M22308 – BEHAVIOR OF

STRUCTURES
Coursework

University of Portsmouth
UP2225686
 Table of Contents

Task 1: Deflections of Pin Jointed Structures and Beams ..........................................................2

Part 1: Experimental Task ...........................................................................................................2
Introduction ............................................................................................................................... 2
Procedure ...................................................................................................................................3
Result and Discussion.................................................................................................................3
Part 2: Theoretical Task ..............................................................................................................6
Vertical Deflection of Pin-Jointed Truss .....................................................................................6
Deflection of Cantilever Beam at Free End ................................................................................8
Discussion.................................................................................................................................10
Task 2: Behaviour of Compression Members (Buckling)..........................................................12
Introduction ............................................................................................................................. 12
Theory ......................................................................................................................................12
Procedure .................................................................................................................................14
Results and Discussion .............................................................................................................15
Case A & Case B: Comparison Table ....................................................................................... 17
References ................................................................................................................................ 20

1
Task 1: Deflections of Pin Jointed Structures and Beams

Part 1: Experimental Task

Introduction

Plane trusses, with their diverse configurations, are a mainstay in structural

engineering. From roofs and bridges to cranes, they play a critical role in supporting
various structures (Horikx, 2022). The present study delves into the design
considerations for deflection in plane trusses. It serves as a practical demonstration of
how any pin-jointed structure like a truss deflects during increasing load conditions. In
structural design, calculating deflection is crucial for any type of structure. Deflection
refers to how much an element deforms or bends under applied forces. Accurate
deflection calculations are necessary to control how much a structure bends, ensuring
it functions as intended (Williams, 2020). The deflection of structures with pinned
connections can be analyzed using the principles of structural mechanics such as “Unit
load method”, which work on virtual work concept (Fuchs, 2016). These structures,
known as pin-jointed frames, consist of members joined at their ends by hinges or
pins. Unlike rigid connections, pins allow for movement and rotation between
members, but prevent any axial force transfer between them. In this study, the principle
of virtual work was used to determine the deflection of a truss structure (Figure 1). The
unit load method was employed to calculate this deflection.

This experiment has two main objectives. The first is to explore how a pin-jointed
structure deflects under load. The amount of bending or deformation the structure
undergoes will be measured. Secondly, this measured deflection will be compared to
the theoretical value obtained by virtual work method. To further solidify our
understanding, the deflection of the pin-jointed structure will be compared to a beam
with similar material properties, loading conditions, and span. This comparison would
help us estimate how different structural arrangements respond to forces.

2
 Fig. 1 Truss structure

Procedure

In the experimental phase, a point load was applied on the G joint point of pin-jointed
frame, and its deflection was determined through a gradual increase in the load. The
deflection was systematically examined during both unloading and loading cycles.

To ensure accuracy, dial gauge was correctly set, and a gentle tap on the gauge's front
was done at the experiment's beginning. Load hanger was firstly treated as having "no
load," and the corresponding gauge readings was measured. Subsequently, the load
was increased in 20N increments, and the gauge readings were measured for each
individual increase, with particular attention to the sudden increase in load. The
process was then reversed by decreasing the load in 20N decrements until it reached
zero, and the gauge readings were recorded at each step. The collected data was
arranged following the calculation of average deflection values corresponding to 0N-
100N.

Result and Discussion

Our experiment employed a staged loading and unloading process to investigate

deflection behavior. Table 1 summarizes the collected data, while Figure 2 depicts the
graphical representation of deflection curves for both unloading and loading phases.
As expected, the results demonstrated a clear trend: deflection increased
proportionally with the applied load. This approves our hypothesis that the structure
deflects more significantly under heavy load conditions. In addition, the data reveals

3
that the maximum deflection values were similar during both loading and unloading
cycles.

