EFFECT OF LATERAL TORSIONAL BUCKLING ON

WEB TAPERED I-BEAMS

Nilesh B Salukhe1, Suhasini N Madhekar2

1. Research Scholar, Civil Engineering Department, College of Engineering Pune,

Maharashtra
2. Applied Mechanics, College of Engineering Pune, Maharashtra

ABSTRACT. Lateral torsional buckling is considered as an ultimate limit state related to

member buckling resistance. The buckling resistance is obtained by using a reduction factor.
This reduction factor is the function of two other parameters, viz. the imperfection factor and
the non-dimensional slenderness. The imperfection factor related parameter takes into
account the initial member imperfection, residual stresses and other nonlinear effects. The
non-dimensional slenderness depends upon the elastic critical moment (Mcr) for lateral
torsional buckling. Mcr can be calculated as per IS 800 (2007) (Annex-E, Clause 8.2.2.1). A
comparative study between the values of Mcr and Md (design bending strength) considering
IS 800 (2007) code and AISC (1989) code is performed in the present paper. It indicates that
the values of Mcr and Md as per IS code are on the conservative side, as compared to the
AISC code. It is therefore observed that there is a need for development of design formulae
for tapered structural members in IS 800 (2007), considering the effect of web tapering ratio
on the lateral torsional buckling of web tapered I-beams. Hence, it is essential to find out a
factor in the Mcr equation, which will consider the interactions of the above parameters. The
influence of the taper ratio on the critical load of web-tapered I-beams is seen to be very
significant and must be taken into account when designing such members against buckling.

Keywords: Buckling, Lateral Torsional Buckling, Taper Ratio, Web-tapered Beams.

Nilesh B Salunkhe is a Research Scholarin Department ofCivil Engineering, College of

Engineering Pune, Maharashtra. His research interests include Design of steel structures,
Structural Engineering and computational work on software.

Dr Suhasini Madhekar is Associate professor, Applied Mechanics at College of

Engineering Pune. Her research interests include Structural Dynamics, Vibration Control,
Bridge Engineering and protection of structures from earthquake. She has published a book
on “Seismic design of RC buildings - Theory and Practice”, with Springer. The book received
award from Association of Consulting Civil Engineers, India, the ACCE (I) - Nagadi award
as ‘Best Publication of the year 2017- useful to structural consultants.
 INTRODUCTION
The flexural capacity of beams with large unbraced lengths i.e. unrestrained beams is often
limited by the mode of failure, known as Lateral Torsional Buckling (LTB). A beam is
considered to be unrestrained when its compression flange is free to displace laterally and
rotate. When an applied load causes both - lateral displacement and twisting of a member,
lateral torsional buckling will occur. Figure 1 shows the lateral displacement and twisting
experienced by a beam when LTB occurs. The applied vertical load results in compression
and tension in the section flanges. The compression flange tries to deflect laterally away from
its original position; whereas the tension flange tries to keep the member straight. The lateral
bending of the section creates restoring forces that oppose the movement, because the
tendency of the section is to remain straight. These restoring forces are not large enough to
prevent the section from deflecting laterally, but together with the lateral component of the
tensile forces, they determine the buckling resistance of the beam.

Figure 1 Lateral Displacement and Twisting (NSC 2006) [3]

In addition to the lateral movement of the section, the forces within the flanges cause the
section to twist about its longitudinal axis. The twisting is resisted by the torsional stiffness of
the section, which is dominated by the flange thickness. That is why a section with thicker
flanges has a larger bending strength than that of the same depth section, with thinner flanges.
LTB can be avoided by properly spaced and designed lateral bracings. The other factors
affecting LTB are the proportions of beam cross sectional dimensions, material properties
such as modulus of elasticity and shear modulus, length of the beam, section slenderness,
support conditions, initial geometry imperfections, and the type and application of loading.

The effect of a destabilizing load is considered by the use of effective length given in Table
15 of IS 800 (2007), where the effective lengths are longer for destabilizing loads, as
compared to those of the non-destabilizing loads. Factors C1, C2 and C3 are included to allow
for the effect of different bending moment distributions and end restrained conditions.

In practice, beams are laterally braced in a variety of ways, in order to increase their buckling
strength. Determining the brace force requirements for a system generally requires a large
displacement analysis of an imperfect system. Most bracing studies usually focus on
determining the maximum brace forces that are likely to occur in typical applications. This is
particularly true for beam bracing in which the brace location and distribution of the loading
can have a significant effect on the brace forces (Gill and Yura 1999) [4].

