Accepted Manuscript

An experimental investigation into failure mechanism of a full-scale 40 meters

high steel telecommunication tower
Jacek Szafran
PII:
DOI:
Reference:

S1350-6307(15)00148-X
http://dx.doi.org/10.1016/j.engfailanal.2015.04.017
EFA 2565

To appear in:

Engineering Failure Analysis

Received Date:
Revised Date:
Accepted Date:

6 February 2015
22 April 2015
23 April 2015

Please cite this article as: Szafran, J., An experimental investigation into failure mechanism of a full-scale 40 meters
high steel telecommunication tower, Engineering Failure Analysis (2015), doi: http://dx.doi.org/10.1016/
j.engfailanal.2015.04.017

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

An experimental investigation into failure mechanism of a full-scale 40 meters high steel

telecommunication tower
Jacek Szafran
Department of Structural Mechanics, Chair of Reliability of Structures, Faculty of Civil Engineering, Architecture and
Environmental Engineering, Lodz University of Technology, Al. Politechniki 6, 90-924 d, Poland
email: jacek.szafran@p.lodz.pl, tel. 48-42-631-35-64

Abstract: The main objective of this paper is to present and discuss failure mechanism, failure mode, as well as plastic
deformation of a lattice telecommunication tower obtained during a full-scale pushover test. The manuscript consists of
a detailed description of the tested structure and the experimental site. Displacements of particular nodes of the tower
are presented as a function of the external load. The main conclusion is that the rigidity of joints between particular
elements, which depends on thickness and diameter of connecting flanges and number and quality of bolts, determines
the failure mode of the compressed tower legs. In the article, values of axial forces under compression, taken directly
from the conducted test, were compared with the standard buckling resistance. On the basis of this comparison,
discussion about the effective slenderness factor has been generated, and the proper determination of this coefficient has
been proposed.
Keywords: failure mechanism, buckling, plastic deformation, structural failures, breaking load, code of practice,
deflection

1. Introduction
Experiments on full-scale engineering structures, although difficult to perform because of, both technical and
financial reasons, can produce results that are impossible to obtain in a different way. Structure response
searching in the context of, e.g. stresses, strains, displacements, and last but not least, failure mechanism and
failure mode, may help enrich the knowledge about structures like lattice towers.
Steel lattice towers are extensively utilized in telecommunication industry as supporting structures providing
services such as telephoning, wireless internet or television. The more pressing the mobile networking needs
of todays customers are, the more requirements are being imposed on telecommunication devices which, in
turn, causes repetitive replacements, upgrades, and modernizations. Antennas and radio units which get
attached to telecommunication towers have different shapes, dimensions and weight. They are installed at
various heights which considerably alters forces distribution of a supporting structure as well as carrying
capacity of tower elements. Taking into consideration constant change of operating conditions, there is
recurring need of determining bearing capacity of such structures. Due to the advancements in
telecommunication engineering, including a growing number of new technological solutions, the scientists
and civil engineers also contribute to the effort of meeting the new demands with numerous scientific
publications devoted to problems in the field of telecommunication structures..
As an example of current state of knowledge of high, slender engineering structures and, in particular, of
their dynamic behavior and fatigue, the works of Repetto and Solari [1, 2] may be mentioned.
One of the most practical and comprehensive scientific manuscripts dealing with communication structures
is one written by Smith [3].The author thoroughly elaborates on utilization of structures of that type, taking
into account a variety of design aspects: meteorological parameters, strength, fatigue, and etc., where the
conclusions coming from structural failures of masts and towers caused by external loads are particularly
important. One type of load that determines cross-section size of particular elements of steel lattice towers is
obviously wind load. The effect that the wind has on various engineering structures is a subject of many
research articles. The comparison of results taken directly from wind tunnel tests and those predicted by
Eurocode [4] for a slender tower structure can be found in [5]. Various problems in design, and analysis of
telecommunication structures in particular, were published in [6] by Travanca et al. From the standpoint of
engineering practice, a significant problem was tackled: the comparison of standards records and definitions
from various codes of practice, which change over time, with their impact on subsequent estimation of

