Engineering Structures and Technologies

ISSN 2029-882X / eISSN 2029-8838

2019 Volume 11 Issue 1: 11–16
https://doi.org/10.3846/est.2019.8856

COMPARATIVE ANALYSIS OF THE BUCKLING FACTOR

OF THE STEEL ARCH BRIDGES

Sigutė ŽILĖNAITĖ*

Vilnius Gediminas Technical University, Saulėtekio al. 11, LT-10223, Vilnius, Lithuania

Received 17 October 2018; accepted 01 February 2019

Abstract. The dominant axial compressive force makes the arches become extremely sensitive to the loss of stability. Their
stability analysis was first initiated in the late 20th century. The first stability research of single arches was carried out in-
plane at the elastic stage of the arches. Later the behaviour of arches in the elastic-plastic stage, the initial stresses and
geometric imperfections before the arch buckles were also assessed, the effective length of the arches and the out-of-the-
plane arch strength conditions were being identified as well as the effect of the temperature on the stability of the arch. The
expression of the critical force of the arches connected by vertical hangers with a chord and its dependant elements were
defined by Petersen in the late 20th century. The design methodology for the formal design of arches connected by vertical
hangers with a stiffening girder is presented in Annex D of the Eurocode 1993-2. Nevertheless, the area of application and
the main assumptions are not defined. The first part of the comparative analysis identifies the assumptions for arch bridge
modelling under which the buckling factor β dependence curves in Figure D.4 of Annex D to Eurocode 1993-2 can be
applied. In the second part a comparison of the the normative βEC factor value and the one established by the numerical
experiment with the increase in the number of hangers and change in the hanger network form is presented.
Keywords: buckling length factor, comparative analysis, steel bridge, network arch.

Introduction
The arch bridges are some of the most popular, the old- Pi et al., 2008; Zhao, Guo, & Dou, 2013) were initiated.
est and the most elegant bridges built in the urban and Meanwhile, the latest trend is the effect of temperature
suburban areas. However, the dominant axial compressive on the stability of single steel arches (Backer, Outtier,
force makes the arches become extremely sensitive to the De Pauw, & Van Bogaert, 2010; Guo, Wang, Lu, Zhang, &
loss of stability in plane and out of the plane of the arches. Zhu, 2016; Heidarpour, Bradford, & Othman, 2011; Pi &
The summarized overviews of the stability of single arches Bradford, 2014).
and arch bridges can be found in the published articles The expression of the critical force of the arch bridges
(Pi & Trahair, 1999; Pi & Bradford, 2006) as well as in with vertical hangers was known since the late 20th centu-
books (Galambos, 1988; Ziemian, 2010). The first stabil- ry (Lebet & Hirt, 2013). It is the usual expression of Euler’s
ity research of single arches was carried out in-plane at critical force with an additional coefficient C the numeri-
the elastic stage of the arches. Major contribution to the cal value of which depends on the ratio between the height
analysis of the stability of single arches was made by the of the arch rise and the span length, the ratio between the
following scientists: Pi, Trahair and Bradford. Pi and Tra- arch and the stiffening girder, the number of hangers and
hair (1999) examined the in-plane stability of the single their cross-sectional area. In formal terms in order to en-
two-hinged circular steel arches in the elastic-plastic stage sure the stability of single arches or arch bridges, Table D.4
while Pi and Bradford (2004), Pi, Bradford, and Tin-Loi (single arches) and Figure D.4 (arch bridges) of Annex D,
(2008) analysed the rigidly fixed arches. Gradually the ho- of the Eurocode 1993-2 (Lietuvos standartizacijos depar-
rizons were broadened by the analysis of nonlinear struc- tamentas, 2007) are applicable. In Figure D.4 the buckling
ture behaviour of arches. The assessments of the geometric length factor values β EC are presented as curves, which
imperfections and the initial stresses of the arch behaviour depend on the number of hangers as well as on the height
prior to buckling (Pi & Trahair, 1999; Pi & Bradford, 2004; of the arch rise and the span length.

