 - 2,43:1 +2f1l6 -I.:m!l -0.127!) + 12:1.", -2.673 +1116.", -IO.IfI.",7 -lIllhf, -nO.IR2 -OIl.U:1 -H(M,:.!II .~ 261 - ;t 7".", + 021fi -I n 1.",'11 Pi"
-2.1111 -0.12<11 + [,12:; -O,R'lCl2 +21:"Jf) -111.1181 -0077 11 -111112.", -f n 111 ml -{lIMlJO +- I:1R'2 - 11f.7 + :1:1 11 1 +n ;-,flRI '"i
20 - fi881 -021J
10 - 'Ol~ -3.",tO -0.001 -0.0724 + 3.212 +10011 +1.707 +00810 +0.:1"28 -(10."... , -tu UHn7 +0000;' + 111111 ... [)1!',1t + 2.ror.8 -t 0 7\'11"
0 +203,1 0 +0.766 0 -13.50 0 ~0.6·13 0 +1.-1511 +0.1000 Il 0 - l' 80 0 0 fl.O\MI n
-- --
-'-- ~
"'o...."
.j:>
oI\)
Cl
(ASCE Table 2A) (continued) m.
VEB11CAL Epos LOAD HORIZONTAL EpQII: LOAD
::I
SREAR EDOE LOAD EDOE MOWENT LoAD
11>
T.
::I
• s T. 101. T. s T. M. T. s T. At. T. s T. M. 0-
(I) (2) (3) (f) (5) (6) (7) (8) (9) (10) (11) (12) (lJ) (lf) (15) (16)
I I I I I ;J>
I(f) (C..ti.ued)
I I I I I I ::I
e:.
,. .. -35: ~.
35 + 2.471 0 -1.677 -0.1788 - 2.776 0 +0.395 +0.0816 -0.2558 0 -0.0962 -0.0039 - 14.99 0 - 3.345 +0.3692 VI
30 + 1.493 +0.587 -1.758 -0.1758 - 2.22a -0.7101 +0.495 +0.0816 -0.2428 -00837 -00875 -00037 - 12.12 - 4.6G8 - 2.786 +0.3888
20 - 4.209 -0.060 -2.002 -0.1480 + 1.246 -1.062 +1.056 +00775 -0.0879 -0.1706 -00227 -00018 - 51107 + 0453 +0.5266 o
10 - 4.426 -3090 -1.055 + 6.40 ......
-0.0837 + 2.986 +0.3737 +1.252 +0.0535 +0.4733 -00921 +0.0629 +0.0000 + 17.12 + 1978 + !.D31 +0.7345
0 +22.54 0 +0.571 0 - 7.555 0 +0.819 0 +1.867 +0.5000 0 0 - 4f 56 0 0 (f)
... -40: +1.000
40 + 5.472 0 -0.358 -0.1324 - 4.282 0 -0.220 +0.0682 -0.1143 0 -00775 -00034 - 17.17 0 - 6.060 +0.2033
::r
30 + 1.404 +2.232 -1.069 -0.1354 - 1.839 -1.889 +0.363 +0.0754 -0.1408 -0.0683 -0.0585 -00034 7.86 - 7.678 - 2.645 +0.2912
20 - 6.661 +0.757 -2.062 -0.1254 + 3.336 -1.488 +1.415 +0.0819 -0.0853 -0.1408 +00027 -00014
-
+ 13.14 - 6.372 + 1.904 +O.SOOI ~
10 - 5.826 -3.503 -1.2U -0.0751 + U96 +1.017 +1.528 +0.0593 +0.4218 -0.0802 +0.0780 +0.0003 + 18.72 + U29 + 2.821 +0 7347
0 +26.12 0 +0.643 (f)
0 -12.17 0 +0.766 0 +1.908 +0.5000 0 0 - 502 0 0 +1.000
•• -45: ;:r
f5 + 6.655 0 +0.771 -0.0750 - 4.823 0 -0.923 +00f05 +0.0153 0 -0.0477 -0.0038 - 15.73 0 - 6.128 +0 05Z3
40 + 5.749 +1.741 +0.524 -0.0793 - 4.625 -1.375 -0.727 +00448 -0.0054 +0.0023 -0.0480 -0.0037 - 14.17
C
- 4.167 - 5.525 +0076 I")
30 - 0.652 +3.364 -1.053 -0.1039 - 0.365 -2.902 +0.573 +00709 -0.1266 -0.0318 -00417 -0.0028 - 2.11 - 9.122 - 1.418 +02487
20 - 8.443 +0.730 -2.362 -0.1125 + 5.266 -1.497 +1.929 +0.0901 -0.1395 -0.1180 +00012 -0.0011 + 1612 - 5.253 + 3.200 +0501 E"
10 - 5.758 -4.033 -1.326 -0.0708 + 4.705 +1.781 +1.798 +0.0670 +0.3807 -0.0882 +0.0726 +0.0003 + 17.30 + 5.fi58 + 3.294 +0.742 '"1
0 +27.76 0 +0.707 0 -16.12 0 +0.707 0 +2.002 +05000 0 0 - 58.57 0 0
... -50: +1.000 l}l
50 + 6.074 0 +1.(16 -0.0253 - 5.122 0 -1.462 +0.0088 +0.1028 0 -0.0139 -0.0027 - 11.38 0 - 6()!l4 -0052
40 + 3.738 +2.900 +0.642 -0.0469 - 3.478 -2.507 -0.713 +0.0302 +0.0147 +0.0398 -0.0269 -00027 - 8.59 - 5.742 - .. 363 +0.033
