Euler Buckling

Arnold Mukuvare

University of Surrey

Work carried out on: 5th March 2013

Report completed: 10th April 2013

Abstract:

In designing members in engineering it is of vital importance that the members can satisfy
specific strengths, deflections and stability requirements (Hibbeler R.C., 2011). A column is
a long, straight and slender member that is experiencing compressive axial loading.
Catastrophic failure can result from the columns deflecting laterally at enormous levels. This
effect is known as buckling. The purpose of this experiment is to investigate the effect of
increasing load on members, with different supports, critical loading, the derivation of the
load equations, buckling and effects of length of the members. Calculations of the moment of
inertia (I) were necessary because of how irregular beams are in the real world an error of
0.12% for the shorter column and 2.76% for the longer column. Figure (6) showed a direct
proportion relationship between the load and deflection for a beam resting horizontally from
the graph the elastic modulus obtained was 4.3% from the actual value. Loading columns
while vertical produced asymptotic graphs of the load against the deflection with the both
graphs approaching critical loading value Figure (8). The graphs produced reliable results as
would be expected from Euler buckling. Errors of 4.75% and 2.77% were recorded between
the theoretical and experimental values of the 550mm and 750mm column respectively. Such
errors are acceptable but it is important to note that in the case of the 750mm column the
elastic modulus was assumed to be 200GPa as noted by the demonstrator. Concept of
slenderness ratio reviewed and the effect of lengths given that all the other variables remain
constant.

Page | 1
Contents
Contents ................................................................................... Error! Bookmark not defined.
Introduction ................................................................................................................................ 3
Theory ........................................................................................................................................ 4
Procedures and Equipment ........................................................................................................ 6
Results ........................................................................................................................................ 8
Discussion ................................................................................................................................ 10
Conclusion ............................................................................................................................... 11
References ................................................................................................................................ 12
Appendix .................................................................................................................................. 13

Page | 2
Introduction
A column is a structural member experiencing compressive loading at the either end, given
however, that the cross sectional dimensions are considerably smaller than the length which
will be the direction in which a load is applied. The cross-sectional area is to be kept constant
in order to analyse the possible deflection. Buckling is the phenomenon that happens when a
column is experiencing an axial load and deflects due to the loading being big enough.
Buckling can lead to failure if the compressive loading is big enough. It is important to note
that buckling failure is not as a result of the material since after applying the loads the
material retains its original shape hence the elastic limits will not have been reached. So
buckling failure occurs mainly to loads that are smaller than the elastic yield strengths. To
engineers it is important to be able to predict buckling levels due to how destructive,
dangerous and sudden it can occur.
The critical load of a column is the maximum axial load that a column can support before
failure and any load greater than the critical load will cause the beam to deflect laterally and
bow out.

Masses of the beams are neglected and focus is on the elastic modulus and the cross sectional
area. Equation (1) shows the moment of a beam about a certain distance from the support, the
beams in this experiment will be either pinned or fixed at both ends. An ideal column is
assumed meaning that the column is made of homogeneous material, is perfectly engineered
(cross-sectional area constant throughout the structure), material is linearly elastic and the
load is applied exactly in the middle of the cross-sectional area. Upon reaching the critical
load the column tends to be unstable and the structure becomes reliant on how well the
column can restore its natural structure.

This report analyses the theoretical and experimental results of the out of plane deflection
which results when the struts are sustaining a load axially. A close to ideal beam situation is
also assumed meaning that, the beam is perfectly uniform and loaded also that the material in
the beam is homogeneous. This however, in the real world cannot be the case because of lack
of accurate precision from manufacturing equipment and uneven distribution of load around
the beam.

When a column is exposed to compressive loading starting from zero, the column would start
at a state of balance (equilibrium). Small lateral deflections occur so long as this state of
balance is not exceeded. Meaning that the beam remains below it elastic limitations. Critical
loading is the maximum compressive loading that a column can take before reaching unstable
equilibrium. Any further increase in loading would result in catastrophic failure. Critical
loading for long and slender columns occurs below the elastic limits.