*Corresponding author. E-mail: [email protected]

Copyright © 2019 The Author(s). Published by VGTU Press

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unre-
stricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
12 S. Žilėnaitė. Comparative analysis of the buckling factor of the steel arch bridges

With the emergence of the new arch bridge form in the were used to constitute the dependence curves of the
middle of the 20th century – the network arch bridge – buckling length factors of Eurocode Annex D Figure D.4.
the classical expression of the critical force acquired a new The research presents the buckling length factor depen-
shape (Graβe, Teich, Tveit, & Wendelin, 2004; Schanack, dence on the number of hangers and grid shape.
2009; Tveit, 2005; Brunn, Schanack, & Steimann, 2004).
In the oldest version the buckling resistance of the single 1. The comparative analysis of the buckling
arch is viewed as the performance of the general system of length factor of arch bridges
the arch and the stiffening girder interconnected by elastic with vertical hangers
supports – the hangers (Teich, 2011; Tveit, 2005). So far it
is the only expression of critical force, which also evaluates This chapter deals with the numerical and analytical ex-
the bending stiffness of the stiffening girder. The second periments to determine the limits of the application of
(populist) expression of the critical force is described as Figure D.4 of the Eurocode 1993-2 for steel arch bridges
the sum of the Euler force of the flexibly fixed rod and the with vertical hangers. The buckling length factor values of
total support forces against elastic supports (Heidarpour the single arches with hinged or rigid supports are pre-
et al., 2011). This expression is rich in geometric indices sented in the form of curves in Table D.4 of Annex D,
such as the arch radius, angle of inclination, number of depending on the height of the rise and the span length
buckling half-waves, hanger cross-section factors, etc. as well as on the shape of the arch. Likewise, Annex D,
However, the application of both critical forces is possible Figure D.4 presents the values of the buckling length fac-
only when the stiffening girder is absolutely rigid. There tor of arches with a tension chord and vertical hangers
are no solutions for the versions with the flexible chord. connected to it, which, in addition to this, are depend-
In general, the scope and the results of the research on the ent on the number of hangers. As the amount of hangers
stability of network arch bridges are poor. Due to the large increases, the arch supporting effect grows. The value of
static uncertainty the engineering calculation methodol- the buckling length factor decreases respectively, and the
ogy is quite primitive. It is known that the arch buckling critical buckling strength of the arch increases.
resistance depends on the amount of hangers, the nature In order to determine the conditions and assump-
of the load and the arch bending stiffness (Schanack, tions of the application of the Figure D.4 of the Eurocode
2009). It is stated that the effect of bending stiffness of 1993-2, stability analysis was carried out with the use of
the stiffening girder on the arch stability is low (Schana- “SOFiSTiK” – the finite elements software. Three arch
ck, 2009). It has been established that the critical force of bridges with vertical hangers are modelled: in the first one
the network arch bridge arches is approximately 4 times the arch and the stiffening girder have separate founda-
the critical force of the arches with vertical hangers (Lars- tions, in the second one the arch and the stiffening girder
sen & Jakobsen, 2011). In one of the most recent articles have a common foundation and one movable support, and
(Backer, Outtier, & Bogaert, 2014) a numerical experiment in the third one both supports are rigid (Figure 1). The
was performed to determine the in-plane arch buckling arch is connected flexibly to the stiffening girder. For each
length factor at the elastic-plastic stage while modelling a model of the arch the values of axial forces and buckling
real arch bridge by finite plane elements. parameters are obtained. The critical buckling force given
Meanwhile, the application area of Figure D.4 of An- in Annex D of the Eurocode is analogous to the expression
of Euler’s critical force where the arch’s calculated length
nex D, Eurocode 1993-2 is not defined. The ratio between
is equal to the product of multiplying of the half length of
the bending stiffness of the arch and the girder, the iden-
the arch and the buckling length factor. The expression of
tification of buckling length factor values for the cases of
the buckling length factor is obtained by converting the
more than 11 hangers, the static structure of the bridge
buckling critical force expression given in Annex D of the
and the shape of the arch are not known as yet. The ap-
Eurocode 1993-2, where the critical force is equal to the
plication possibilities for the curves of buckling length
product of multiplying of the arch’s axial force and the
factor for the purpose of establishing the buckling stabil-
buckling parameter calculated by numerical methods:
ity of the arch bridges with inclined hangers and arches
of the network arch bridges which had long been known π N Ed ψ
have not been defined as well. Dutch researchers have also β= , (1)
s EI y
highlighted the limited possibilities for the application of
Figure D.4 (Romeijn & Bouras, 2008). In order to extend where ψ is the buckling parameter received from the
to the application boundaries of the mentioned Eurocode FEM program analysis results, NEd is the axial compres-
Figure a numerical experiment was conducted for a mod- sion force acting in the arch, s is the half length the arch,
el of arch bridges with pre-tensioned hangers and a rigid EIy is the arch bending stiffness where the arch in-plane
arch and stiffening girder connection. buckling takes place.
The article presents a comparative analysis of the buck- While interpreting the buckling length factor curve,
ling length factor of the steel arch bridges with vertical when the arch is connected with the chord by a single
hangers and network arch bridges with the use of numeri- hanger, it may be assumed that Figure D.4 is based on
cal methods. The main assumptions are established which the first graph of Table D.4: a two-hinged parabolic arch.
Engineering Structures and Technologies, 2019, 11(1): 11–16 13