30 - 2.427 +3.394 -1.371 -00900 + 1.103 -3.275 +1.025 +0.0759 -0.1664 -0.0015 -0.0410 -00022 + 092 - 8.235 - 0093 +0.253
20 - 8.536 +0.212 -2.512 -0.1065 + 6.245 -1.164 +2.385 +0.1024 -0.1868 -0.1144 -00079 -00010 + 15.24 - 3.756 + 3776 +0.515
10 - 4.760 -4.221 -1.261 -0.0672 + 4.596 +2.387 +1.966 +0.0752 +0.3719 -0.1018 +0.0671 +0.0003 + If 96 + 6.079 + 3305 +0.749
0 +26.68 0 +0.766 0 -1851 0 +0.643 0 +2.065 +0.5000 0 0 - tJG.51 0 0 +1.000
---- -- ----
~.-30:
30 + 0.890 0 -2.477 -0.1648 - 2.37& 0 +0.609 +0.0838 -0.4130 0 -0.101& -0.0031 - 17.24 0 - 2.909 +0.3912
20 - 3.149 -0.438 -2.432 -0.1447 - 0.079 -0.8580 +0.895 +0.0605 -0.2349 -0.1448 -0.0424 -0.0010 0.99 - 6.327 - 0.707 +0.4746
10 - 5.088 -3.429 -1.334 -00837 + 2.720 +0.0560 +1.160 +0.0431 +0.5284 -0.1530 +00739 -0.0001 + 16.89 + 0.297 + 1.831 +O.88U
0 +26.76 0 +D.SOO 0 - 8.379 0 +0.866 0 +2.481 +0.6000 0 0 - 50.48 0 0 +1.000
(ASCE Tabl. 2A) (col/tln"rd)
----
Vr.anc.u EDOJ. LoAD lJOIll&Ol'lU.L EDol. LoAD SR LAR LOGt: LOAD El'olt MOUI:N'T LoAD
• T. s T. M. T. s T. M. T. s T. M. T. s T. J/.
(1) (2) (3) (') (5) (6) (7) (8) (9) (10) (11) (ll) (U) (14) (15) (16)
I I I I I I
(Q) (Cont.lDued)
I I I I I I ~
~
•• -35:
m
<n
35 + 5.530 0 +0.999 -0.1294 - 4187 0 +0085 +00577 -0.2245 0 -0 (li137 -0.0036 - ZU4 0- 4 _ +0.2286
30 + 3.964 +1.371 -1.194 -0.1291 - 3.3G8 -1.072 +0.231 +0.058" -0.22;8 -00617 -0.OSl9 -00034 -17M - [, 539 - 4 ()Q7 +O.2.",J4
20 - 6.521 +1.142 -2.096 -0.1187 + 1 8J.~ -1.(;25 +1.0fl8 +0(l{;28 -0.1434 -0.1738 -001(,3 -00014 + 878 - 87{)!1 + 0 r,(lg +04257 ö'
10 - 7.862 -3.492 -1.452 -0.0731 + 4 fil7 +05579 +1438 +0.0477 +0.4825 -01215 +00821 +00005 + 2570 + 2 ;78 + 2.10(,. +0.6789 '"1
0 +31.211 0 +0.574 0 -11.74 0 +0.891 0 +2.453 +O.GOOO 0 0 - 67.87 0 0 +1.000
•• -oW:
40 + 8.029 0 +0.412 -0.0781 - 5.611 0 +0.G32 +0.0378 -0.0378 0 -0 Of,53 -00034 - 20.fH 0 - 6413 +00741 ?,>
30 + 2.640 +3.389 -0.660 -0.0907 - 2.518 -2.498 +0.140 +0.0500 -0.1275 -00385 -00561 -00020 - 10 34 - 9.508 - 3.449 +0.1748 Il>
20 - 8.5gB +1.754 -2.347 -0.1016 + 4.328 -2.036 +1.563 +<lOG;3 -0.1799 -01354 -0.0078 -00012 + 16.02 - 831\.1 + 2.316 +04147
10 - 8.783 -4.118 -1.651 -0.0674 + 5859 +1.341 +1 421 +0.0538 +0.4136 -0 1I~2 +0.0787 +00004 + 2& 53 + 5071 + 3(086 +00842 <"
0 +34.71 0 +0.643 0 -17.02 0 +0.766 0 +2.558 +O.GOOO 0 0 - 7[. 81 0 0 +1000 <n
•• -45: Vi·
45 + 7.004 0 +1.346 -0.0293 - 5.855 0 -1.258 +0 0115 +0.1009 0 -002(',7 -0.1>026 - 15 66 0 - 6.(':;5 -0.03e.-l
40 + 6.935 +2.078 +1.051 -00352 - 5.259 -\.550 -1.038 +0.0166 +0.0669 +0.0245 -00302 -00025 - 1436 - 4.156 - G057 -0.0122 o
.....,
30 - 0.196 +4.195 -0.884 -0.0712 - 0.727 -3.373 +0.459 +00490 -0 14M +0.0082 -004:19 -00022 - 3 !16 - 9.685 - 181. +01f>13
20 - 9841 +1.352 -2.657 -0.0959 + 6025 -1.905 +2.094 +0.OH5 -0.2510 -0.1174 -00148 -00010 + 16.43 - «) 537 + 3.3"1 +0 421\.S
10 - 7.893 -4.563 -1682 -00659 + 6150 +2.089 +2.020 +0.0623 +0.3813 -0.1339 +00714 +00004 + 2253 + 6.087 + 3873 +0.69'7 n
~.