Over the past three decades, construction of buildings with frames comprising of web-tapered
I-beams, manufactured from high tensile steel has become a standard practice. Their cross-
sectional profiles are intended to match the flexural strength close to the bending moment
diagram, so that the requirement of the cross sections is well optimized. Despite several
advantages of tapered structural members, they lack the appropriate simple and accurate
design formulae in most of the codes of practice. The design solutions for tapered structural
members are limited, because the available approaches consist of elastic design formulae,
where taper effects are not properly accounted for. The stability of loaded tapered rafters was
investigated by earlier researchers with proposals to design tapered rafters as uniform
prismatic members, using additional factors.

Mcr can be calculated as per IS 800 (2007) Annex-E, Clause 8.2.2.1. Here, C1, C2 and C3
factors depend upon the loading and end restrained conditions. When the loading is not due to
a single central point load or due to full-length uniformly distributed load (udl), the published
values IS 800 for C1 and C3 may be inaccurate and in some cases non-conservative. In
addition, the C3 factor is only required for asymmetric sections. A comparative study as
shown in Table 1 between the values of Mcr and Md considering IS 800 (2007) code and
AISC code indicates that the values of Mcr and Md as per IS code are on the conservative side
as compared to the AISC code.

It is therefore observed that there is a strong need for the development of design formulae in
IS 800 (2007), for tapered structural members, considering the effect of geometric
imperfections, effect of transverse loading applied at different heights with respect to the
mid-height of the cross section and; effect of web tapering ratio, on the LTB of web tapered
beams. Hence, it is essential to find out a factor in the Mcr equation, which will consider the
interactions of these parameters. With the inclusion of this factor, it will be ensured that the
material is used to its fullest capacity by optimizing the steel quantity, considering all the
influencing parameters and thus reducing the steel consumption and saving the natural
resources for future.

REVIEW OF BEAM DESIGN AS PER IS 800 (2007) SPECIFICATIONS

Lateral torsional buckling is considered as an ultimate limit state related to the member
buckling resistance. The buckling resistance is obtained by multiplying the resistance of the
cross section by a reduction factor . This reduction factor is the function of two other
parameters; the imperfection factor αLT and the non-dimensional slenderness LT. The
parameter αLT takes into account the initial member imperfection, residual stresses and other
nonlinear effects. The non-dimensional slenderness 𝜆LT depends upon the elastic critical
moments for lateral torsional buckling.
The elastic critical moment may be calculated from Equation (1) of Eurocode 3, derived from
the buckling theory:

Mcr =C1 - }
(1)
 Table 1 Values of Mcr and Md using IS 800 (2007) and AISC code

IS 800 (2007) IS 800(2007)

Member AISC-LRFD
Member Section Clause 8.2.2.1 ANNEX-E
Size
Description type Mcr Md Mcr Md Section Mcr Md
(mm)
(kNm) (kNm) (kNm) (kNm) type (kNm) (kNm)
Web
152.4 × 2.54 Plastic Compact
(dw × tw)
6.64 4.98 6.36 4.80 6.64 5.97
Flange
101.6 × 6.35 Compact Compact
(bf × tf)
Web 152.4 ×
Plastic Compact
(dw × tw) 2.54
81.68 32.57 63.06 29.04 49.17 44.25
Flange
304.8 × 6.35 Slender Slender
(bf × tf)
Web
152.4 × 6.35 Plastic Compact
(dw × tw)
56.24 36.23 55.92 36.07 56.27 50.64
Flange 101.6 ×
Plastic Compact
(bf × tf) 19.05
Web
152.4 × 2.54 Plastic Compact
(dw × tw)
531.86 207.93 504.73 203.69 308.27 277.44
Flange 304.8 ×
Compact Compact
(bf × tf) 19.05
Web Non
304.8 × 2.54 Slender
(dw × tw) compact
20.43 15.17 15.98 12.17 20.43 18.38
Flange Non
152.4 × 6.35 Slender
(bf × tf) compact

In Equation (1), L is the beam length between points which have lateral restraint. For simply
supported beams with intermediate lateral restraint, the effective length LLT, as per the code is
1.2 times the length of the relevant segment in between the lateral restraints. This is how the
code takes into account the destabilizing effect of the top flange loading. yg is the distance
between the points of load application and the shear centre.

When a beam buckles and twists, the displacement of the flanges, combined with the
longitudinal bending stresses (compression and tension) set up a torque. The term yj
represents the coordinate of the centre about which the two torques rotate. This leads to yj
equal to zero for symmetric sections. For asymmetric sections yj will be positive when the
larger flange is in compression and negative when the smaller flange is in compression. K
and Kw are effective length factors of the unsupported length accounting for boundary
conditions at the end lateral supports. The effective length factors K and K w vary from 0.5 for
full fixity (against warping) to 1.0 for free (to warp) case and; 0.7 for the case of one end
fixed and the other end free.