a)
? 1.2

m 1.0

f
m=1

Buckling length factor, b

0.8
m=3
L m=5
0.6
m=7
m = 11
0.4
b)
0.2
m

f
0
0.000468 1 2135
EIA/EIS
L
Figure 2. The comparative analysis of the numerical and
analytical values of the buckling length factor
c)
are valid in the case of the tied bridge and the bending
m
f
stiffness of the stiffening girder is close to zero compared
to the arch bending stiffness. The difference in numerical
analysis results reaches up to 5.19%.
L
When the arch is connected with a stiffening girder
Figure 1. (a) a two-hinged single arch connected
by a single hanger, the stability behaviour of the arch is
with the stiffening girder by hangers; (b) a tied arch bridge; close to the stability of the single arches: an insignificant
(c) an arch bridge hanger support effect is obtained, while most of the total
load is taken over by the arch. The calculated buckling
Therefore, for the purpose of comparative analysis three length factor values are close to the values presented in
different static schemes of two-hinged arch bridges were Figure D.4, regardless of the static scheme of the bridge
modelled: a single two-hinged arch connected by vertical (difference up to 0.65%).
As the number of hangers m increases, the difference
hangers with a stiffening girder (Figure 1a), a tied arch
between the theoretical and the numerically calculated
bridge (Figure 1b), an arch bridge (Figure 1c). The num-
buckling length factor values grows depending on the
ber of vertical hangers m is analogous to that of the Euro-
static scheme of the bridge. In case of 11 hangers on the
code Figure D.4: m = 1; 2; 3; 5; 11.
arch bridge the difference is between 30% (bending stiff-
Since the bending stiffness ratio between the arch and
ness ratio of the arch and stiffening girder equals 1) and
the stiffening girder is not defined, marginal ratios are se-
170% (bending stiffness ratio of the arch and stiffening
lected. The bending stiffness ratio of the first (Figure 1a)
girder equals 0.0005).
and third (Figure 1c) scheme arch and stiffening girder is
The increase in the bending stiffness of the stiffening
equal to 2135.
girder reduces the part of the load attributed to the arch.
The ratio of the second scheme (Figure 1b) is respec- Correspondingly, the buckling length factor of the arch
tively equal to 2135, when the arch is rigid and the girder decreases as well, while the critical force increases. The
is slender; 1, when the bending stiffness ratios of the arch numerical experiment has established that the reserve of
and stiffening girder are equal; 0, when the arch is slender value of the buckling length factor given in Figure D.4
and the stiffening girder is rigid. of the Eurocode is 30−152%. In the Figures 3–6 are il-
Arch rise height and span length ratio f/L = 0.2. The lustrated bending moments and axial forces in the arch
bridge load is uniformly distributed along the stiffening and girder.
girder. Deck of the bridge is evaluated for calculation of For the further calculations a tied arch bridge scheme
the buckling length. Arch and girder section – IPE and is used with a bending stiffness ratio of the arch and stiff-
HEB. For each arch rise height and span length ratio f/L ening girder equal to 2135
are selected different size of the cross section.
Table 1 presents the calculation results for each arch
2. The comparative analysis of the buckling
model: the values of the buckling length factor from
length factor of the network arch bridges
Figure D.4 of the Eurocode and their respective critical
forces, and the values of the β factor obtained by means This chapter also deals with a numerical experiment
of the numerical experiment and their respective critical carried out with the use of the finite elements software
forces. The right column of the table presents the calcula- “SOFiSTiK” to determine the stability assessment possibil-
tion errors. ities of the network arch bridges with the use of the curves
The results of the comparative analysis presented in presented in Figure D.4 of the Eurocode 1993-2. The
Table 1 and Figure 2 show that the buckling length factor network arch bridges are characterized by a larger num-
β curves given in Annex D, Figure D.4 of the Eurocode ber of hangers compared to the traditional arch bridges.
14 S. Žilėnaitė. Comparative analysis of the buckling factor of the steel arch bridges