0 +34.56 0 +0.707 0 -20.64 0 +0.707 0 +2.659 +0 GOOO 0 0 - 74.68 0 0 +1.000
•• -50: n
50 + 6.029 0 +1.647 +0.0038 - 4.847 0 -1.571 -0.0114 +0.1678 0 +0.0073 -00015 826 0 - 56t6 -00950 ~
40 + 4.096 +2.955 +0.283 -0.0192 - 3.656 -2.446 -0.851 +0.0100 +0.0523 +00699 -00151 -00017 800 - 4483 - 4344 -00164
30 - 1.803 +3.754 -1.264 -0.0627 + 0.458 -3.469 +0.919 +0.0574 -0.2030 +0.0305 -00491 -00018
---I.P3 - 7.812 - 0 G~3 +0 1835 ~
20 - 9.145 +0.659 -00245 -00010 3 !",,9 +0 4425 '"1
-2.736 -00932 + 6.530 -1.539 +2.469 +00897 -0.2862 -O.1~25 + 1386 - 5044 +
10 - 6.445 -4.528 -1.501 -0.0624 + 6.010 +2.570 +2.141 +00696 +0.38P9 -0.1463 +00675 +00003 + 2005 + 5.P42 + 3 ('58 +06986
0 +31.83 0 +0.766 0 -22.64 0 +0.643 0 +2.688 +0.6000 0 0 - 7037 0 0 +1000 n
(Al ./1 - 200 .nd'/l - 0.1 S·
-
0-
'"1
~.-3O:
30 - 5.388 0 -3.482 -0.3278 - 0.0129 0 +0.9979 +0.1310 -0.0905 0 -00207 -0.0009 - 0.1005 0 - 00164 +0.9730 n'
20 - 3.600 -2.628 -2.1166 -0.2779 + 0.0014 -00043 +0.9845 +0.1162 -0.0307 -0.0403 -00080 -00006 + 0.0123 - 0.0332 - 00024 +0.9761 ~
10 + 1.787 -3.290 -0.781 -0.1511 + 0.0164 +0.0020 +0.9410 +0.0722 +0.1117 -0.0251 +0.0142 -0.0002 + 0.1340 + 00177 + 00099 +0.9852
0 +10.83 0 +0.500 0 - 0.0447 0 +0.8660 0 +0.3604 +0.1000 0 0 - 0.3807 0 0 +1000 Cf)
::T
~
§l
o
ur
-1:>0
o
w
404 Design and Analysis of Shell Structures
~~§~~ ~~~~§~
====-
+++--r
ecce=-
++++++
::i~~~ ~~~~ ~!::=~~
i~~~~=
~~~
ög3~= ~~3=g= ~g==
01:0: d;~;= QC;C::l ce::>=::> ce:
! I+...L- II.J-+ 111++ 11+++ 11+
ii::;
w:,..- ~g~ ?:6~=
,,~--=
:§:::!~ ~~
ce::: =---:-0 o::!;;::=i; C~-.--:= er-==
oe: oe: =:0:1 c::;: 0=
11+ 11+ ' I ,..:.. I I I'+' 1+
:;~~ «I"''':COC'''':!
~=~;=;
-..
1ji>!
o~!::~= 0--0-00 ceS:
c==
1 I.,.
ce::
11 I~
00
11
=:::;~
.: =0:;01ccc==
::::"8==~ ~!::=~~~
c:=:i:= c:====
cle!'I ...,
ci::-=
11+"-1 11++1 11+++1 II+,J..+I 11+1
rt;g:.1~;':
"'-"I"~-
·
"" ... :0::'\1-=
d::::
c 11' I I
'"·
oS .... :,,":«12)"-
t:=:~~
~~~~~ ==l2~:~
C'I .Ieot-===
~ =;OQO:-: ..~
C'I'!'I-CO C"I:'I_== C"::'I:=
1111.+ I I I t+ I I "T"'.,.. 111+
osig:== =i
>~ ., ~
0-:-10
ce-- :'IM
I' I I 11
"
;~~~~a :'":"- ..
a;",,:,\-,..:'I
;:;:;e~3
~r:;~=]§:
f":- ..-:t-:: -'Z:':-;ol
~-==
i~~~
:-,-c--.
CIOCO,...~=:~
C':M _ _ "'::
---o-~ ce===:"2 .,.:-:--j
111 I...L.+
.
I II+.J. I I I+...L- 1 I I 1++ 11++
~~o-==~e=-==~~=o===S=oo===
...
~~~~_ ... ...... ~ ~~_
.
~ ~~_
..
~~"'M~-
(ASeE TabJ. 2A) (rcmMure/)
VERTICAL EDOI: LoAD JlORII.ONTAL EDO& LOAD SMrA. EbGL LOAD EDO~ MOMENT LoAD
.,-30.