Table. 1 Deflection recorded at point G

Load
Deflection while Loading Deflection while Unloading
(N)
(mm) (mm)
0 0 0

20 0.25 0.29

40 0.39 0.42

60 0.43 0.45

80 0.48 0.49

100 0.53 0.53

Deflection Curve
0.6

0.5
Deflection (mm)

0.4

0.3

0.2

0.1

0
0 20 40 60 80 100 120
Load (N)

Loading Unloading

Fig. 2 Deflection curves of Loading-unloading cycles

Our results demonstrated that deflection consistently increased with higher loads in
both scenarios, slight differences between loading and unloading, like 0.25mm at 20N
versus 0.29mm, revealed a small hysteresis loop, indicating potential energy
dissipation due to factors like minor plastic deformation or friction. Additionally, the
minimal deflection at zero load after unloading (0.53mm), closely resembling the initial

4
zero-load deflection (0mm), suggests the structure's ability to regain its original shape,
emphasizing its integrity (Mukhopadhyay, 2020).

In theory, the deflection curve for this structure was predicted to be linear (Ju et al.,
2011). However, as our experimental findings show a non-linear behavior. Several
factors could contribute to this discrepancy. Firstly, the structure itself might exhibit
non-linear behavior due to large deformations or material properties. Secondly, the
theory assumes perfectly pinned joints where loads act exactly at the connection
points. In real-world scenarios, this might not be the case (Thai and Kim, 2011).
Factors like the eccentric loading (loads not acting perfectly at the center), members
self-weight, or unevenly distributed loads can introduce deviations from the ideal pin
behavior. Additionally, manufacturing tolerances and slight imperfections could lead to
limited fixity in structure joints that might also contribute to non-linear deflection (Wang
et al., 2006).

The experimental results reveal a significant difference between the actual deflection
and the theoretical predictions. Particularly, the vales of theoretical deflection exhibit
a perfectly linear behavior, whereas the experimentally measured deflections showed
a non-linear trend. This difference becomes more pronounced for loads exceeding
35N, where the experimental deflection is (roughly 0.3mm) greater than the theoretical
values. This can be due to several reasons as external factors, equipment or material
issues (Wang et al., 2006).

There are several strategies to address deflection in trusses (Fawzia and Fatima,
2010). The theoretical equation suggests that deflection is inversely proportional to
both the material's modulus of elasticity (stiffness) and the member's second area of
moment (resistance to bending). Therefore, using larger members with a greater
second are of moment or employing a material with a maximum modulus of elasticity
can effectively decrease deflection. Generally, trusses with more members tend to
experience less deflection. However, including members might not always be feasible
due to architectural considerations and design constraints. Exploring alternative truss
geometries can be another approach to minimize deflection. However, this strategy
might be limited by architectural necessities (Rand and Givoli, 1995).

5
Part 2: Theoretical Task

This theoretical task comprises of solving the deflection of a pin-jointed frame and a
cantilever beam using the unit load method and virtual work principles.

Vertical Deflection of Pin-Jointed Truss

Vertical deflection at joint G is calculated theoretically in this section.

The pin-jointed truss data to use is given below and also shown in Figure 3.
• a = 2.1 m (Based on last digit of ID = 6)
• b = 2.1 m (Based on last digit of ID = 6)
• W = 50 + 8 x 10 = 130 kN ((Based on second last digit of ID = 8)

A B C

= 2.1 m

G
D E F
W = 130 kN
3 @ a = 2.1 m

Fig. 3 Dimensions and Load on Frame

The free body diagram of the pin-jointed frame is shown below in Figure 4.

Fig. 4 Free Body Diagram of Truss

6
Figure 5 shows the real member forces with their magnitude highlighted by the colour
bar shown on the right. The unit is kN.

Fig. 5 Real Member Forces of Pin Jointed Frame

Figure 6 shows the member forces due to unit load with their magnitude highlighted
by the colour bar shown on the right. The unit is kN.

Fig. 6 Unit Load Member Forces of Pin Jointed Frame

Now, a table is constructed below that utilises real force, force due to unit load, area,
length and elastic modulus of each member to determine the value of PrPvL/AE. The
sum of these values gives the final value of vertical deflection at Joint G of the frame. It is
shown in Table 2.