carrying capacity of the structures with emphasis being placed on existing object evaluation. This problem
seems to be particularly significant for telecommunication towers due to changing nature of their load
conditions (technological requirements). Additionally, the subject of tower strengthening and upgrading,
where maintaining an object in a fair condition over years is concerned, is extremely valid. It is not
uncommon that these objects are utilized for several dozen years. Hence the constant need for modernization,
renovation, or upgrading to fulfill the above requirements [7]. The replacement of structures elements with
ones of larger cross-sections, manufacturing additional truss elements or adding weight to the foundations
are very common, exemplary attempts of upgrading telecommunication support structures.
Reliability modeling is the next aspect of research and development in this branch of engineering. The
requirements concerning the reliability are included in [8]. They impose a structural design process which
ensures compliance with top quality standards.
Computational probabilistic analysis and reliability assessment of steel telecommunication towers subjected
to material and environmental uncertainty can be found in authors previous works [9-12] and book of
Kamiski [13].
One of more interesting and challenging problems is either analytical or numerical determination of carrying
capacity of steel tower elements made of cold-rolled or hot-rolled L-sections. In particular, it involves the
determination of support conditions influence, load applying manner, failure mechanism, and failure modes
which results in carrying capacity estimation. Many scientific publications tackle the problem of structural
elements behavior and their plastic deformation in particular [14]. An analysis of such elements often used in
steel structures may involve full 3D behavior; consideration of axial, bending and shearing actions; various
slenderness ratios; loading and displacement eccentricities. The comparison of numerical calculations with a
full-scale destructive test may be found in the work of Lee and McClure [15]. The manuscript elaborates on
highly relevant aspects of structures composed of hot-rolled single angle shapes with eccentric connections
and, most importantly, allows for comparison of the obtained numerical results with full-scale, destructive
test data. The authors of [15] should be also noted for pointing out that, in the context of mechanical security
where the assessment of modes of failures and the confinement of failure are key, it becomes important to
predict the actual strength and failure mechanism of such towers with a reasonable accuracy for failure
scenarios in both static and dynamic regimes. Some consideration about non-local and local buckling and
their influence on overall stability of truss and arched structures can be found in [16, 17]. The use of FEM

analyses with the experimental study findings considerably extends the knowledge of buckling phenomenon
for structures elements.
2. Experimental, full-scale test
2.1. Experimental study objective
One of the main objectives of the experiment was to reveal failure mechanism and failure mode of the 40
meters high, steel, lattice telecommunication tower under breaking load.
The experiment has been conducted in such a manner that the external load was exerted on the tower and the
maximum compression forces in its legs were created. It was achieved by applying the load in the least
favorable direction for a tower of a triangular cross-section causing the maximum value of the compression
stresses at one of the legs.
Particular attention was paid to the observation of the tower legs behavior. One of the goals of the performed
test was the determination of their buckling resistance. Another purpose was to determine the nature of
destruction of diagonal bracings comprising hot-rolled angle sections. For these purposes, a tower of a
complex structure and specific bearing elements was chosen.
2.2 Tower description
The tower is a 40 meter lattice structure with a triangular cross-section. It is divided into seven sections. The
sections may be classified into two types, with convergent and parallel legs. The bottom part of the tower (up
to 34th meter) forms a pyramid frustum of convergence of 5%. The centerline dimension is 4.90 m at its base
and 1.50 m at the top. The upper part of the tower is a parallelepiped of a height equal to 6.0 m with the
cross-section of an equilateral triangle of side length equal to 1.50 m. Geometric division of sections and
their heights, the layout of elements in two bottom sections as well as the structure of the tower under
average exploitation are shown in Fig. 1. The general construction manner of particular sections shall be
presented as the following: the legs were made with round solid bars, the diagonal bracing members with
hot-rolled single angles, both symmetric and non-symmetric. The profiles of particular tower elements are
presented in Tab. 1 below:

Section
S-1
S-2
S-3
S-4
S-5
S-6
S-7

Legs
65
65
65
80
80
90
90

Diagonal bracings
60x60x6
60x60x6
60x60x6
60x60x8
90x60x8
90x60x8
90x60x8

Tab. 1. Selected tower element profiles, units are mm

Intersections of the diagonal bracings (of type X) are made with a spacer and a single bolt. The joints of the
diagonal bracing and legs were realized with gusset plates and bolts (two for a node). The legs at the
connections of individual sections of the structure were made using round plates (connecting flanges) welded
at their ends and an adequate number of bolts as presented in Fig. 2. Data concerning individual joints,
thickness and size of the flanges, and the number and type of bolts in individual sections were presented in
Tab. 2.
Section