:10
:w 1.124\ -0 CItH~ \-2<79\-02211
- 3.4Ii:' -2481 -fI I!J.17 \- 2625\
- () 0
UIlH -0.9430 \+11..\1-13\+008118
+n OOIJ2 +0.0810 \-112817\ 11
-u 1.",8:1 -0.1:130 \-0.0710
-0.0302 \-0.002"\-
-0.0018 - 20.1.
0.87 0
\ - 7.313 \-
- 3.173
0.619 r-o.5913
+06610
IU - .-.S:!8 -34118 -1."u7 -0.1117 2U71 +OU700 +I.IIJI +Ofl5:,)3 +1I.:Ui:l:l -lI.lOt!1 IO(HU;j -O.tXXlJ + 23.57 + 0.435 + 1.999 +0 8155
U
r
+27 1.i8 0 +O.MXJ 0
+
- n 8~" 0 +0.8l1UO 0 + 1.03' -t-O."ooo 0 0 - 50.5,2 o 0 +1.000
(AseE Taille 2.0\) (nl/lll"urcf)
Yn.nCA.L EnGE LOAD HOJIIZOh"UL LDOL LoAD Snu.1l l:DOE LOAD EDOE MOr.U:NT LOAD
o"'"
.....
o"'"
Cl
a
(1)
~,
(ASCE Ti.lhlc lAI ((1111111111('(/)
----- ------ ::s
VtUriCAI. EDa~_ I~OAD HOIl11.0N'rAl. E'lOIl: LOAD SUEAII 1':Dol: 1.04D EpoE htO\U:HT l.OAD P>
::l
. T. ,5 1'. 7', s 1'. 1', .~ 1'. 7', s 0-
M. M. M. 1'. M.
(I) (2) (l) (4) (S) (6) (7) (8) (9) (10) (11) (12) (Il) (14) (lS) (16) :t>
I I I I I I I I I I I I ::s
(I) Conhnucd ~
[Jl
k... 35' ü;"
-I I:! g~ 0 -10520 -0.1106 - 7.700 -0 MS7 +11.0486 +00566 0 -00528 -00027 - 41.17 0 - H,181 +0.)851
+:I,HI:J -10.0:17 -U.1151 - 11.371 -1.952 -0.31103 +O.nfIiS -U1I817 -0.0181 -00r,05 -t)UU:l5 - :t:J7!J
• :15:w
+10." ° -10 flZ - 7.611 -to 2:122
20 - ~ w:t +1.0:!:! -2 ·125 -O.I:WU + :1:"17 -30:111 +1.25:1 t00670 -(1.18IH -0,0006 -OO:.!O~ -00015 + 11i.5;j -16 110 + 1 .. ~u +O .. fili8
o......
lO -HiOg -,14tH -2417 -0.O8~7 + 8.6"2 tl.030 +U'26 +0.0564 tO.27oo -0,\200 +00510 -00001 + !In 25 + 5:.!6 + 5B7 +O.7!)(J.I
0 ·1·52.~\ 0 +0514 0 -22.Hi 0 tO.8\92 0 -12.:IH '10.5000 0 -13~.8 0 0 tl 000 U'l
k=40 ° ::r
• 40 ·HJ9:J 0 tl.963 -00443 - 88U8 0 -1520 tO.0185 tO.10\)3 -0.0175 -0.0020 - 3296 -)0:\0 -00072
30 + 53111 t5888 +0.146 -0.11720 - • (~J7 -3897 -03:!59 +00:198 -0.0475 tO.0298
° -00204 -00010 - 17.70 -15.2\!
° - 5586 tO.1389 ~
20 -I~ 7!; +3U:J8 -;IUt:.! -O.112~) + O.ti2~ -3.303 +1.051 tO.on7 -0.2613 -0.0652 -0.0242 -01lO12 t 24.15 -14:t:i + 39US to 4658
10 -153\ -5.476 -2 n~.u -00866 t 9721 +2099 +2.:150 +0.0045 tO.21:<6 -0.1310 +0.(1133 -00000 + ".05 t 813 + tj,422 +07611 U'l
0 .. 53 f.3 0 +0.043 0 -28.17 0 +07600 0 +~.LUO +05OUO 0 0 -133.2 0 0 tl 000 q
~ ......'l
• 45 +10 ~:! +2135 +000-15 - 7.196 0 -1.959 -0.0106 +0.1830 +00132 -00011 - 1831 0 - 8981 -0.1089 ~
40 +- !I ;18:, +2.771 +2(U5 -0 OU!.2 - 01S7 -1.907
° -1.675 -0.0031 +0.1440 tOO100
° to (KIOO -00012 - 1778 - 498 - 8:m3 -0.0771 n
:10 -I- U 9:.7 +,581i8 -0.505 -OOtilli - 1.610 -4.438 +0.2l86 tOOl31 -0.110\) tOOß37 -00:10:1 -00014 - 8.110 -13.02 - 2.87U tO.1523 8"
20 -12tH +2 GHG -32\18 -0.1102 + 7723 -2.850 +25fJ,'j +0.0881 -03155 -0.0670 -OO~"'I -00011 t 20.73 -10 69 + 4.006 +0.4876 '"i