7
 Table 2. Vertical Deflection at Joint G
Pr Pv A L E
Member PrPvL/AE
(kN) (kN) (mm2) (mm) (kN/mm2)
AB 260 2 1000 2100 205 5.33
BC 130 1 1000 2100 205 1.33
AE 183.85 1.414 1000 2960 205 3.75
BE -130 -1 1250 2100 205 1.07
BF 183.85 1.414 1000 2960 205 3.75
CF -130 -1 1250 2100 205 1.07
CG 183.85 1.414 1000 2960 205 3.75
DE -390 -3 1250 2100 205 9.59
EF -260 -2 1250 2100 205 4.26
FG -130 -1 1250 2100 205 1.07
34.97 mm

The vertical deflection at Joint G is computed as 34.97 mm. It is positive as it is in the

direction of unit load.

Deflection of Cantilever Beam at Free End

A cantilever beam is considered using the dimensions of pin-jointed frame above. The
span of this beam is same as the lower members of the frame which is equal to 6.3
meters. The same point load of 130 kN is acting at free end of the cantilever beam as
in the far-right joint of the frame. The total material volume of the pin-jointed frame is
computed to be 26,205,000 mm3. The total volume along with length of 6300 mm gives
a cross-sectional area of the beam as 4159.52 mm 2 ≈ 4160 mm2. Using this cross-
sectional area, the width and breadth of the steel beam are computed as 55 mm x 75
mm.

55 mm

75 mm

8
The free body diagram, shear force diagram, bending moment diagram of the beam
considered is shown in Figure 7.

Fig. 7 FBD, SFD and BMD of Cantilever Beam

9
Using the unit load method and virtual work, deflection for a beam can be computed
from following equation.

𝐿
𝑀𝑟 𝑀𝑣
𝑖𝛿 = ∫ 𝑑𝑥
0 𝐸𝐼
Assuming the point load as W and section distance as x then,

𝐿
(𝑊 x 𝑥)(1 x 𝑥)
𝛿=∫ 𝑑𝑥
0 𝐸𝐼
𝑊𝐿3
𝛿=
3𝐸𝐼

130 𝑥 63003
𝛿=
55 𝑥 753
3 𝑥 205 𝑥 12

𝜹 = 𝟐𝟕, 𝟑𝟑𝟓 𝒎𝒎 = 𝟐𝟕. 𝟑 𝒎

The deflection at free end of cantilever beam is computed to be a hugely large value
of 27.3 meters. Practically, such a large value of deflection is not possible. This huge
value can result from a number of factors such as human error or the equivalent
conversion of frame into beam dimensions.

Discussion

The substantial disparity in deflections observed between the cantilever beam and the
cantilever pin-jointed frame, subjected to an equivalent load, is predominantly
attributable to their respective structural configurations and mechanisms of load
distribution.
Cantilever Pin-Jointed Frame:
Deflection at the extreme right joint: 34.97 mm
Pin joints connect each member of a pin-jointed structure, permitting rotational motion
while limiting translational motion. A localized deflection occurs at the joint where the
load is applied due to the redistribution of the load among the members of the frame
when a load is applied at the extreme right joint. Over time, the deflection decreases
as one progresses away from the laden joint.The diminished deflection observed in

10
the frame in comparison to the beam can be attributed to the redistribution of load
across the interconnected members, which effectively restricts the deflection at any
one location.
Cantilever Beam:
Deflection at the free end: 27.3 meters
In contrast, a cantilever beam is a simple structural component that is loaded at one
extremity and supported at the other. The bending stresses generated by the load
propagate along the entire length of the beam, leading to a substantial increase in
deflection as one move further away from the support. The substantial difference in
deflection between the frame and the cantilever beam can be attributed to the
concentrated load at the free end, which generates a high bending moment and a
considerable deflection. Furthermore, in a nonlinear relationship with the square of the
distance from the support, the deflection of a cantilever beam increases significantly
towards the free end.

11
Task 2: Behaviour of Compression Members (Buckling)

Introduction

In structural engineering, buckling describes a sudden and excessive bending of a

slender member subjected to compression. This occurs even though the compressive
force might be well below the material's capacity to handle compression (de Miranda
Batista, 2005). The “critical buckling load” is the specific axial compressive force that
triggers buckling. Several factors influence this critical load, including the member's
length, end conditions, material properties, cross-sectional properties, the eccentricity
of applied load, and any initial curvature (Southwell, 1932).