S-1
S-2
S-3
S-4
S-5
S-6
S-7

Diameter of
lower/upper
connecting flange
180/180
180/180
200/180
220/200
230/220
260/230
270/260

Thickness of
lower/upper
connecting flange
25/25
25/25
30/25
30/30
30/30
35/30
35/35

Bolts in lower/upper
connecting flange

Number of bolts
in joint

M16x75(8.8)/M16x75(8.8)
M16x75(8.8)/M16x75(8.8)
M16x85(8.8)/M16x75(8.8)
M20x85(8.8)/M16x85(8.8)
M20x85(8.8)/M20x85(8.8)
M24x100(8.8)/M20x85(8.8)
M24x100(8.8)/M24x100(8.8)

6
6
6
6
6
6
6

Tab. 2. Data concerning leg joints of individual tower sections, units are mm
The tower was equipped with a climbing - cable ladder. The ladder is attached at every section to the tower
legs with two angle braces and bolts (Fig. 3).
In order to determine the mechanical properties of the structural steel tensile testing of the 14 specimens (5
for L-sections and 9 for full round bars), according to standard [19], were performed. Mechanical properties
obtained during the laboratory tests are listed in Tab. 3.
Mechanical properties
Young modulus
Lower yield strength
Upper yield strength
Tensile strength

Tab. 3. Mechanical properties of structural steel

Values
L-sections
205,0GPa
323,6 MPa
333,8 MPa
447,0 MPa

Legs round bars

203,3 GPa
284,7 MPa
290,0 MPa
467,4 MPa

2.3 Full-scale experiment

Full-scale experiments are the type of scientific and engineering effort which allows for understanding of
authentic behavior of a structure under ultimate load. As mentioned in [15], experimental tests are frequently
used as a validation procedure in the development process of numerical models. Taking engineering
structures into account, the complexity level, scale, and nonlinearities of any kind are considerably large
making it impossible to give analytical solutions in most structural problems. Thus, in engineering practice,
approximated numerical formulae and elements of lower complexity levels are being used. As one can see, it
may be concluded that conducting full-scale tests, although particularly difficult, laborious, and most of all,
costly, cannot be overstated enough. There is apparent need for testing with those features provided. Taking
the above into consideration, as well as analyzing the tower structure described in 2.2, the tested tower
undeniably complies with all the requirements above.
The experiment was carried out in November, 2014. As it is known to the author of this manuscript, no prior
experiments have been conducted for this type of telecommunication towers. There are about 2000 such
structures in Poland, which was the reason for choosing this particular type of the tower.
In order to conduct the test, the structure was provided with a diaphragm to which external load was applied.
It was designed, constructed and welded to the legs of the tower especially for the experiment. Its steel
elements, their dimensions, and attachment manner were designed to apply the pulling force to the whole
cross-section of the tower rather than a single joint connecting the bracings and legs. Such application of
load results in the most unfavorable load distribution for a tower structure of a triangular cross-section. The
technical solution of the diaphragm is depicted in Fig. 4. The element during assembly and testing is
presented in Fig. 5.
The experiment consisted of slow line pulling action with one end of the line attached to the diaphragm
described above and the other to a towing truck. The load was being increased in steps. It was caused by the
intermediate necessity of geodetic measurements at individual tower nodes. A load cell measuring the force
in the line was located right at the diaphragm.
The testing site is presented in Fig. 6 (scheme) and Fig. 7 (photo taken during the experiment). The tower
along with specially laid foundation constituted its main part. The structure was attached to a steel frame
equipped with special anchors enabling fixing into the foundation. Everything got loaded with pavement
slabs of dimensions 3.0 x 1.5 x 0.15 meters which summed up to the total weight of about 108 tons. The