10 -I:! 5J -5053 -2."57 -001332 + UA98 +2.791 +2.M:,·1 +00750 tO.22 11 -0.1407 tO 0389 -00001 + 38:11 + 83:1 + 6.068 +07681 (1)
0 +-JSUÜ 0 +0.707 0 -30."'1 0 tO.707 I 0 1-2.539 tO.5IJOO 0 0 -12J I 0 0 tl 000 [Jl
,=50
• 50 i- 5.720 11 +2.151 +00279 - 4333 0 -1.931 -002n5 tO.17:1O 0 +0.020fl +00062 - 4.66 0 - 6.575 -0.1404
40 + !l1U5 -1<1 (170 +1.:.!1i5 -ooms - 44:11 -2438 -1.2:18 -00037 tO 0786 tOU777 tU 0051 -(lOIlO7 0.77 - 3.1;0 - 5029 -00518
:In - 0025 f I h5:! -101-1 -OOIl[;ß - 1.252 -4.161 +OO8IG +00509 -0.1578 +0 U5!J3 -O.O:UJ7 -00013 - fl06 - 1.fiS2
-- 10:11 -tO.1U15
20 -10 .n! +1 ~N7 -:111:1 -U.IOtI:) + 7 .... 2 -21i:m +281:' -I n.1U13 -0 2~WJ -(071)" -OU:U,I -u tHJII + 16.18 - tl.Ga + 4675 +05157
\11 -121).", -5 U:W -2 Hl:t -u USS2 -HI..",O +3.115 +:I.I~'''''I -10.0913 ·IO.:0:!141 -01221 +UO:lfili -()UUOO +- 41 7M t (\:l0 + 7.575 108127
0 1-12 lU 0 f 0 7,i{j 0 -:n.811 0 t O.(H~8 0 +~""Il f U.5OOU 0 0 -IW.I 0 0 tl.OOU
-----
•• -=30
3011-13111
:lU - :.!!I:I 0
IIIH rllnr.oI-012\51-7 0
..1941 -2.725
-2WIi --I) I:!:.!I - 0:117 0177
+U7r,o 1+11 Ol[,n r02118ß I
+00512 11
-fl:.!:lfI5 -0.12:13 1-007:121-01 .. 21 1-
-uu;ml -IHHIIU - 1 "5
r•.·IUI -!.!OO1. 8721 l-1o.Z589
- 1.8~H -tO.-lOI2
1- 11
10
u
-HI8H -'.1751 -~ti:!1 -OUijli7
·1 bl 811 0
I- 87~,1j .·0.171
t o.~)uu 0 -2U.81 0
·H tiM.!
+O.HtiU
+O.UII:I
o
+U:lIIH)
+:1003
-f).WI;i
10 OOUO
+fJU5J2
U
-UtHIOI
U
+ li'i'!1
-Hili 1.
I + 07\
0
r
+ 556\1
o
-I l),li!I:11
+1.000
(A.SeE Tabl. lA.) (cnnlmurtl)
\'Ear.CAL Eno. LoAD Ho.uonAL EDG. LoAD 8..... EDG. LOAD EDOI MO"."T 1.0••
~
• T. s
I I I T. M. T. s
I I I T. JJ. T. s
I I I T. Jf. T. s
I I I T. M. ~
(I) (l) (~'- _ _ (4_)_ _ (5) CO). (7) (a) (9) _~IO)_ _ Ol~ _(~ _~) (14) (15) (16) f1)
VI
Cm) CCoaliauod)
-g
• .11-35.
35 +17.48 0 +1.S81 -00576 - 9.794 0 -1.165 ~,0231 ~.0814 0 -00363 -00020 - 4868 0 -11,31 ~0433
30 +13.78 +4442 ~.942 -OORU - 8004 -2.520 -0.794 ~0270 ~0010 ~01l3 -0,0381 -0,0020 - 40.77 -12 82 - 9488 ~0887 ~
20 -1037 HOOO -2.570 -00989 + 4.010 -3973 +1313 ~,0531 -0.2759 -006&3 -0.0294 -00012 + 17,1\ -2122 + 1.522 ~3710
10 -22.33 -5,09Q -3136 -00827 +11.70 +1.270 +2.252 ~0523 ~1830 -0.1613 ~.0454 -0.0000 + r.1122 + 5.83 + 7.338 +0,7020 ~
0 +68.85 0 ~.574 0 -3032 0 +0810 0 +3.204 +0.6000 0 0 -130,0 0 0 +1.000 <
.,-40. VI
40 +14,46 0 +2.530 -00052 - 8Il00 0 -1.788 -0,0029 ~209a 0 ~.0030 -0.0012 - 29.22 0 -1053 -008.'10 lji'
30 + 6.70 H.504 ~510 -00:ID7 - 4840 -4.143 -0527 ~0213 -0,0157 ~.07lD -0.0224 -0.0013 - 2021 -14 50 - G.IOI +0,0578
20 -1324 H092 -3.187 -00948 + 8,008 -3889 +2.003 ~0630 -0.3733 -0,0458 -00384 -0,0010 + 10.37 -16.37 + 3.594 ~,3\1()9
10 -ID32 -5810 -3.114 -0.0812 +12.10 +2.231 +2.r,01 ~081' ~.1500 -01707 ~.0373 -00001 + ~09 + 7.56 + •. 285 ~713D a.