The present study investigated the buckling behavior of a slender strut with initial
curvature under both eccentric and symmetrical/axial loading conditions. A graphical
method known as the “Southwell plot” was used to analyze the experimental data.
This method allows us to measure the critical buckling load, the eccentricity of the
applied load, and the initial curvature of the strut. The discussion section also explored
the buckling behavior differences without and with eccentricity, the benefits of using
the Southwell plot method, and the influence of the strut's end conditions.

Theory

The critical buckling load (Pcr) is the axial compressive force that triggers buckling in
a slender structure. It can be calculated using the following formula:

In this;

I: Second moment of area of the member's cross-section around buckling axis

E: Material modulus of elasticity

Le: Effective length of member, including end conditions

Strut with Initial Curvature:

A strut with initial curvature is shown below. It is represented by following equation with
maximum central amplitude as ao.

12
   .x 
y0 = a0 .sin 
 l 

x
ao yo

Following figure shows compression load P being applied to strut.

P x P
ao y

ymax y

y*
The elastic bending behaviour of strut is given by the following equation:

d 2 ( y − y0 ) M
=−
dx 2 E. I

Using the boundary conditions, the solution of differential equation is given by:

2   .x 
. a0 .sin 
l 2
 l 
y=
 2
 2 − 
l 

where:

P
=
E.I

The rearrangemenet of above equations gives:

y0
y=
P
1−
PE

13
 a0
ymax =
P
1−
PE

P
a0 .
PE
y* =
P
1−
PE

Finally, additional central deflection is given as:

 y *
y* = PE .   − a0
 P

Strut with Initial Curvature and Eccentric Force:

A strut with initial curvature and eccentric force is shown below.

P P
x e
ao y

ymax y

y*

Additional central deflection for this case is given by:

 y *  2 
y* = PE .   − a + . e
 P  0 8 

Southwell Plot:

Southwell plot could be used to analyze experimental data related to buckling. This
plot involves graphing y* (central deflection multiplied by load applied) versus y*/P
(central deflection divided by load applied). The slope and intercept of this plot could
be used to analyse the factors discussed above.

Procedure

The experiment method involved setting of the strut inside the frame as specified in
the instructions. However, various components were excluded. 2 knife-edge brackets
were then positioned inside the groove equivalent to zero-eccentricity. The

14
counterweight upon loading arms was corrected to ensure the arm's weight did not
affect strut. A 0.98 N (100g) weighed load hanger was used, and its weight was
factored into the total weight measurements. The scale was then set to a designated
reference reading. The experiment proceeded by gradually putting weights on the
hanger while gauging the corresponding deflections. Recommended weights included
20N followed by seven additional weights of 5N, 1N, and 0.5N each. The stop bracket
limited the maximum deflection to ensure the strut remained within the elastic range
and avoided damage. The actual load put on strut was calculated and the entire
procedure was then repeated with knife-edge brackets positioned inside groove
equivalent to maximum eccentricity. For this scenario, a weight sequence of 10N,
followed by five weights of 5N each, a 1N weight, and lastly, a 0.5N weight was used.

Results and Discussion

Table 3 and 4 shows the load-deflection results for strut with axial load and
eccentrically applied load respectively.

The experiment revealed an interesting trend in deflection behavior, as shown in Table

3. Up to a weight of 21 N (corresponding to an actual load of 28 N), the deflection of
the strut remained zero. This observation suggests that the strut behaviour was rigid
within this load range. However, as the weight was increased beyond 21 N, the
deflection started to increase progressively, reaching a value at 35.2 N.

Table 3. Axial loading deflection

Deflection t*
Weight Load P
(mm) y*/P
(N) (N)

1 1.3 0 0.00
21 28.0 0 0.00
26 34.7 0.5 0.01
31 41.3 1 0.02
32 42.7 1.5 0.04
33 44.0 3 0.07
34 45.3 5 0.11
35 46.7 26.5 0.57
35.2 46.9 46 0.98

15
The data in Table 4 indicates a positive correlation between weight and both deflection
and the calculated ratio for eccentrically loaded strut. In other words, as the weight
applied to the structure increased, the deflection and the ratio also increased.