mass of the foundation and the geometrical dimensions were determined beforehand on the basis of tower
stability calculations as well as on an analysis of over a dozen towers of the same type and height. As the
comparative analyses indicated, the total mass of the foundation of existing towers with height equal to 40
meters amounted to, with ground above, 60-100 tons. The foundation of the tower prepared for the test is
depicted in Fig. 8.
3. Failure mechanism of the tower
Yielding of one of the compressed towers legs at section S-5 is presented in the movies which can be
accessed

online

at

https://www.youtube.com/watch?v=qtLBw_RumVA&feature=youtu.be

and

https://www.youtube.com/watch?v=abDk-VRVTmc&feature=youtu.be. During the test, the changing values

of the external force were controlled and recorded. The data of the measurements of the external load were
gathered every 1.7 s. The measuring device was synchronized with special computer software which
correlated the data from the load cell and electric resistance strain gauges placed on the elements of the
structure. The leg plastic deformation occurred when the force recorded in the line reached about 108.7 kN.
It is worth noting that the loss of stability occurred in the leg of section S-5 where the diameter of the round
bar was equal to 80 mm. In the recorded movie mentioned above, the failure mechanism and plastic large
deformations of particular elements of the tower can be observed. The buckling of the leg occurred along the
plane parallel to the wall opposite to that bar which indicates that, in accordance with the assumptions of the
experiment, the force applied to the transverse cross-section of the structure remained alongside the bisector
of the angle between the walls of the tower. In the movie - the displacement of individual nodes and tower
deflection may be observed best if compared with the initial position of the structure.
3.1 Tower nodes displacements
Geodetic measurements of the displacements of structure nodes are highlighted in Fig. 9 and Fig. 10. It
should be emphasized that the measurements were taken for two different locations of the total station
theodolite. Arithmetic average of measurement values (planes X and Z of nodes A, B, and C) are shown in
Tab. 4.

Force in line [kN]

0
20
30
40
52
60
65
70
75
80
85
90
95
100
105

Point A
X
0

Point B

Point C

Z
0
3
7
6
6
8

X
0

Z
0

34
58
82
115
134

0
0
0
0
7
6

0
56
97
138
193
226

72
128
180
250
298

4
3
7
9
9

148
166
183
204
229
268

6
3
4
2
3
5

251
282
311
349
391
455

10
12
12
11
12
8

330
370
410
464
518
608

13
20
19
19
21
16

324
449

-10
-26

549
750

34
48

735
1012

49
83

588

-30

975

71

1320

101

Tab. 4. Displacements of monitored structure nodes, all measurements are given in mm

Due to the elastic characteristic of the line, the force recorded in the line decreased each time the pulling
effort by the towing truck paused. Therefore, the results obtained are approximated with an accuracy of
around 1.0 kN. One significant observation concerns the displacement along Z axis of node A which was at
the top of the leg that lost its stability. The buckling effect itself started when the force in the line was
reaching around 90kN. A change of displacement direction was the sign that the buckling eventually started
leading to the stability loss of the bar. Since then, the displacements of all the measuring points increased
dramatically.
During the process of line pulling, the back edge of the foundation started to rise. Taking into account the
mass and the geometry, such a behavior revealed a truly global effort of the tower. The vertical displacement
of the back edge of the foundation is presented in Tab. 5 below.

External load [kN]

0
40
73
80
85
90
95
100
105
110

Vertical displacement of the

foundation [mm]
0
0
10
10
10
40
60
120
170
190

Tab. 5. Vertical displacement of foundation with corresponding force in line

Deformed state of the tower after the experiment is shown in Fig. 11, the front view (left) and the base view
(right). The difference in the impact of bending on the upper part of the structure (above S-5) and the rest of
the tower, and the state of the leg after the stability loss may be observed.
Fig.12 depicts how the displacement of the leg bar occurred the main plastic deformations can be found in
the compressed leg of section S-5. Most importantly, the rotation of the cross-sections of the legs does not
occur at joints. It also happens not only in the buckled leg of section S-5, but also at the legs of neighboring
sections, S-4 and S-6, which is particularly visible in Fig. 13. The stiffness and thickness of the connecting
flanges, the number of bolts, and the attachment manner of neighboring diagonal bracing elements determine
the character of the stability loss. The plastic hinges of the legs under ultimate load occurred below the node
connecting sections S-6 and S-5, as well as above the node connecting sections S-4 and S-5, which is
precisely shown in Fig. 14.
In Fig. 15, the deformed state of the leg is shown from the view inside of the tower. Measured deformations
of individual leg cross-sections in section S-5 are presented in Fig. 16. It may be noticed that the largest
plastic deformations occurred on plane XZ. The plastic hinges were located at and of the span of the
section wherein larger deformations were observed for a cross-section located above, at of the span. The
fact worth noting is that, for the lower cross-section, the maximum displacements of 70 mm occurred on the
plane YZ. It should be mentioned that the measured values were recorded after the disassembly of all tower
elements took place. Therefore, the deformations are shown on the plot from the start of the leg, which
results from the fact that the disassembly of bolts in connecting flanges of particular tower legs caused
rotation of these elements and loss of elastic deformations.

