Buckling Loads of Columns With Gradually Changing Cross-Section Subjected To Combined Axial Loading
Buckling Loads of Columns With Gradually Changing Cross-Section Subjected To Combined Axial Loading
ABSTRACT
In this paper, the elastic stability of a column with variable cross-section subjected to combined
distributed and concentrated axial load is presented. The analysis is carried out by using the
minimum potential energy technique. The moment of inertia and the load intensity at different
cross-sections are varied according to power of the distance along the column length. The
buckling loads for perfect columns are given in explicit expressions for a wide range of the
power of the function of the moment of inertia and the power of the load intensity function. Also,
a simple, yet accurate expression, for the critical buckling coefficient is obtained instead of using
the explicit expression.
A comparison between the results of the present analysis and that of some previous studies shows
the accuracy and the simplicity of the developed method to find the critical load in such cases.
INTRODUCTION
The case of a fixed-free column is important for many practical applications as the column of the
lift bridges and the column which carries the crane girder. These columns are usually long
columns so that the buckling load is the most effective aspect in the design. Thus, the critical
buckling load under combined action of the own weight (distributed axial load) and the
concentrated load applied at the top should be calculated accuratly by a simple expression.
Also, in order to decrease the weight of these columns, gradually changing cross-section is
sometimes a good solution. As a result, the intensity of axial load due to own weight of the
column and the moment of inertia of the cross section are also gradually changed. Both of the
1
moment of inertia of different cross sections and the intensity of the load varry according to a
power function of the distance along the column length.
There are many researches concerning the compression members with gradually changing cross-
section [1,2,3]. In this paper, the study was performed to determine the critical load of the fixed-
free column with a gradually
n
changing cross section (the moment of inertia at any section is
given by I x l x
. IThe
0
column is subjected to both distributed and concentrated axial load.
l energy is used to establish the necessary equation taking into
The minimum potential
consideration two terms of the Fourier Series for the deflected shape of the column.
p = 1
p = 2
n = 0
n = 1
n = 2
qx qx qx Ix Ix Ix
l
q0 q0 q0 I0 I0 I0
2
Deflection Function
The buckling shape must be firstly assumed in order to apply the minimum potential energy
method. The best approximation can be obtained by taking the deflection y as a function which
satisfies the geometric conditions. Two terms of the deflection curve are assumed as follows:
3
y a1 (1 cos x) a 2 (1 cos x) (3)
2l 2l
Method of Analysis and Theoretical Equations
By applying the energy method using the deflection function (3), one can obtain the following:
(a) The strain energy takes the form
l
E
U I x y 2 dx (4)
20
Thus, E
l
(l x) n
3 3
U
2 I 0
0
l n
[( ) 4 a12 cos 2
2l 2l
x ( ) 4 a 22 cos 2
2l 2l
x
2 3 2 3
2a1 a 2 ( ) ( ) cos x cos x] dx
2l 2l 2l 2l
EI 0 4 2 l l
1 (l x ) n cos x dx )
n
n4
[ a ( (l x ) dx
64l 0 0
l
3
l l
81 a 22 ( (l x ) n dx (l x) n cos x dx)
0 0
l
2
l l
18 a1a 2 ( (l x) n cos x dx (l x ) n cos x dx )
0
l 0
l
IE 0 4 2 \ \ 2 \
(5)
Where:
\
1 U 3 [a1 2a1a2 a2 ]
J 1n
\
n 1
64 l
9 ( J 1n J 2 n )
1
\
81 ( J 3n )
n 1
J 10 0 J 20 0 J 30 1
n n( n 1)
Jhn Jh 2 2
( h ) 2 ( h ) 2
n 2
J 11 2 J 21 0 J 31
(3 ) 2
3
2 1 2 l 9 2 l 3
[ a1 ( x sin x) a2 ( x sin x)
8l 2
2 l 2 3 l
l l 2
3 a1 a 2 ( sin x sin x )]
l 2 l
l
(l x) p
T q0 p
u x dx
0 l
Thus, q 2 l l
T 0 p 2 [a12 (l x) p ( x sin x )
16 l 0 l
l 3
9 a 22 (l x) p ( x sin x)
3 l
l l 2
6 a1a2 (l x) p ( sin x sin x ) ] dx
Finally, l 2 l
q0 2
T [ a12 2 a1 a 2 a 22 ] (7)
16
Where: 1 1
I1 p
( p 1) ( p 2)
1 1
3 ( I1 p I2p )