• 0 +6523 0 +0643 0 -35.15 0 +0.768 0 +3.297 ~.6000 0 0 -100.2 0 0 +1,000
•• -45: 0.
45 +800 0 +2397 +00217 - 5,729 0 -1.858 -0.0205 ~.2347 0 ~.0288 -0.0004 838 0 - 7.554 -0.1241 "1
40 + 8.15 +2.310 +2088 +0.0133 - 5,022 -1.562 -1.836 -0.0140 ~.1024 ~OOO5 ~.OIgg -00005 - 1038 - HO - 7.198 -0.0988 n
30 + 2.68 +5.666 -0311 -0.0418 - 2.980 -4.165 ~041 ~0293 -0.1065 ~,0944 -0.0306 -0001\ - 15.36 -10.1\ - 3688 ~.OO30 s::
20 -1168 +3500 -3237 -00948 + 6.764 -3,438 +2418 ~.0750 -0,4030 -00596 -0.0478 -00011 + 1242 -1303 + 3667 ~,4113
10 -1602 -5.641 -2,752 -0.0172 +12.12 +2.709 +2752 ~.0895 ~.11lO6 -0.1895 ~0361 -80001 + !>/l.54 + 6,80 + 8668 ~.7Ir,o iiJ'
0 +5853 0 +0.707 0 -3804 0 ~.707 0 +3.268 ~.6000 0 0 -158.3 0 0 +1.000 "1
•• -50:
50 +309 0 +1.709 ~.O284 - 1.816 0 -1.467 -0.0273 ~.I854 0 ~,0382 ~,0002 + 718 0 - 3.972 -0.1088
40 + 442 +I.ASI +1.170 ~0063 - 3,003 -1.374 -1.120 -00084 ~,IOfil ~0872 ~01\8 -0,0003 577 + 140 - 4 &34 -00572 1-2
30 + 2.57 H.287 -0.638 -0.0476 - 3650 -3.722 +0.325 ~0400 -0,1481 ~.0826 -0.0419 -00012 - 20,5\ -683 - 3347 +0,1208
20 - 9,53 +2775 -2920 -00008 + 6,372 -3.425 +2.581 ~,0862 -0.3819 -0.0757 -0.0497 -00012 + 778 -12 83 + 3041 ~.4149 S'
20 -14.34 -5.142 -2383 -00708 +13,11 +2.955 +2.877 ~on7 ~.2199 -0.1883 ~031J4 -00001 + 5100 + 585 + 6371 ~,7134 0..
0 +53.46 0 ~.766 0 -41.80 0 +0.643 0 +3.218 ~,6000 0 0 -156,6 0 0 +1000 "1
n'
~
;9
~
g
üI'
.1:10
o
CO
.1:>0
......
o
(ASCE Tahle 211) Symmelrieal Edge Load, on Simply Supported
Cylilldrical Shd .. ; J)j'I,llIcelllellt, or Edgc at <I' = 11 CJ
ro
~.
\'tIIlJeA'. EHO~. I.IUIl lIuu'loN'lAL Euu..: LOA.D 811 t An Euot; LOA LI Mm.ra:rfT ELIOE LOAD
I I I ::s
VClllC'ul l)h(llacclUcnt Av-~(I)owllwllrtl 1·....::.ltl\·C)- III
%
::s
", ;/,~ R }' ('"I (I) ~< I'>III;! I 'I
1I/';li'H X (,,,I. (-I) Xhili
..
T ~h ;A:R X ('ul. (7) X hlll T I Alt. y.;!;~ X ('01. (10) X IIIn T 0-
I
lIullzolIllll I.h::.II)UCI'IIICIII. Au -(Iu"nrcl l'u:"It.lvc)-- >
14 ... r I' .. z
::s
l'L r't
'I f: X Cu!. (:!) X ~III
.
Tr 111,,.1, ß' X Col. (f,) X sill T ,,>,. 1"1'; K >< Cul. (8) X :.iu T ML X r4lE XC'ul. (11) XaiDT a
HatnluJII (J- H.,tntiun 11-· HulntiulI'- Rotation l - ~.
VI
I".~; X l'ul. CI) X Hili T r'
1/ J. H 1 X Col. (Ii) X Kill
.. 7
T SI. -k-'i X Col. (1) X hlll T ML X Fi X Cul. (12) X ein T o......
~,. dll A,· All Ay All , AU
• • A" • ~
~.
(I) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (U)
--
J_ I I I I I I I ~
(f)
(a) r/' - 100 q
~
r/I - 0.1; n
30 1~ r,:t 01115 01108 0,1445 -0.04556 o 31UI 0.000480 0.000363 - 1.330 OM6~ -0.5144
:15 11 :!li:1 U 2,)118 0.14:!1I 02[,118 0.12:11 -0.07108 0.2178 O.tMIlOOß ., UOUM4 1.7tll 08531 -0.500 .. 2"
.0 a (BI o "'Ot~, o l7r,:! --- () 1(K~' U 2:125-
--- 00585 -O,1U3U 0.144:1 o ()(12000 0.tMX1822 2102
--- 1.217 -0.6749 ....