Page | 3
Theory
The theory of buckling of columns under a compressive axial load was discovered by
Leonard Euler (1707-1783). Columns are to remain long, straight and slender with specific
supports at the ends. A compressive load P is applied and has to pass through the centre of
the cross-sectional area. If a column is pinned the pins are assumed frictionless. Ideal beams
are assumed meaning that these columns have no defects. Figure (1) is of an ideal column
which is symmetric with deflections occurring only in one plane.

Figure 1 (a) Buckling of a pin-jointed column under an axial load P, (b) the Free-body force
diagram showing internal forces (Civil Engineering Lectures., 2013)

If a pin jointed beam is experiencing a certain load P while taking a distance x from the top
pin and a displacement (y(x)). The bending moment M(x) about the point will be the load
multiplied by the displacement (P*y(x)). Applying this to the equation for beams we would
have an expression of modulus of elasticity and load times the deflection. The derivation of
this equation is beyond this report.
Instead of using y(x) to denote deflection will be used;

(1)

Where E is modulus of elasticity and I is the area moment of inertia

Applying differential equation for the column would see the introduction of new term k2,
which is represented in Equation (1.1) the full derivation of the differential equation is found
in the Appendix and the final solution would show the load equation as shown by Equation
(2).

(1.1)

(2)

Where n=1, 2, 3, … depending on the support.

Page | 4
The above analysis helps in determining the minimum load P at which buckling occurs. In
this experiment the beams are either pin-ended or fixed which means that n=1 or 2
respectively and substituting these n values would give critical loads. This critical load can
also be called the Euler buckling load.
(3)

(4)

In the introduction it was noted that the critical load is independent from material strength, Equations
(3) and (4) shows that the critical load is dependent on the modulus of elasticity E, moment of inertia I
and length l. The first buckling mode is related to n=1 and is given in Equation (3), so a different
value of n would mean that theoretically an increase, but this is not the case since columns buckle
once they have reached the critical load so n has no practical interest. Figure (2) shows different
supports to 4 columns.

Figure 2. Buckling on columns with different support systems (Birch D., 2012)

If a column has a rectangular cross-section as shown in Figure (3) below would fail in a
certain way.

Figure 3. Rectangular cross-sectional area of a column (Den Hartgog J.P. 1949. Pg 56)

Buckling failure will occur at the lowest value of moment of inertia. Figure (3) would buckle
about the x-plane rather than the y-plane. Achieving a balance that the ( )
would result in better preferred columns. Calculation of moment of inertia is dependent on
the width and height, both I’s should be calculated to obtain the smaller value.

Page | 5
 (5.1)

(5.2)

Euler buckling load for ideal conditions is reached instantaneously and the failure is
immediate, but this is not the case in the real world deflections are noted with the increase in
load until reaching a critical load as shown by Figure (5).

Figure 5: A column under a load (a); ideal Euler load (P), deflection (δ) curve (b); actual
observed results (c) (Henslee E. and Ward S., 2013)

When a beam is loaded mid-span calculations of bending moments would leave a relationship
between the load P and deflection. Equation (6) shows a direct proportion relation between
the load and the deflection. Figure (6) in the Results section shows a straight line from the
first experimental results.

(6)

The Euler buckling and the arguments presented earlier will only work if the material
behaviour stays elastic. Moment of inertia (I) can be defined as the cross-sectional area A and
the minimum radius of gyration r.
(7)

Substituting Equation (8) into Equation (3), then diving both sides by the area a formula for
critical stress for a column Pc/A, taking also the minimum radius of gyration (r) as a safety
feature since (I) was minimum. From the critical stress a ratio between the length and the
radius of gyration (L/r) would determine the critical stress since the critical load, cross
sectional area and elastic modulus are all constants. Equation (9) will be always applicable so
long as the material limits are not exceeded.

⁄
(8)

Page | 6
Figure 6. Critical stress against slenderness ratio (Dr AULD D.J., (2010))

The slenderness ratio is how flexible a column can be, this helps explain the actual effect of
length (long, intermediate or short). Figure (6) shows the relation while the material remains
below the proportional limits hence the column still behaves in an elastic manner. If the
slenderness ratio from Figure (6) is less than 89 (L/r)<89 the yield stress will be exceeded
before buckling so the Euler formula can not be used. Table (1) in the appendix also shows
the lowest slender ratio for different types of steels.