Table 1. The comparative table of the numerical and analytical values of the buckling length factor

SOFiSTiK EC 3-2 %
Scheme No. EIA/EIS
Ncr (MN) β Ncr (MN) β
A 2135 3.53 0.962 3.57 0.65
2135 3.53 0.961 3.57 0.52
B 1 3.54 0.961 3.57 0.956 0.56
0.0005 0.00153 0.998 0.00167 4.40
C 2135 3.53 0.962 3.57 0.63
A 2135 3.81 0.925 6.46 30.10
2135 6.27 0.722 6.46 1.47
B 1 7.88 0.644 6.46 0.711 10.47
0.0005 0.00495 0.556 0.00302 27.93
C 2135 3.82 0.925 6.46 30.02
A 2135 3.93 0.911 9.07 51.85
2135 8.82 0.608 9.07 1.39
B 1 11.48 0.533 9.07 0.6 12.51
0.0005 0.00638 0.490 0.00425 22.53
C 2135 3.95 0.910 9.07 51.59
A 2135 4.02 0.901 12.5 76.34
2135 12.19 0.517 12.5 1.25
B 1 17.84 0.428 12.5 0.511 19.47
0.0005 0.014 0.330 0.00585 54.71
C 2135 4.04 0.899 12.5 75.87
A 2135 4.11 0.891 15.5 94.11
2135 17.15 0.436 15.5 5.19
B 1 26.3 0.352 15.50 0.459 30.26
0.0005 0.0512 0.173 0.00726 165.51
C 2135 4.14 0.888 15.5 93.97

Figure 3. Axial forces and bending moments of a tied arch bridge

Figure 4. Axial forces and bending moments of an arch bridge

Figure 5. Axial forces and bending moments of two-hinged single arch connected with the stiffening girder by hangers
Engineering Structures and Technologies, 2019, 11(1): 11–16 15