45 2 :,nl o S·IH7 (12088 O.5H81 - 0·1010 -0.1444 0.1084 o Otl377 1 0.00122:1 2.!',O5 1.734 -0.7542
--- - ~
50 :.! 04:' 081:11 U.:NIO U 8134 -U.1925 IJ.07U2 0.006516 0.001776 - 2.892 2.312 -0.82110
-- - 0.650,"'8 -
r/l ... 0 2.
:10 17 :18 2 n:II () IJ'IR2 - 2 Oll - CI 8742 -00.1288 o Ciü[,O 0.00726 o ()()O353 - 18.86 8233 -O.ten3
a.U~ 1J.12:)tI :U,:18 1 -0.00[,:10 047:13 0.02065 - 2-1.18
It 87 :, 1.7 '1 o OtlOtl7H
-- 12M. -0.550"
:'.s
551 .. 0.1521
- 5.M1 -0,(",256 0.:1715 0,045\10 0.00114.5 - 29.25 - 17.77 -0.6159
40 12:H - -- 3.282
45 13 ·17 7.\lf,1l 017-18 7 USO 5,417 -O.12:JJ 0.3402 008862 0.001757 - 33.5(1 - 23.67 -0.0(151
-0.1646 0.1520 - 29.70
-- 105U
- 8.270 0.3610 0.002478 - 36.46 -0.6986
50 1".54 1050 0.IH"9 -
./1 ~ 0.3:
30 :1524. 11.4'" 00891 0.474 4010 -0.03866 1.187 0.00357 0.000536 - Bß.nt - 37.58 -0.4400
I!l.fi,", 7.8UO -O.Ur,OOO O.!Ifi83 0.1474 0.000023 -100.2 -0.4965
:If) ;17 :11 0.10!.2
-- 15.05 - M.MI
-0,075U5 O.U501 0.21131 0.tH1I413 -121.0 - 73.83 -0.6288
40 42 :18 22 UII 0.1215 - 22.UIJ
--- 1:1.59
45 4tillli aO.12 0.1317 - 3U.12 - 20.09 -0.09381 1.073 0.0020 0.001027 -127 .• - 01.21 -0.6-1:14
50 45.75 35.11 0.1285 - 35.11 - 27 .•3 -0.1070 1.209 0.7422 0.002334 -124.9 -10-1.0 -0.5402
rll - 0.4:
30 7506 20.10 0.07700 - 26.ltf - 11.06 -0.03483 I.fl40 0.2256 0000632 -239.2 -104.0 -0.4049
35 87.57 40.75 000046 - 40.75 - 20.25 -0.04716 I.U24 0.5080 0.000094 -277.7 -/45.2 -0.4320
40 96118 M.OO 0.0951,16 - 55.00 - :12.68 -005921 2.1811 0.!I067 0.001384 -293.8 -181.4 -0.4417
45 Oti.IO 6".69 O.()U256 - 0 .. UJ - 44.05 -O.OU7-IO 2.500 1.:172 0.001641 -284.1 -207.1 -0.4348
8,).1:', 07.32 008360 - 67.31 - .51.7" -007216 2.591 1.731 U.001732 -~~m5 7 -220.8 -0.4228
50
(ASCE Tahl. 28) (""'/li""'!)
- 0 S.
'1' 30 J4:!2 S4.14 006634 - 54.14 - 22.P8 -0.021>00 3218 o 5~9S o {l(l()r.4.& -497.6 -21G 6 -031102
:t.; 11;:\ 2 78.77 0.07239 - 7877 - 3'1 U7 -003811 3.588 1.202 o OtIO~J15 -5-12 ~ -285,9 -03700
4(1 167.7 97.43 0.07124 - gj 43 - 5847 -OOHM 4141 1938 0001160 -53~, 1 -33(, :; -O:tftfi8
.;, 1524 104.9 O.O(,w:, -1049 - 74.31 -00488.1 4418 2566 o ()(J1242 -48i .~, -36fi.2 -03568
50 127 :; 103.~ 005747 -103.9 - 87.21 -0.05166 .( 246 2.945 0.001109 -431.0 -387.5 -03493
,/I-OG.
30 2385 88 .7 o M,.",59 - 8847 - 3fl78 -002434 ".112 1.~18 O()O()('.:!O -8(,1 r, -3786 -03247
35 257.7 126.7 00:,714 -1267 - j'i4 74 -003049 5.97G 2.242 00(10834 -8S87 -4742 -(I :118:,
40 NS.4 14B o ar,3R6 -J.4·tJ. - 8810 -00:l39(j (j 62Q 32W O~13Z -8146 -b~S '2 -01171
H 2123 149.5 00478.1 -149 :; -1083 -003700 6.roOl 3.942 00(lOl'18 -7-13,9 -S7!. .( -0 ~Ot2
00 1758 1437 0.04304 -141.9 -127.9 -003\'67 6678 4324 00008r.5 -Gö9 .. -6187 -0 :W8G >-l
~
(bI 'It - 200 ro
U'I
-'=-
.....
.....