Procedures and Equipment

Equipment
Steel beam 1 dimensions (550x19x3)mm
Steel beam2 dimensions (750x19x3)mm
Ruler
Micrometre Dile
Micrometre screw gauge
Strut rig machine
Masses

Procedure
Before starting the experiments, instruments have to be calibrated, with the smallest degree of
measurement noted for error calculations. Three experiments were carried out in this lab and
each case is explained in this section. The rig was adjusted to fit the columns for each
experimental case Figure (7) shows the strut machine used in the experiment.
Table (1) shows the modulus of elasticity and vital in error calculations.

Case 1: Horizontal lying knife edge (pin) setup (550mm)

The 550mm column was placed horizontally on the knife edge and the rear of the positions of
the strut gauge adjusted. On the marked mid-span a stirrup was placed perpendicular to the
length of the bar, here loads were going to be attached on the load carried suspended
underneath the stirrup. The dial gauge was mounted vertically above the stirrup just touching
and calibrated to zero carefully. Loads were carefully applied starting with 1N until 10N and
the deflection given on the dial noted in Table (2). A plot was made of the load (W) against
the deflection (δ). Since the moment of inertia has been calculated, length known the gradient

Page | 7
of this curve can be used to find the modulus of elasticity. The working out and further
explanation is carried out in the results section. The elastic modulus can be obtained from
calculating bending moments or graphically (gradient of the slope) Figure (6).

Case 2: Vertical lying knife edge (pin) setup (550mm column)

The piston was first adjusted to releases any forces. The bar was carefully taken and placed
vertically on knife edge supports. The dial was placed horizontally and at the mid-span of the
column. Pressure was applied to make sure the beam deflected away from the dial. The load
was taken off completely and the dial reset to zero. The starting deflection readings were
taken at 25N and recorded in Table (3). A graph was plotted of the load W against the
displacement δ and comparisons made against the Euler formula this is also carried out in the
results section.

Case 3: Vertical lying clamped (fixed) setup (750mm)

The strut rig was adjusted to fit the 750mm span. Using the rule the middle of the span was
determined and the column was fixed into position using the clamps. The piston was turned
to allow load and observe if the column deflected away from the dial. The pressure was taken
off after the observation and the dial was calibrated to zero. Since the length of the column
had increased increments began at 50N and deflections were recorded in Table (3), and a
curve was plotted of load W against displacement δ.

Results

A column is never ideal, so using a micrometre on three different points of the span
dimensions of width and height can be taken and averaged as shown in Table 4 below.
Equation (10) was used to calculate the means.

Table 4: Micrometre readings of the width and height readings at different spans
Beam 1 2 3 AVERAGE
550mm Width (b) 19.95 19.97 19.97 19.96
±0.005mm
Height (d) 2.95 2.99 2.92 2.95
±0.005mm
750mm Width (b) 19.92 19.92 19.92 19.92
±0.005mm
Height (d) 2.98 2.98 2.98 2.98
±0.005mm

∑ (10)

Using the aide of Table (4) and Equation (6.1), the moment of inertia was calculated and
recorded.

Page | 8
Table 5: Experimental moments of inertia (I) including instrument error
550mm 42.70±(6.25x10-10)mm4
750mm 43.93±(6.25x10-10)mm4

The loads and deflections of the horizontal 550mm column on a knife edge support were
carefully measured and recorded in Table (1). A plot of the load W against the deflection δ
showed a linear relationship with all the points lying on the straight line apart from one of the
points. This agrees with Equation (6), which shows that the line should cross the x-axis at
zero and the constants in the equation making up the gradient. The slope can be calculated as
( ) which is 0.21N/mm. Since all the other quantities making up the gradient are
already known for the experimental values apart from the Elastic modulus we can use
Equation (8) to find the experimental value of the elastic modulus.

(8)

Substituting the values of length and moment of inertia using this value the Elastic modulus
would be 209GPa. A 4.5% error is obtained between the theoretical and experimental value
of the elastic modulus. Figure (7) shows the values from Case 1

Figure 7. Load against deflection with 5% error bars and a line of best fit

In Case 2 the 550mm beam was pinned and load was gradually applied, results are tabulated
in Table (3). Knowledge obtained from the Theory section that when a column is pinned
Equation (3) can be used to find the critical load.
Table (3) also contains the values of the fixed 750mm beam applying Equation (4) since n=2
would give the experimental values of the critical load.