One observation, captured also in the attached movie, is that tower legs, not bracing elements are crucial in
determination of reliability of a structure as understood in the serial reliability system. It is worth underlining
that the deformation of the diagonal bracings of section S-5 occurred at the time of the yielding of the
compressed leg. The effects of that deformation can be observed in Fig. 17 and Fig. 18.The main cause of the
deformation of L-section diagonal bracings is the vertical displacement of gusset plates located at the center
of the section. At the moment of adoption of a large deformation state, deflection of those plates occurs,
which causes bending of bracing elements. That way a very complex load state is created. Apart from an
eccentric (due to attachment manner) compression and relative tension, a moment which bends bracing
elements is created by the buckling behavior of the leg. Deformed gusset plates at joints of L-sections of
section S-5 are presented in Fig. 19 and Fig. 20. The confirmation of the thesis concerning diagonal bracing
bending can be witnessed in Fig. 21. A crack in material occurred near one bolt hole. It was also caused by
the element attachment manner which involved two bolts.
3.2 Standard buckling resistance versus experimental compression in the tower legs
According to standard [18], buckling resistance of a compression member in a lattice tower should be
determined as:
N b,Rd

Af y
,
M1

(1)

where is the reduction factor for the relevant buckling mode, A is the cross section of the tower element,
f y is the yield strength, and M 1 is the partial safety factor.

The most important issue in calculations of the buckling resistance of the tower elements is the
proper determination of . For constant axial compression in members of a constant cross section
the reduction factor should be calculated as follows:

1
2
2 eff

(2)

where

2
0,5 1 eff 0,2 eff
.

(3)

In formula (3) is the imperfection factor equal to 0.49 for full round rods and eff is the effective
slenderness ratio defined by:

eff k

,
1

(4)

where k is the effective slenderness factor, is the slenderness for the relevant buckling mode defined in
[18], and 1 93.9 235 f y .
In order to complete the presentation of the standard calculation procedure, the buckling resistance of the
tower legs in section S-7 and S-6 is introduced below. All the parameters, both geometrical and mechanical,
are collected In Tab. 6. Standard definition of the geometrical division of legs is presented in Fig. 22.
Length of the leg members:

L 600 cm

Distance between the diagonal bracing members and both ends of the legs:

L1 300 cm

Diameter of the leg member:

d 9 cm

Cross sectional area:

A 63.6 cm 2

Moment of inertia

J 321.9 cm 4

Radius of gyration:

i 2.25 cm

Grade of steel

S 235
E 210 GPa

Standardized value of the Youngs modulus:

Standardized value of the yield strength (reduced because of the thickness of the
element):
Partial safety factor:

f y 215 MPa

M1 1

Tab. 6. Geometrical and mechanical parameters of the tower legs in sections S-6, S-7
Considering the slenderness of leg members for the analyzed tower that are braced symmetrically, the
following formula should be implemented:

L1 300

133.3.
i 2.25

(5)

Effective slenderness factor k for the leg members of the analyzed tower is equal to 1.0 according to the
definition given in standard [18].Taking that value into account, the effective slenderness ratio is equal to:

eff 1.0

133.3
93.9 235 215

1.36.