Index 413
Subject Index
Bearn-Arch method 74
Buclding of shells
concrete domes 184
concepts 344
circular cylindrical shells 352-366
formulas 360-366, 368, 371
gerneral stabilty considerations 22
hyperbolic paraboloids 371
shells of revolution 368
Concrete shells
cylindrical roof 63-101
design exarnples 89-99, 186,239
domes 169-194
hyperbolic paraboloids 215-247
folded plates 265
liquid retaining 291-341
Conical shells 137
Containment shells 3D, 31
Cylindrical shells
axisymmetric loading 53
axisymmetric probe1ms 55
bending analyis 43-61
buckling 353
concrete, see concrete shells
deformation 31, 34
design exarnples 89-99
displacement equations 50
internal pressure 57
membrane equations 26
membrane behavior 25-42,
membrane displacement 31
liquid retaining, 58
qualitatice discussion 36
qualitative behavior 34
roofs, 65-103, also see concrete shells
vaults 28, 33
414 Index
Domes
bending analysis 141-167
design example 186
dome-ring analysis 171
dome-ring roof 186
membrane analysis 103-139
qualitative behavior 130
reinforced concrete 169-194
spherical 111
with skylight 112
Edgebeams
dimensions 67
HP shells 223
vaults 29
Folded plates 249-289
Force method
general 20
cylindrical shells, see cylindrical shells
domes 160,171,183
liquid containers 304
Hyperbolic paraboloids
design 215-247
geometrical description 217
membrane analysis 222
Influence coefficients
cylindrical shells 54
shells of revolution 158
Liquid containers 291-341
Membrane behavior
cylindrical shells 25-42
general 15
general shells 197-205
HP shells 222
shells of revolution 103-139
Pipes
external pressure 366
internal pressure 31,57
Pressure vessels 120
Reinforcement
concrete Cylindrical Roofs 70
domes 186
Hp shells 244
liquid retaining shells 329
Index 415
Roofs
buekling 184,367,368
design examples 89-99, 169-192
geometrie Design of Cylindrieal Shell Roofs 66
analysis of Cylindrical Vaults by ASCE Tables 89
Shells
classification 4, 8
codes of practice 23
cylindrical, see cylindrical shells
general structural features 8
geometry 3
general desi gn considerations 21
hyperbolic paraboloid, see hyperbolic parabolois
Internal Force System in a Shell 12
liquid containers 291-341
liquid retaining 58
qualitative Description ofbehavior 15,34,130,227
revolution 103-139
with arbitrary geometry 195-214
shallow 212
Stability, see buckling of shells 22
theories 19
thin 12
uses 2
Shells of revolution
bending analysis 141-167
buckling 368
membrane analysis 105-140
1. M.A Krasnoselskii, P.P. Zabreiko, E.I. Pustylnik and P.E. Sbolevskii: Integral
Operators in Spaces of Summable Functions. 1976 ISBN 90-286-0294-1
2. V.V. Ivanov: The Theory of Approximate Methods and Their Application to the
Numerical Solution of Singular Integral Equations. 1976 ISBN 90-286-0036-1
3. A Kufner, O. John and S. Pucik: Function Spaces. 1977 ISBN 90-286-0015-9
4. S.G. Mikhlin: Approximation on a Rectangular Grid. With Application to Finite
Element Methods and Other Problems. 1979 ISBN 90-286-0008-6
5. D.G.B. Edelen: Isovector Methods for Equations of Balance. With Programs for
Computer Assistance in Operator Calculations and an Exposition of Practical Topics of
the Exterior Calculus. 1980 ISBN 90-286-0420-0
6. R.S. Anderssen, F.R. de Hoog and M.A Lukas (eds.): The Application and Numerical
Solution of Integral Equations. 1980 ISBN 90-286-0450-2
7. R.z. Has'minskil: Stochastic Stability of Differential Equations. 1980
ISBN 90-286-0100-7
8. AI. Vol'pert and S.I. Hudjaev: Analysis in Classes of Discontinuous Functions and
Equations of Mathematical Physics. 1985 ISBN 90-247-3109-7
9. A. Georgescu: Hydrodynamic Stability Theory. 1985 ISBN 90-247-3120-8
10. W. Noll: Finite-dimensional Spaces. Algebra, Geometry and Analysis. Volume 1.1987
ISBN Hb 90-247-3581-5; Pb 90-247-3582-3
MECHANICS OF CONTINUA
Editors: W.O. Williams and G.lE. Oravas
1. G.C. Sih (ed.): Methods 0/ Analysis and Solutions 0/ Crack Problems. 1973
ISBN 90-01-79860-8
2. M.K. Kassir and G.C. Sih (eds.): Three-dimensional Crack Problems. A New Solution
of Crack Solutions in Three-dimensional Elasticity. 1975 ISBN 90-286-0414-6
3. G.C. Sih (ed.): Plates and Shells with Cracks. 1977 ISBN 90-286-0146-5
4. G.c. Sih (ed.): Elastodynamic Crack Problems. 1977 ISBN 90-286-0156-2
5. G.C. Sih (ed.): Stress Analysis 0/ Notch Problems. Stress Solutions to a Variety of
Notch Geometries used in Engineering Design. 1978 ISBN 90-286-0166-X
6. G.C. Sih and E.P. Chen (eds.): Cracks in Composite Materials. A Compilation of Stress
Solutions for Composite System with Cracks. 1981 ISBN 90-247-2559-3
7. G.C. Sih (ed.): Experimental Evaluation 0/ Stress Concentration and Intensity Factors.
Useful Methods and Solutions to Experimentalists in Fracture Mechanics. 1981
ISBN 90-247-2558-5