Table 4. Eccentric Loading Deflection

Weight Load (P) Deflection Actual Deflection (y*) y*/P

(N) (N) (mm) (mm) (mm/N)

1 1.33 6 0 0.00

11 14.67 10 4 0.27

16 21.33 13 7 0.33

21 28.00 18 12 0.43

26 34.67 28 22 0.63

27 36.00 31 25 0.69

28 37.33 36 30 0.80

Figure 8 complements the above findings by presenting the Southwell plots for both
zero as well as highest eccentricity. Southwell plot determined the critical buckling load
and the highest load a strut can withstand before buckling occurs.

Southwell Plot
120

100

80

60
y*

y = 47.005x - 0.1245
40
y = 40.14x - 3.8465
20

0
0.00 0.50 1.00 1.50 2.00 2.50 3.00
-20
y*/P

Initial Curvature and Axial Load

Initial Curvature and Eccentrically Applied Load
Linear (Initial Curvature and Axial Load)
Linear (Initial Curvature and Eccentrically Applied Load)

Fig. 8 Southwell Plot for axial and eccentric loading

16
The essential information extracted from Southwell plot is as follows in Table 5.

Table 5. Euler Buckling Load

Case Euler Buckling Euler Buckling Load Comments
Load (Theoretical)
(Experimental)
Case A: y = 47.005x - PE = 𝜋 2 EI/L2 Theoretical PE >
Axial Load 0.1245 E = 205 kN/mm2 Experimental PE
PE = 47.005 N I = bd3/12 = 25.4 x (Case A and B).
a0 = 0.1245 mm 1.63/12 = 8.67 mm4 Yielding = 250 x 40.64
A = 25.4 x 1.6 = 40.64 = 10.16 kN
Case B: y = 40.14x – mm2 Strut will fail at the
Eccentrically 3.8465 L = 600 mm Euler buckling load as
Applied PE = 40.14 N PE = 48.73 N buckling load is
Load a0 + 𝜋2e/8 = significantly lower
3.8465 than the yielding load
0.1245 + 𝜋2e/8 = of the strut.
3.8465
e = 3.01 mm

Figure 9 depicts the deflection and load curves for both experimental scenarios.
Results showed that the strut exhibited greater buckling resistance under zero
eccentricity conditions compared to the highest eccentricity scenario. By analyzing the
graphs, we can estimate the approximate buckling loads:

• Zero eccentricity: 45 N
• Maximum eccentricity: 35 N to 40 N

17
 Load vs Deflection Plot
50

40
Deflection y* (mm)

30

20

10

0
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0

-10
Load P (N)

Initial Curvature and Axial Load Initial Curvature and Eccentrically Applied Load

Fig.9 Load curve and deflection curve

The Southwell plots provide valuable insights into the strut's behavior (Southwell,
1932). The plot of zero-eccentricity exhibits a steeper slope compared to the plot of
maximum eccentricity. This steeper slope indicates a higher or maximum “critical
buckling load” for the zero-eccentricity condition (Gregory, 1960). Furthermore, the
load-versus- deflection curves highlight the significant impact of eccentricity. The zero-
eccentricity curve shows minimal deflection until reaching the buckling point.
Conversely, the curve of maximum eccentricity exhibits enhanced deflection from the
start, with values significantly higher than the zero-eccentricity case throughout the
loading process. Additionally, the load of buckling failure for the zero-eccentricity
condition is significantly higher compared with maximum eccentricity condition. These
observations collectively demonstrate how eccentricity in loading can drastically
decrease the stability of a lateral compression member. Eccentricity introduces
bending, creating a vulnerability to both Euler and lateral buckling, ultimately reducing
the strut's compression capacity (Howlett, 1972).

Southwell plot is a valuable graphical technique for analyzing the buckling behavior of
slender compression members. It offers several key advantages which are as follows
(Blum and Rasmussen, 2016):

18
1. Versatility: This method allows for the determination of several crucial parameters,
including the “critical buckling load” (also known as the “Euler buckling load”), the initial
curvature of the member, and the applied load eccentricity.