(6)

Using all the previously determined values, the buckling resistance of the analyzed leg members can be
defined as:

N b,Rd

0.36 63.6 104 215 103

492.3 kN .
1

(7)

The maximal compression forces obtained during the experiment for the value of the external load equal to
108.7 kN are collected in Tab. 7; in fact, this is the breaking load for this particular structure.
Measuring point located on tower legs at:
middle of section S-7
top of section S-7
bottom of section S-6
middle of section S-6
top of section S-6

Experimental compression forces [kN]

521.3
461.0
756.5
654.3
410.2

Tab. 7. Experimental compression forces in the tower legs in sections S-7, S-6
There is quite significant scatter of the results. In the opinion of the author, the reference values of the axial
forces shown in Tab. 7 are for centers of individual sections, which are 521.3 kN for section S-7 and 654.3
kN for S-6. Taking into account that the legs in both sections, S-7 and S-6, have the same diameter and the
same system of diagonal bracing (resulting in the same L1 parameters), it can be concluded that the greater
value can be treated as buckling resistance of such elements. It should be underlined that the loss of the
stability of the tower leg took place in section S-5, not in S-7 or S-6, where the axial compression forces
were measured. Therefore it can be assumed that real buckling resistance is greater than 654.3 kN for these
particular leg members.
In order to compare the results of the axial compression forces, taken directly from the experiment with the
standard buckling resistance, the values of N b,Rd as a function of

effective slenderness factor k are

presented in Tab. 7.
Effective slenderness factor k

Standard buckling resistance N b,Rd [kN]

0.95
0.90
0.85
0.80
0.75
0.70

527.9
569.0
623.2
661.9
713.5
768.2

Tab. 7. Standard buckling resistance of the tower leg members in sections S-7 and S-6 as a function of the
effective slenderness factor
Is worth noting that for the determination of the buckling resistance, mechanical parameters like Youngs
modulus or yield strength were taken from the standard assumption for the particular grade of steel: S235 in

this case. There are significant differences between standardized values of these parameters and ones
obtained during the laboratory tests (see Tab.3 and Tab.6). Higher values of the mechanical steel properties
have, without any doubt, additional and positive influence on the real buckling resistance of the tower legs.
Thats why, in opinion of the author, comparable value of the standard N b,Rd to experimental axial forces is
the one obtained for slenderness factor k equal to 0.85.Taking that assumption into account, one can
compare the experimental force during the compression of the tower legs equal to 654.3 kN and standard
buckling resistance (obtained for k 0.85) equal to 623.2 kN. Achieving the experimental results allowed
for comparison with the analytical ones based on the standard assumptions.
4. Conclusions and perspectives
The manuscript presents the results of a full-scale, push-over test of a 40 meters high lattice
telecommunication tower. The focus has been placed on the overall behavior of the structure under the
ultimate load. One of the purposes of the experiment was to reveal the failure mechanism and failure mode.
This goal has been achieved. The buckling of the leg member in section S-5 of the tower has been registered.
After observing progress of the displacements increase, plastic deformations and, above all, failure mode of
the tower leg member, one can conclude that joints between particular sections of the tower determine the
form of the stability loss. Connecting flanges at both ends of the leg remained rigid, whereas cross-sections
of the full round rods above and below these nodes underwent the plastic deformation. It confirms that the
rigidity of the connections resulting from the flange thickness, number and diameter; and bolts quality is
significantly higher than the rigidity of the leg members.
A comparative analysis of the axial compression forces in the tower legs and standard buckling resistance
have been presented. The results of N b,Rd calculations using standard procedure are significantly lower than
the ones obtained experimentally.

It is worth pointing that standard procedure of buckling resistance

determination does not take into account quality of joints of particular tower legs. The value of buckling
resistance, what was proven during the experiment, depends not only on the diagonal bracing distance
between the elements, cross-sectional area of the legs, and mechanical properties of the material, but on the
quality of the joints as well. Consequently, it is postulated that effective slenderness factor k should take the
value of about 0.85 in buckling resistance analysis in the context of the standard records.