2
1 1
9 ( I3p )
( p 1) ( p 2) 3 2 2
I10 I 20 0 I 30
1 p ( p 1) 3
I hn I h p2 1 1 1
h (h ) 2 I11 I 21 I 31
2 3
Thus, the critical value of the load P is found from the equation;
UT (8)
Which represents the condition when the equilibrium configuration changes from stable to
unstable mode, so that:
EI0 2 \ \ 2 \ q0 2 2
4 2
[a 2a a a ] [a1 2a1a2 a2 ]
The conditions for (q0 l)cr are:
(ql ) cr (ql ) cr
3 1 12 2 0 and 0 (9)
Or;
64l EI 16 a1
2
2 EI 0 \
a2
and
a1[ 2 2(ql) cr ] a2 [ 2 2(ql) cr ] 0
0 \
2l 2l
The possibility of buckling
a2 [ 0 \2 that the
\
EI
2 2 EI 0 occurs2 when these equations
EI
2 EI 0 \ give solutions for a1 and a2 different
(ql0 )\ cr ] a2of[ the2 equations
2determinant 2(qlmust
) cr ] be 0equal to zero, so that:
from zero. This requires
2l
2 cr 2 c r
2(ql) 2(ql) 0
2l
2l 2l
0
Which finally gives
EI 0 \ EI 0 \
2 2
(q 0 l ) cr
b b 2 4ac
2a
2(ql) 2(ql) 0
2 EI 0
2 cr 2 cr
2l 2
4
2l 2l
(10)
Where
a 4 ( 2
) \
The summation of the distributed load P cr can be calculated by integrating the intensity of the
load qx along the length of the column
b 2 ( \ \ 2 \ )
(l x) p
l
P q0 p dx \
cr
Where: l
c \P \ (q l)
\
\2
0 cr
0
m
EI (11)
cr
p 1 2 l2 b b 2 4ac
m ( )
2 ( p 1) 2a
In this case, the critical buckling load is found for combined axial distributed load, (q0 l/p+1), and
axial compressive force (P) applied at the ends as shown in Fig. (2). The case of column with
variable cross section subjected to an axial force (P) only has to be studied first. Then, the
combination with the previous case can be introduced. To study the case of axial compressive
force, the previous assumptions and the previous deflection curve can be also considered.
P
p= 0
p=1
p=2
n=0
n=1
n=2
qx qx qx Ix Ix Ix
l
q0 q0 q0 I0 I0 I0
5
Fig. (2) Column submitted to combined concentrated and distributed axial loading
The work done T by the axial load P at the end is:
l
P (17)
T y 2 d x
2 0 we obtain
Substituting Eq. (3) in Eq. (17),
P 2 2 (18)
T ( a1 9 a22 )
16 l
The strain energy is:
E I 4 (19)
64l T U (20)
IE 0 2 \ \ 2 \ P 2 2
4 2
(21)
[a 2 aa a ] (a1 9a2 )
Thus, the critical load Pcr is a function of the constants a1 , a2 and Pcr is found from:
3 1 12 2
64 l
Which gives,
Pcr
a1
0 16l
and
Pcr
a2
0 (22)
EI
Pcr 2 (23)
l
Where,
2
1 1
2
1( b b 144c )
144
\
, b 2 ( 9 )
\
When a distributed axial load acting on the column for which the intensity varies according to a
1
\
power of the distance along the column, Eq.
\ 2 (2), the distributed axial load (P ) (0wn weight)
\ \
, c
reduces the critical value of the load (Pcr) so we can put
1
EI (24)
Pcr 2
l
where the factor (), smaller than the factor () obtained in Eq. (23), gradually diminishes when
\ \
the distributed axial load (P ) increases and approaches zero when (P ) approaches the value
given by Eq. (11). The study of this situation depends on the ratio between the sum of the
\
distributed axial load (P ), which can be obtained by integrating the intensity of the load along
the length of the column, and the critical load of the column subject to a force (P) applied at the
ends only. Assuming this ratio is where:
P \ (own weight)
(25)
EI / l 2
The distributed axial load reduces the critical value of the load (Pcr). The value of the reduction
\
of the load (Pcr) is equal to the load P divided by the factor (m) and multiplying by the factor ()
. The critical load (Pcr) after reduction due to the effect of the distributed axial load can be put as
Eq. (24) where
(1 )
m
6
(26)
CRITICAL STRESSES
The preceding analysis gave the values of the critical loads for various types of cases of loading,
but the designer usually works with the stresses. In the case of a uniform cross section with end
conditions as the column under study, one end fixed and the other is free, the critical stress is
given by:
P
cr cr (27)
A0
2
Where, i = I / A0
Putting = l / i, then the critical stress is:
2E
cr (28)
4 ( ) 2
EI
2
E 2
Or cr 2
4l A 0 4 (l / i) 2
Therefore, the critical stress of a column is dependent only on the Young’s Modulus and the
slenderness ratio of the column.