2. Simplicity: It is a straightforward graphical technique, making it easy to understand

and interpret the results.

Experiments showed that initial curvature can slightly increase a strut's critical buckling
load, while eccentricity significantly reduces it. Eccentricity creates bending, promoting
both lateral as well as Euler buckling, which together decrease the strut's compression
capacity. These findings emphasize the importance of considering both factors in
compression member design to ensure safety and optimal load-bearing capacity
(Ostertag et al., 2019).

Compared to a pin-ended strut, a fixed-end configuration offers additional support.

This reduces the effective length of the member, leading to a significant increase in its
“critical buckling load”. Consequently, the strut exhibits greater stiffness, potentially
resulting in a drastic reduction in lateral deflection. Conversely, pin-ended struts have
a tendency to buckle within one plane. In contrast, fixed-end struts might buckle in
several planes. However, the member's cross-sectional properties significantly
influence this multi-planar buckling behavior. Generally, altering the end conditions
from pinned to fixed ends could substantially improve a strut's lateral stability, leading
to considerably reduced deflections (Ziemian, 2010). In nutshell, Southwell plot
technique proves to be a valuable tool in structural design and engineering. It offers a
simple and efficient way to determine the critical of slender strut members.

19
References

Blum, H. B. and Rasmussen, K. J. 'Buckling and design of columns with intermediate

elastic torsional restraint'. Proceedings of the Annual Stability Conference Structural
Stability Research Council Orlando, Florida.

de Miranda Batista, E. (2005) 'Modelling buckling interaction', Phenomenological and

Mathematical Modelling of Structural Instabilities: Springer, pp. 135-194.

Fawzia, S. and Fatima, T. 'Deflection control in composite building by using belt truss
and outriggers systems'. Proceedings of the 2010 World Academy of Science,
Engineering and Technology conference, 25-27.

Fuchs, M. B. (2016) Structures and their analysis. Springer.

Gregory, M. S. (1960) The buckling of structures: with particular reference to structures

containing bolted angle-section members and lightweight structures. University Of
Tasmania.

Horikx, M. P. (2022) 'Conceptual structural design: understanding structural

performance'.

Howlett, J. H. (1972) An investigation into the structural behaviour of thin-walled

aluminium alloy welded battened struts. Durham University.

Ju, S., Jiang, D., Shenoi, R. and Xiao, J. (2011) 'Flexural properties of lightweight FRP
composite truss structures', Journal of Composite Materials, 45(19), pp. 1921-1930.

Mukhopadhyay, R. (2020) Development of Autonomous Truss Modules Utilizing

Shape Memory Alloys for Actuation/Sensing Functionalities. North Carolina State
University.

Ostertag, A., Toader, T.-N., Bertsche, B. and Sobek, W. 'System-safety in the

application of adaptive load-bearing structures'. 2019 Annual Reliability and
Maintainability Symposium (RAMS): IEEE, 1-6.

Rand, O. and Givoli, D. (1995) 'Reduction of the periodic thermoelastic deformation in

truss-structures by design refinements and active loads', Computers & structures,
54(4), pp. 757-765.

20
Southwell, R. V. (1932) 'On the analysis of experimental observations in problems of
elastic stability', Proceedings of the Royal Society of London. Series A, Containing
Papers of a Mathematical and Physical Character, 135(828), pp. 601-616.

Thai, H.-T. and Kim, S.-E. (2011) 'Nonlinear inelastic time-history analysis of truss
structures', Journal of Constructional Steel Research, 67(12), pp. 1966-1972.

Wang, C.-Y., Wang, R.-Z., Chuang, C.-C. and Wu, T.-Y. (2006) 'Nonlinear dynamic
analysis of reticulated space truss structures', Journal of Mechanics, 22(3), pp. 199-
212.

Williams, M. (2020) Structures: theory and analysis. Bloomsbury Publishing.

Ziemian, R. D. (2010) Guide to stability design criteria for metal structures. John Wiley
& Sons.

21