The results shown in the article demonstrate a significant raise of the back edge of the foundation under
ultimate load. It can be concluded that, for this particular case, the buckling of the tower leg occurred before
the loss of the stability of the foundation, which have caused overturning of the structure. It is worth
underlining that, for other practical realizations (e.g. for lower values of the foundation mass or other
geometrical properties), this behavior can be completely different which can cause stability loss of the
foundation(s).
The results of the conducted experiment will foster development of behavior analyses of similar steel
telecommunication towers, which concerns mainly carrying-capacity analyses of already existing structures
with a prospect of telecommunication equipment addition. In such cases, nonlinear analysis like the one
carried out in work [20] is planned to be conducted. Moreover, knowing the results of an experiment with a
full-scale structure, it is further planned to develop a numerical model for predicting structural failure.
Behavior of diagonal bracing elements will be the next aspect subjected to computer simulation. In L-section
modeling with complex strain state and complex geometric nonlinearities, aspects involving number and
disposition of bolts, edge effects and tightening torque will be examined as given in [21]. As observed during
the experiment, the elements suffer damage at the point where they are connected to legs, the fact which will
also have to be taken into consideration in future perspectives.
5. Acknowledgements
The author of this document would like to express his deepest gratitude to T-Mobile Poland S.A. whose
financial support made this project possible. He would like to express his gratefulness to Mr. Micha
Wjcicki. Acknowledgements should be given to Professors Kazimierz Rykaluk at Wrocaw University of
Environmental and Life Sciences and Marcin Kamiski at Lodz University of Technology for their support
in the research project.

References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]

Repetto MP, Solari G. Dynamic along wind fatigue of slender vertical structures. Eng Struct 2001;23:16221633.
Repetto MP, Solari G. Wind-induced fatigue collapse of real slender structures. Eng Struct 2010;32:38883898.
Smith BW. Communication structures. London: Thomas Telford Publishing; 2007.
Eurocode 1: Actions on structures-Part 1-4: General actions Wind actions. Brussels: CEN; 2005.
Belloli M, Rosa Z, Zasso A. Wind loads on high slender tower: Numerical and experimental comparison. Eng
Struct 2014;68:24-32.
Travanca R, Varum H, Real PV. The past 20 years of telecommunication structures in Portugal. Eng Struct
2013;48:472-485.
Albermani F, Mahendran M, Kitipornchai S. Upgrading of transmission towers using a membrane bracing
system. Eng Struct 2004;26:735-744.
Eurocode: Basis of structural design. Brussels: CEN; 2002.
Kamiski M, Szafran J. Random eigenvibrations of elastic structures by the response function method and the
generalized stochastic perturbation technique. Arch Civ Mech Engrg 2009;9(4): 5-32.
Kamiski M, Szafran J. Stochastic analysis of the forced vibrations of steel telecommunication tower. Journal
of Civil Engineering, Envirnoment and Architecture 2014;61: 46-56.
M. Kamiski, J. Szafran. Stochastic Finite Element Analysis and Reliability of Steel Telecommunication
Towers. Computer Modeling in Engineering & Sciences 2012;vol.2060, no.1;1-25.
Kamiski M, Szafran J. Eigenvalue analysis for high telecommunication towers with lognormal stiffness by
the response function method and SFEM. Computer Assissted Mechanics Engrg Sci 2009; 16: 279-290.
Kamiski M. The Stochastic Perturbation Method for Computational Mechanics. Chichester: Wiley & Sons;
2013.
Lee PS, McClure G. A general three-dimensional L-section beam finite element for elastoplastic large
deformation analysis. Comp&Struct 2006;84:215-229.
Lee PS, McClure G. Elastoplastic large deformation analysis of a lattice steel tower structure and comparison
with full-scale tests. J Constr Steel Res 2007;63:709-717.
Cybulski R, Walentyski R, Cybulska M. Local buckling of cold-formed elements used in arched building
with geometrical imperfections. J Constr Steel Res 2014;96:1-13.
Jankowska-Sandberg J, Koodziej J. Experimental study of steel truss lateral-torsional buckling. Eng Struct
2013;46:165-172.
Eurocode 3: Design of steel structures Part 3-1: Towers, masts and chimneys Towers and masts. Brussels:
CEN; 2006.
EN ISO 6892-1: Metallic materials Tensile testing Part 1: Method of test at the room temperature.
Brussels: CEN; 2009.
Albermani F, Kitipornchai S, Chan RWK. Failure analysis of transmission towers. Eng Failure Anal
2009;16:1922-1928.
Dang Hoang T, Herbelot C, Imad A. On failure mode analysis in bolted single lap joint under tension-shering.
Eng Failure Anal 2012;24:9-25.