Table 5. Critical loading experimental values of 550 and 750mm columns

Column Pinned (n=1) Fixed (n=4)
550mm 291.2N -
750mm - 616.64N

Page | 9
The values from Table (5) show the experimental values for both columns with their different
supports. The theoretical values are calculated in the Appendix; errors of 4.75% and 2.77%
were obtained between the 550mm and 750mm column respectively.

The results obtained from the two experiments shows graphs that are asymptotic after
reaching a certain load. The 750mm column has a steeper slope and reaches the maximum
loading of 600N. The load W against deflection δ graphs for both the 550mm span and
750mm column are plotted in Figure (9) which is even more accurate in its approach to the
critical values than the experimental calculation.

Figure 10: Load against deflection comparison of (750 and 550)mm columns including 5%
error propagation

Discussion
Human errors are inescapable whenever an experiment is carried out. Tackling this however,
would be best done by tasking individuals to certain tasks and repeating those tasks three or
more times. Hysteresis error is visible on the strut machine which was slow to react to initial
inputs of force. There is a 4.5% error between the theoretical modulus of elasticity and the
experimental value, reduction of this error would be achieved by repeating the experiment
three or more times also using a more sensitive dial. Errors of 0.12% and 2.76% were
recorded for the moment of inertia of the 550mm and 750mm beams respectively. For the
550mm beam such an error is within reason but increasing the points where the readings are
taken would certainly lower the error.
A 4.75% and 2.77% error was obtained for the critical loading of the 550mm and 750mm
beam. Again such errors are acceptable, but however, it is important to note that for the
750mm column the elastic modulus was assumed to be 200GPa and not calculated.

Page | 10
From all the errors that are obtained the results are overall with reason and solidify that the
experimental methods used are a good way in obtaining the critical loading. The small error
of 4.75% for the 550mm column also reinforces the fact that different elastic modulus do not
play much of an effect to the critical loading. The modulus that was obtained from Figure (7)
was very good and coincides with those in Table (1). To add Figure (8) and (9) produced an
even more accurate estimate of the critical load.

Figure (6), shows that buckling will occur if the material is long and remain with the elastic
limits. Shorter columns have higher buckling ratios than longer, slender columns and the
relationship between the critical stress and length2 is inverse which means there is an
accelerated fall in the critical stress as the length increases assuming that the radius of
gyration stays constant and the graph obtained is hyperbola. It is also worth noting that Euler
buckling does not act as a safety factor but just the maximum load a column a beam can take
before bowing out.

Conclusion
Upon starting this experiment the demonstrator had mentioned that the elastic modulus of
both columns was 200GPa, and Case 1 gave a result which was very close to the actual value
of the modulus. All the measurement errors from instrument calibrations were in cooperated
in the calculations. Ideal conditions were assumed for the columns although this is not
possible in reality, from these calculations however the small percentage errors between the
theoretical and experimental values were still very reasonable.
Improvements can be definitely made to the experiment starting by using more modern
instrument that have no hysteresis errors due to over use, or maybe even laser dials can be
used to measure the deflection in that way the possibility of the dial falling off will be
eliminated.

Page | 11
References
AULD D.J. (2010). Buckling of columns. Available:
http://web.aeromech.usyd.edu.au/AMME2301/Documents/mos/Chapter09.pdf. Last accessed
8th April 2013

Civil Engineering Lectures. (2013). Buckling of columns. Available:

http://lecture.civilengineeringx.com/building/structural/FIGURE-3.88-Buckling-of-a-pin-
ended-column-under-axial-load.jpg. Last accessed 25th Mar 2013

Beardmore R. (2012). Useful tables. Available:

http://www.roymech.co.uk/Useful_Tables/Mechanics/Struts.html. Last accessed 25th Mar
2013

Den Hartgog. J.P., 1949, Strength of Materials, First Edition, Constable and Company
Limited

Hibbeler R.C., 2011, Mechanics of Materials, Eight Edition, Pearson Prentice Hall

Jensen A. and Chenoweth H.H, 1983, Statics and Mechanics, Library of Congress Catalog in
Publication data, Fourth Edition

Benham P.P., Crawford R.J., and Armstrong C.G., 1996, Mechanics of Engineering
Materials,Second Edition, Pearson Education Group Limited

Philpot T.A., 2008, Mechanics of Materials; An integrated learning system, John Wiley and
Sons Inc

Birch D., 2009, Design and Component production; A reference guide for engineering
students, Surrey University

Engineering toolbox. (2010). Young modulus. Available:

http://www.engineeringtoolbox.com/young-modulus-d_417.html. Last accessed 28th Mar
2013.