Moreover, the bridges of this type are much more rigid in The provided results show that as the number of
plane. Only the application for arches with vertical hang- hangers increases, the value of the buckling length fac-
ers is graphically defined in Figure D.4 of the Eurocode. tor decreases and the critical force increases accordingly.
Therefore, while designing steel arch bridges with criss- The difference is up to 1.6 times or 61.05%. The value of
cross or diagonal hanger arrangement according to the the β factor of the arches with fourteen vertical hangers
buckling length factor values β presented in the Eurocode, is approximately 19.84% less than the standard one on
the reserve can reach several times that amount. Figure D.4 of the Eurocode 1993-23. Therefore, the ap-
For the purpose of comparative analysis, five tied net- plication area of the Eurocode is very small in the design
work arch bridges are modelled (Figure 3). The original of network arch bridges. The presented results show that
angles between intersecting hangers are selected which are Figure D.4 must be supplemented not only by the number
α = 30°, 40°, 45° and the respective number of hangers: of hangers but also by angle α which characterizes the in-
m = 14, 20, 28 and 40. The geometry of the grid of the net- tersection of the hangers.
work arch bridges is circular (Backer et al., 2014; Brunn &
Schanack, 2003): the hangers intersecting with the arch at Conclusions
constant angle divide it into equal parts. Thus, the vertical
loads are moved as far as possible in the radial direction The article provides a comparative analysis of the buckling
and a local curvature of the arch radius occurs, while all length factor. The values presented as curves in Eurocode
of this results in small bending moments in the arch and Annex D Figure D.4 and calculated by means of finite el-
buckling resistance. ements software “SOFiSTiK” were compared. The aim of
Similarly to the first part of the analysis for each model the first part of the analysis is to determine the assump-
of the arch the values of axial forces and buckling parame- tions for arch bridge modelling under which the depen-
ters are obtained in a numerical way. The product of these dence of the ratio between the buckling factor and the
two is equal to the critical force of the arch. The value of height of the rise as well as the span length is valid. The
the buckling length factor is calculated according to for- numerical experiment result error is up to 5.20% in case
mula no. 1 discussed in the first part of the article. The of a tied bridge and the ratio of the bending stiffness of
buckling length factor values β calculated numerically are the stiffening girder relatively equal to zero. The increase
compared with the buckling length factor value selected of the bending stiffness of the stiffening girder causes the
in Figure D.4 of the Eurocode 1993-2. It is assumed that buckling length factor values to be 30−152% greater than
βEC = 0.459, while the number of hangers equals 11. Arch the normative βEC value.
rise height and span length ratio f/L = 0.2. The bridge load In the second part of the comparative analysis the
is uniformly distributed along the stiffening girder. Buck- buckling length factor value was established by a nu-
ling length factor difference is expressed as a percentage. merical experiment with the increase in the number of
The calculation results are presented in Table 2. hangers and changing the hanger network form. It was
The first part of Table 2 presents the constant buckling established that as the number of hangers increases, the β
length factor value obtained from Figure D.4 of the Euro- factor decreases (from 14.20% up to 61.05% compared to
code 1993-2 while the number of hangers equals 11. The the normative βEC, m = 11), and the critical force increases
second part of the table shows the buckling length factor accordingly.
values obtained and calculated numerically. The third part In conclusion, the application area of Figure D.4 of
of the table presents the difference in percentage form. Annex D, Eurocode 1993-2 is limited. In order to increase
the economic efficiency and design accuracy, it is recom-
mended that the dependence of the buckling length factor
a values should be supplemented by a bending stiffness ratio
f

between the arch and the stiffening girder, a number of

hangers greater than 11 and a new form of hanger grid.
L

Figure 6. Network arch bridge

Table 2. The comparative table of the numerical and analytical values of the buckling length factor

EC3-2
0.459 0.459 0.459 0.459 0.459
Network arch bridge, FE
m = 14, α = 45° m = 28, α = 45° m = 28, α = 30° m = 20, α = 40° m = 40, α = 40°
0.383 0.338 0.285 0.348 0.297
Difference, %
19.84 35.80 61.05 31.90 54.55
16 S. Žilėnaitė. Comparative analysis of the buckling factor of the steel arch bridges

