Introduction to elastic

stability
By Girma K. (MSc)
introduction

 Failures of many engineering structures fall into one of two simple categories:
1. material failure
2. structural instability
The first type of failure, treated in introductory courses on the strength of
materials and structural mechanics, can usually be adequately predicted by
analysing the structure on the basis of equilibrium conditions or equations of
motion that are written for the initial, undeformed configuration of the structure.
By contrast, the prediction of failures due to structural instability requires
equations of equilibrium or motion to be formulated on the basis of the deformed
configuration of the structure. Since the deformed configuration is not known in
advance but depends on the deflections to be solved, the problem is in principle
nonlinear, although frequently it can be linearized in order to facilitate analysis.
Cont’d

 The study of structural stability is often confusing because the definition of

structural stability itself is unstable.
 Stability can be roughly defined as the ability of a structure to support a
given load without experiencing a sudden change in its configuration.
 Consider the following figure.
Cont’d

The figure illustrate various forms of equilibrium. We can easily see that if it
is displaced slightly, the ball on the concave spherical surface,(a), will return
to its original position upon the removal of the disturbance. On the other
hand, the ball on the convex spherical surface, (b) will continue to move
farther away from the original position if displaced slightly. A body that
behaves in the former way is said to be in a state of stable equilibrium, while
the latter is called unstable equilibrium. The ball on the horizontal plane, (c),
shows yet another behaviour: it remains at the position to which the small
disturbance has taken it. This is called a state of neutral equilibrium .
In this chapter, stability will be discussed chiefly in relation to columns,
more specifically in relation to buckling.
Buckling of columns

Buckling Vs Stability Buckling, however, does not occur as a

Change in geometry of structure under result of the applied stress reaching a
compression that results in its inability to certain predictable strength of the
resist loads is called buckling. Under an axial material. Instead, the stress at which
compressive load, a column that is buckling occurs depends on a variety of
sufficiently slender will fail due to deflection
factors including the dimensions of the
to the side rather than crushing of the
material. This phenomenon, called buckling, member, the way in which the member
is the simplest prototype of structural stability is supported, and the properties of the
problems, and it is also the stability problem material out of which the member is
that was historically the first to be solved. made.
A slender column is defined as a column that The goal in this topic determining the
has a significant reduction in its axial-load critical load at which the structure loses
capacity due to moments resulting from its neutrally stable equilibrium.
lateral deflections of the column.
Cont’d
Critical load of the Euler column
Even as simple a structural element as an
axially loaded member behaves in a fairly
complex manner. It is therefore desirable to
begin the study of columns with a very
idealized case, the Euler column. (Consider
the column in the figure below)
In this analysis, the following basic cases are
considered.
1. The ends of the member are simply
supported.
2. The member is perfectly straight, and the
load is applied along its
centroidal axis.

Buckled column in a laboratory

3. The material obeys Hooke's law.
Cont’d
We first recall from elementary
calculus that the curvature of a plane A third equivalent equation can be
curve at a point Q (x,y) of the curve can added here which gives us,
be expressed as

This equation represents the equation

of curvature. where dy/dx and are the
first and second derivatives of the
function y(x) represented by that curve.
But, in the case of the elastic curve of a
beam, the slope dy/dx is very small,
and its square is negligible compared to
unity. We write, therefore,
Cont’d
Setting, the value of the moment of
This equivalence gives rise to a second inertia I to be, where A is the cross
order differential equation of, sectional area and r is the radius of
gyration,
EI.y’’- P.y=0
Solving for this differential equation
and calibrating, we obtain the critical Or
load P that causes buckling, with a
value of The quantity L/r is called the
slenderness ratio of the column. It is
clear, that the minimum value of the
The value of the stress corresponding radius of gyration r should be used in
to the critical load is called the critical computing the slenderness ratio and the
stress and is denoted by critical stress in a column.
Cont’d
The above derivation represents the critical
buckling load for Euler’s column which is
the simplest form of stability problem with
its boundary condition limited to pin
supports only.
When the boundary conditions are altered,
it is obvious that the inflection points also
shift from their previous location on the
simply supported cases.
The figure below represents a few selected
scenarios with different boundary
conditions. In these cases, it can be
observed that due to the shifting of the
inflection points, the effective length
considered in the critical buckling
calculation also changes.
Buckling of frames

In the previous case, the stability of columns was discussed in an individual

manner in a very idealized boundary condition. Although this facilitates the
understanding of column buckling, it does not represent the actual case of
most of the columns in actual structures.
Instead, these members are usually part of a larger framework, and their
ends are elastically restrained by the members to which they are attached. it
is thus desirable that we consider the
behaviour of framed columns as well as the characteristics of isolated
members.
In large structural frames, no single compression member can buckle alone
without deforming the other members in the frame. Hence, to obtain the
critical load of one or more compression members that are part of a larger
framework, it is thus necessary to investigate the stability of the entire
frame acting as a single unit.
Modes of buckling