Page | 12
Appendix

Important tables:

Table1. Limitations of Euler’s formula (Jensen A and Chenoweth H.H, 1983. pg327)

Lowest value of L/k

for which Euler’s
Modulus of Proportional limit Formula Applies
Material Elasticity, (MPa) (MPa)
Structural steel, low alloy 200,000 342 76.0
Structural steel, carbon 200,000 214 96.0
Douglas fir, select 11,000 31.0 59.2
structural

After carrying out the experiment results were collected and recorded in tables.

Table 2. 550x19x3mm steel column on horizontal knife edges (CASE 1)

Load Deflection(δ)
±0.5N ±0.05mm
1 0.2
2 0.4
3 0.6
4 0.8
5 1.1
6 1.3
7 1.5
8 1.7
9 1.9
10 2.1

Table 3. Load and deflection readings of a 550mm pinned column and 750mm fixed column
550mm column pinned 750mm column fixed

Load(W) Deflection(δ) Load (W) Deflection(δ)

±0.5N ±0.05mm ±0.5N ±0.05mm
25 0.1 50 0.2
50 0.2 100 0.2
75 0.4 150 0.2
100 0.7 200 0.4
125 1 250 0.5
150 1.4 300 0.7
175 2.2 350 0.9
200 4.2 400 1.3
225 7.6 450 1.9
250 13.9 500 2.9

Page | 13
275 14 550 4.2
277 14 600 14.9

Diagrams excluded from the text:

Figure 6. Strut rig machine and a column fixed vertically

Following the experiment Case 2 and 3, Figure (8) and (9) were plotted for the load against deflection
of the 550mm and 750mm columns

Figure 8. Plot of load against deflection of 550mm column vertically pinned

Page | 14
Figure 10. 750mm fixed column load against deflection plot

Important Equations in the paper:

(A1)

Where E is modulus of elasticity and I is the area moment of inertia

Introducing partial differential equation formula and a constant k;

(A1.1)

√ (A1.1a)

Removing the –k2δ to the left hand side and adding it to the second partial differential
(A1.1b)

Applying the knowledge of second order partial differential equations, the general solution to
the equation

) (A1.2)
Where A and B are constants which can be determined from the boundary conditions which
are
δ(0) = 0 and δ(L) = L

Hence A=0 and 0=Bsin(kL)

If B=0, this would mean that the beam is not being deflected. But for any other value of B,
would mean that sin(kL)=0, would mean that kL=nπ, where the n is an interger. Rearranging
this would lead us having a definition of solution for k.

(A1.2a)

Substituting back into Equation (1.1) and rearranging

Page | 15
 (A2)

The Euler load for a column which is pinned at both ends has n=1. Hence the formula for
column with both ends pinned is

(A3)

For a column that is fixed at both ends n=2. So the formula for a column with both ends fixed
(A4)

Moment of inertia, were the

(A5.1)

(A5.2)

(A6)

Moment of inertia by definition is:

(A7)
Substituting Equation (7) into Equation (3)

⁄
(A7.1)

Dividing the critical load by the area, critical stress

⁄
(A8)

To calculate mean
∑ (A9)

Theoretical values of moment inertia (I):

550mm ideal beam:

750mm ideal beam:

Theoretical values of critical load (Pc) if E=200GPa

550mm pinned beam:

Page | 16
750mm fixed beam:

Calculating Errors:

Elastic modulus for 550mm:

550mm moment of inertia:

750mm inertia (I):

Critical load on 550mm column pinned:

Critical load on 750mm column fixed:

Standard deviation error: √ ∑ ̅

Standard deviation error of 550mm and 750mm columns:

550mm 750mm
Width (d) 0.01224 0
Height (b) 0.03536 0

Page | 17
Page | 18