References Pi, Y.-L., & Bradford, M. A., & Tin-Loi, F. (2008). In-plane
strength of steel arches. Advanced Steel Construction, 4(4),
Backer, H. D., Outtier, A., De Pauw, B., & Van Bogaert, Ph. 306-322. https://doi.org/10.18057/IJASC.2008.4.4.3
(2010). Long-term monitoring of temperatures in steel box gird- Romeijn, A., & Bouras, C. (2008). Investigation of the arch in-
ers. Paper presented at the 34th International Symposium on plane buckling behaviour in arch bridges. Journal of Con-
Bridge and Structural Engineering (IABSE − 2010) (pp. 1-8). structional Steel Research, 64, 1349-1356.
Venice, Italy. https://doi.org/10.1016/j.jcsr.2008.01.035
Backer, H. D., Outtier, A., & Bogaert, Ph. V. (2014). Buckling de- Schanack, F. (2009). Berechnung der Knicklast in Bogenebene
sign of steel tied-arch bridges. Journal of Constructional Steel von Netzwerkbögen. Stahlbau, 86(5), 249-255.
Research, 79, 159-167. https://doi.org/10.1002/bate.200910022
https://doi.org/10.1016/j.jcsr.2014.09.004 Teich, S. (2011). Entwicklung allgemeiner Entwurfsgrundsät-
Brunn, B., & Schanack, F. (2003, August). Calculation of a dou- ze für Hängernetze von Netzwerkbogenbrücken. Stahlbau,
ble track railway network arch bridge applying the European 80(2), 100-111. https://doi.org/10.1002/stab.201001395
standards. Technische Universität Dresden, Grimstad. Tveit, P. (2005, August). The network arch, Bits of Manuscript
Brunn, B., Schanack, F., & Steimann, U. (2004). Network arches after lectures in 34 countries. Retrieved from https://home.uia.
for railway bridges. In P. Roca & C. Molins (Eds.), Advances in no/pert
assessment, structural design and construction. Paper present- Zhao, S.-Y., Guo, Y.-L., & Dou, C. (2013). Geometric imperfec-
ed at the Arch bridges IV, International Center for Numerical tion effects on out-of-plane inelastic buckling loads of lateral
Methods in Engineering (pp. 1-9). Barcelona. braced arches. In Proceedings of the 10th Pacific Structural
Galambos, T. V. (Ed.). (1988). Guide to stability design criteria for Steel Conference, 8-11 October, Singapore (pp. 181-186). Sin-
metal structures (4th ed., pp. 669-703). New York, NY, USA: gapore: Research Publishing Services.
John Wiley & Sons. https://doi.org/10.3850/978-981-07-7137-9_072
Graβe, W., Teich, S., Tveit, P., & Wendelin, S. (2004). Network Ziemian, R. D. (2010). Guide to stability design criteria for metal
arches for road bridges. In Advances in assessment, structural structures (6th ed., pp. 762-806). New York, NY, USA: John
design and construction. Paper presented at the Arch bridges Wiley & Sons. https://doi.org/10.1002/9780470549087.ch17
IV, International Center for Numerical Methods in Engineer-
ing (pp. 1-10). Barcelona.
Guo, Z., Wang, Y., Lu, N., Zhang, H., & Zhu, F. (2016). Behaviour
of a two-pinned steel arch at elevated temperatures. Thin-
Walled Structures, 107, 248-256.
https://doi.org/10.1016/j.tws.2016.06.015
Heidarpour, A., Bradford, M. A., & Othman, K. A. M. (2011).
Thermoelastic flexural–torsional buckling of steel arches.
Journal of Constructional Steel Research, 67, 1806-1820.
https://doi.org/10.1016/j.jcsr.2011.05.005
Larssen, R. M., & Jakobsen, S. E. (2011). Brandangersundet
bridge – a slender and light network arch, taller, longer, ligher.
Paper presented at the IABSE-IASS-2011 London Symposium
Report, 20–23 September, London, United Kingdom.
Lebet, J.-P., & Hirt, M. (2013). Steel bridges: conceptual and
structural design of steel and steel-concrete composite bridges
(pp. 461-488). New York, NY, USA: Taylor and Francis Group.
Lietuvos Standartizacijos departamentas. (2007). Eurokodas 3.
Plieninių konstrukcijų projektavimas. 2 dalis. Plieniniai tiltai
(LST EN 1993-2:2007/NA:2010). Retrieved from https://lsd.
lt/index.php?1323599487
Pi, Y.-L., & Trahair, N. S. (1999). In-plane buckling and design of
steel arches. Journal of Structural Engineering, ASCE, 125(11),
1291-1298.
https://doi.org/10.1061/(ASCE)0733-9445(1999)125:11(1291)
Pi, Y.-L., & Bradford, M. A. (2004). In-plane strength and design
of fixed steel I-section arches. Engineering Structures, 26(3),
291-301. https://doi.org/10.1016/j.engstruct.2003.09.011
Pi, Y.-L., & Bradford, M. A. (2006). Elastic flexural-torsional
buckling of circular arches under uniform compression and
effects of load height. Journal of Mechanics of Materials and
Structures, 1(7), 1235-1255.
https://doi.org/10.2140/jomms.2006.1.1235
Pi, Y.-L., & Bradford, M. A. (2014). Effects of nonlinearity and
temperature field on in-plane behavior and buckling of
crown-pinned steel arches. Engineering Structures, 74, 1-12.
https://doi.org/10.1016/j.engstruct.2014.05.006