Similar expressions can be obtained for the column of gradually variable cross section as
follows:
(a) If the column is subjected to axial load P only at the ends :
EI
Pcr
Then, l2
E E
cr
( ) 2 ( / ) 2
E
(29)
( m ) 2
Where, m
(b)If the column is subjected to the combined action of a distributed axial load and a compressive
force P applied at the ends
The total compressive force Pt , which is the sum of the own weight and the applied load at the
ends, will be as follows:
EI EI
( Pt ) cr 2
2
l l
EI
Thus, ( Pt ) cr ( ) 2
l
E E
and cr ( )
( ) 2
( / ( ) ) 2
E
( m ) 2
7
(30)
Conclusions
In this research, the theoretical analysis is developed to determine the buckling loads and
buckling stresses of columns with gradually variable cross section. A relatively simple but
8
accurate method is presented for the prediction of the buckling load and the collapse stresses for
columns under study.
From the previous results the following conclusions may be drawn
1. The critical buckling load for column with gradually changing cross-section under the action
of a distributed axial load, can be accurately determined through an explicit form by means
of energy solution tacking into account two terms of the Fourier Series in the deflection
function.
2. The proposed method of analysis offers many simple expressions from which the buckling
loads, hence the buckling stresses, can be calculated in the case of the action of the axial
distributed load. The results obtained from this work can cover any values of the power of
the moment of inertia function of the cross section and the power of the intensity function of
the load functions (represented by factors n, p). If the power of the intensity of the load is not
an integer (fracture number), the critical load can be found from the given curves.
3. The great advantage of the present analysis is the determination of the buckling loads in the
case of column under combined action of axially distributed load along the column and a
compressive force applied at the ends by simple and accurate equations. The results were
given also in tables and curves, which can be used by the designer engineers for wide ranges
of (q) and (P).
4. The simple concept of the “modified slenderness ratio” is applied for the calculation of the
critical stress for columns under investigation.
REFERENCES
[1] Timoshinco, S. P., and Gere, J. M., “Theory of Elastic Stability ”, McGraw-Hill Book
Company, London, 1983.
[2] Dinnik, A. N., in the book of smith, M. J. and Bell, B. J., “ Theory of Structures”,
George Godwin, London, 1980.
[3] Abdel-Lateef, T. H.,“ Buckling and Collapse Stresses for Eccentrically Loaded and
Imperfect Columns with Variable Cross Section ”, Bulletin of the faculty of
Engineering & Technology - Minia University, June 1993.
[4] Dinnik, A. N., “ Design of Columns of Varying Cross Section ”, translated from the
Russian by M. Moletz, trans. ASME, vol. 54, 1932.
[5] Zhou, Z. H. and Chan, S. L., “ Second-Order Analysis of Slender Steel Frames under
Distributed Axial and Member Loads ”, J. Struct. Engrg., ASCE, vol. 123, 1997.
[6] Robert Englekirk, “ Steel Structures ”, John Wiley & Sons, Inc., 1994.
[7] Salama, M. I., “Buckling of columns of variable cross section and columns with
holes”, M. Sc. Thesis, Minia University, 1999.
9
Table ( 1 ) Values of factor m Eq.( 12 ) P cr mEI /l2
p
0 1 2 3 4 5
n
0 7.84 16.18 27.95 43.78 64.36 90.34
n = 2
p = 0
p = 1
p = 2
n = 0
n = 1
l
q 0 q 0 q 0 I0 I0 I0
m
100
90
n=0
80
n=1
70
n=2
60 n=3
n=4
50
40
30
20
10
0
0 1 2 3 4 5 6
p
Fig. ( 3 ) Values of factor m in Eq.( 12 )
10
2.5
n =0
2.0
1.5
1.0
p=0
p=1
p=2
0.5
p=3
p=4
0.0
10.0
1.0
2.0
5.0
6.0
8.0
0.0
3.0
4.0
7.0
9.0
Fig. 4)
( Values of factor
11
12