Let us consider the single-story single-

bay frame in the figure below. The
external loads P are assumed to act
directly over the columns so that there
is no bending moment in any member
of the frame prior to buckling.
When analysing frames, it is useful to
divide them into two categories: those
in which side sway is prevented (parts
a and b of the figure) and those in
which the upper joints are free to move
horizontally (parts c and d).
Cont’d
We consider first the frame in which If we assume the beam to be infinitely rigid,
side sway is prevented. At the critical the columns can neither rotate nor translate
at their upper ends. Under these
load this frame buckles as indicated by
circumstances the columns behave as if they
the solid line in the figure (a). It is self were fixed at both extremities, and the
evident in this simple structure that critical load of the frame is equal to four
buckling takes place when the applied times the Euler load of the columns.
load is equal to the critical load of the Alternatively, the beam can be assumed to
columns. It is also obvious that the be infinitely flexible. The beam is then
upper end of each column is elastically unable to offer any rotational restraint to the
restrained by the beam to which the upper end of the column (part b). In this
case the columns behave as if they were
column is rigidly connected, and that
fixed at one end and hinged at the other, and
the critical load of the column the critical load of the frame is
therefore depends not only on the approximately equal to twice the Euler load
column stiffness, but also on the of the columns.
stiffness of the beam.
Cont’d
For an actual frame the flexibility of the beam On the other hand, if the beam is infinitely
must lie somewhere between the two extreme flexible, the upper ends of the columns are
conditions just considered. The critical load of free to both rotate and translate, as indicated
such a frame, in which side sway is in part d. In this case the columns act as if
prevented, can therefore be bracketed as they were fixed at the base and free at the
follows: top, and the critical load of the frame is equal
to one fourth the Euler load of the columns.
The critical load of the frame whose upper
where is the critical value of the applied load
joints are free to translate laterally must
and is the Euler load of the columns.
therefore lie between p. and That is,
For frames whose upper joints are
free to move laterally, If the beam is infinitely
rigid, the frame buckles in the manner Comparison of the results for symmetric
indicated in part C. The upper ends of the buckling with those given for side sway
columns are free to translate, but they cannot buckling indicates that the load required to
rotate. Hence the critical load of the frame is cause symmetric buckling is larger than the
equal to the one needed for side sway regardless of the
Euler load of the columns. stiffness of the members.
Critical load of a frame using neutral equilibrium

Side sway Buckling The frame in Fig. a is fixed at the base

and free to translate laterally at the
top. To differentiate the column
properties from the beam properties,
the subscript l is used to denote the
former and the subscript 2 to denote
the latter.
When the frame buckles it assumes
the shape indicated by the solid lines
in Fig. a. The forces acting on each of
the individual members when the
frame is thus deformed are shown in
Fig. b.
Cont’d

the equation of moment equilibrium Symmetric Buckling

for the vertical member is,

solving this second order differential

equation, we obtain the critical load to
be,

This is the critical load for a fixed-base

portal frame free to move laterally at
the top.
Cont’d

If the frame is prevented from This is the critical load far a fixed-base
translating lateral1y at the top, portal frame whose beam has the same
buckling will occur in the symmetric stiffness as the columns and that is laterally
mode, as indicated in Fig. a of the restrained.
above figure. Effect of primary bending and plasticity
on frame behavior
The forces acting on individual
members. when this mode of buckling In the preceding articles the stability of
occurs are shown in Fig. b. frames was investigated assuming that
neither primary bending nor inelastic
Moment equilibrium for the vertical behavior existed in the frame prior to
member (Fig. c) requires that, buckling. Since one or both of these
conditions are often not fulfilled in an
similarly, solving this equation we actual frame, we want to briefly consider
how the presence of primary bending
obtain the critical load to be,
and plasticity affects the behavior of a
frame.
Effect of Primary Bending on Elastic Buckling Load

If a frame is loaded as shown in Fig. a,

no bending is present in any of its
members prior to buckling, and the
frame remains undeformed until The
critical load is reached (curve l, Fig. c).
By comparison, if a frame is loaded as
indicated in Fig. b, primary bending is
present in each member from the onset
of loading, and the frame deforms in
accordance with curve 2 in Fig. c.
Several investigations have indicated
that primary bending does not
significantly lower the critical load of a
frame as long as stresses remain elastic.
Cont’d
The only exception to these findings When carrying out the design, the member
occurs when the beam is exceptionally must be treated as a beam column, and both
long in comparison with the columns. In axial compression and bending must be
that instance, the presence of primary considered.
bending reduces the symmetric buckling Inelastic Buckling Load
load of the frame. The maximum load that a frame can
Therefore it appears safe to conclude that support is equal to the elastically
the effect of primary bending can be determined critical load only if the stress
neglected in determining the elastic does not exceed the proportional limit at
buckling load of a frame. any point in the frame prior to buckling.
In all this, it is important to underline the If the proportional limit is exceeded
fact that primary bending is negligible before instability occurs, the frame will
only in the determination of the critical fail at a load that is smaller than the
loading of framed members and not in elastic critical load (curve 3 in Fig. c).
their design.
Cont’d
The ultimate load of a frame can never the following empirical relation, known as
exceed either the elastically the Rankine equation, can be used to
determined critical load or the plastic predict this failure load:
mechanism load. If instability were the
only factor leading to collapse, failure
would occur at the critical load. On the
where failure load
other hand, if collapse were solely due to
plasticity effects, the frame would fail
when it became a mechanism as a result
of the formation of plastic hinges.
Actually, however, both instability and It has been shown that the results
plasticity are present, and collapse obtained using this equation are usually
occurs due to an interaction of the two at conservative and reasonably accurate.
a load that is lower than either the
critical load or the plastic mechanism
load.
The end