FIGURE CAPTIONS
Fig. 1. View of the tower, static scheme, 3D view for sections S-6 and S-7
Fig. 2. Legs and diagonal bracing joints details
Fig. 3. Attachment of the cable-climbing ladder to the tower legs
Fig. 4. Diaphragm view (left and right elevations)
Fig. 5. Steel diaphragm during the assembly process (left) and the experiment (right)
Fig. 6. Experiment test site with a lattice steel tower
Fig. 7. Top view of the site during the experiment
Fig. 8. Tower foundation
Fig. 9. Displacements of the observed nodes in X direction
Fig. 10. Displacements of the observed nodes in Z direction
Fig. 11. Deformed tower shape after the experiment view from the front (left) and the base (right)
Fig. 12. Deformed leg shape view from the left side (left) and the base (right)
Fig. 13. Deformed leg shape view from the front (left) and the base (right)
Fig. 14. Nodal displacements between sections S-6(S-4) and S-5
Fig. 15. Deformed leg shape view from the inside of the tower
Fig. 16. Plastic deformation measurements of the leg in section S-5, units are mm
Fig. 17. Deformed shape of the diagonal bracings in section S-5
Fig. 18. Deformed shape of the diagonal bracing (view along the element)
Fig. 19. A joint connecting the leg with the diagonal bracings
Fig. 20. Cracked L-section element
Fig. 21. L-section crack at the cross section with the bolt hole
Fig. 22. Standard assumption of the tower legs geometrical division

S-4

S-5

S-6

6000
5000
5000
500

S-7

6000

S-3

6000

S-2

6000

S-1

1500

6000

+ 40,5m n.p.t

1850

+ 42,0m n.p.t

4900

Fig. 1. View of the tower, static scheme, 3D view for sections S-6 and S-7

Fig. 2. Legs and diagonal bracing joints details

Fig. 3. Attachment of the cable-climbing ladder to the tower legs

Fig. 4. Diaphragm view (left and right elevations)

Fig. 5. Steel diaphragm during the assembly process (left) and the experiment (right)

(h = 41,23 m) C
Longitudinal section
diaphragm

(h = 30,25 m) B
F
load cell
wire cable

(h = 19,25 m) A

towing truck
z
D (h = 1,78 m)
118,96 m
x

120,41 m

Top view
towing truck

Fig. 6. Experiment test site with a lattice steel tower

Fig. 7. Top view of the site during the experiment

Fig. 8. Tower foundation

110
100
90
80
70
60
F [kN]
50
40
30
20
10
0

100

200

300

400

500
point A

600

700

800

point B

900 1000 1100 1200 1300 1400

point C

Fig. 9. Displacements of the observed nodes in X direction

ux [mm]

110
100
90
80
70
60
F [kN]
50
40
30
20
10
0
-10

10

20

30
point A

40

50
point B

60

70

80

point C

Fig. 10. Displacements of the observed nodes in Z direction

90

100
uz [mm]

110

Fig. 11. Deformed tower shape after the experiment view from the front (left) and the base (right)

Fig. 12. Deformed leg shape view from the left side (left) and the base (right)

Fig. 13. Deformed leg shape view from the front (left) and the base (right)

Fig. 14. Nodal displacements between sections S-6(S-4) and S-5

Fig. 15. Deformed leg shape view from inside of the tower

Fig. 16. Plastic deformation measurements of the leg in section S-5, units are mm

Fig. 17. Deformed shape of the diagonal bracings in section S-5

Fig. 18. Deformed shape of the diagonal bracing (view along the element)

Fig. 19. A joint connecting the leg with the diagonal bracings

Fig. 20. Cracked L-section element

Fig. 21. L-section crack at the cross section with the bolt hole

L1
L1

Fig. 22. Standard assumption of the tower legs geometrical division

Highlights

1. All the results described in the paper are obtained via full scale pushover test of the 40 meters high
telecommunication tower. As far as author of this study is informed, no prior experiments have been
conducted for this type of telecommunication tower.
2. The mechanism of failure for the structure of the high complexity has been revealed.
3. Formulation of the conclusion for the effective slenderness ratio that is necessary for the bearing
capacity approximation has been proposed.