C1 is equivalent uniform moment factor and; C2 and C3 are the factors which depend upon the
loading and end restrained conditions. The values of C1, C2 and C3 are given in the Table 42
of IS 800 (2007). The factors C1 and C3 are little trickier, as they are not deterministic and
most of them are quoted as values, formulae or graphs. When the loading is not due to a
single central point load or due to a full length udl, the published values of IS 800 (2007) for
C1 and C3 can be inaccurate and in some cases non-conservative. Also the C3 factor is
required only for asymmetric sections.

NUMERICAL SIMULATIONS
In this paper, the effect of taper ratio and cross-sectional geometry on the stability of steel
members that are subjected to bending is investigated. The effect of shear forces is
considered negligible. The problem is studied by focusing on web-tapered I-beams with built-
up cross-sections that are usually required in majority of the steel structures. The beams are
considered simply supported in bending, while other boundary conditions can be easily dealt,
with the approach proposed herein

Nonlinear Finite Element Modeling

To investigate the lateral torsional buckling of beams, nonlinear finite element analysis is
performed using the commercial software package, ANSYS R18.

Material Properties

The properties of steel considered are: Young’s modulus (E) = 2×105 MPa and Poisson’s
ratio (υ) = 0.3.

Geometric Nonlinearities

Due to large deformations, the changing geometric configuration can cause the structure to
respond nonlinearly. Hence geometric nonlinearity is considered in the current scenario.

Finite Element Type and Mesh

A four-node structural shell element (SHELL181) from ANSYS R18 library has been used in
the nonlinear finite element analysis, to investigate the lateral torsional buckling. The element
has six degrees of freedom at each node; 3 translations in the x, y and z directions and 3
rotations about the x, y and z axis. The element is suitable for large rotation, large strain
nonlinear application and load stiffness effect of distributed pressures. The performance of
nonlinear buckling finite element convergence shows that the element mesh size of 25 mm
(shown in Figure 2) is suitable and sufficient to apply in this investigation.

Figure 2 Shell Element Mesh Pattern

Boundary Conditions and Load Application

In order to find the effect of tapering ratio of tapered I-shaped beams on the buckling
analysis, loading and restraint conditions for single span tapered I-beams is considered. Point
load is applied at the centre of the beam. Two typical symmetric tapered beams are
considered in the study. The cross-sectional dimensions at the small end are given in Table 2.
The two sections are identical to the sections originally used by Lee et al. (1972) [5] and M.
L. Morrell & Lee (1974) [6]. The beam end conditions considered are both ends pinned.

Table 2 Section Dimensions

Dimension
Section- I Section- II
(mm)
d0 152.4 152.4
b 101.6 101.6
tf 6.35 19.05
tw 2.54 6.35

Four different lengths of beam viz. 1.735 m, 2.275 m, 2.850 m and 3.415 m are considered in
each case. Since the main objective of this investigation is to study the lateral torsional
buckling of tapered beams, the cross sections are selected to have height to thickness ratio
sufficient, to prevent the local buckling. The taper ratio γ = [(dL-do)/do], (where dL and d0 are
web depth at the large and small end of taper beam respectively) is varied as 0.0, 0.75, 1.5
and 3.0 which covers majority of the practical situations. These results are compared with the
Mcr equation of IS 800 (2007), as per clause 8.2.2 and also as per Annex E.

RESULTS AND DISCUSSION

(a) With reference to Figure 3, Figure 4 and Figure 5 for Section-I, it is observed that as
the taper ratio increases from 0.0 to 3.0, there is drop in the nonlinear moment
capacity of the tapered beam up to 65%. Whereas, as per IS 800 (2007), it is observed
that as the taper ratio increases from 0.0 to 3.0 there is increase in the critical moment
capacity of the beam by 80% (as per clause Annex E) and 100% (as per clause 8.2.2).

(b) With reference to Figure 6, Figure 7 and Figure 8 for Section-II, it is observed that as
the taper ratio increases from 0.00 to 3.00, there is a drop in the nonlinear moment
capacity of the tapered beam up to 42%. Whereas, as per IS 800 (2007), it is observed
that as the taper ratio increases from 0.00 to 3.00 there is increase in the critical
moment capacity of the beam by 14% as per clause Annex E and 39% as per clause
8.2.2

(c) The drop in the nonlinear moment capacity is more dominant in Section type I as
against of Section type II. That is as for section having thinner webs, there is more
drop as compared to the thicker web. As the length of the beam reduces, the drop in
 the nonlinear moment capacity increases in both the type of sections i.e. Section-I and
Section II. However, there is increase in the beam capacity as per IS800 (2007).

Figure 3 Nonlinear Moment Ratio (Mnl/Mnl0) vs. Taper Ratio γ for Various Lengths of
Tapered Beams with Concentrated Load applied at the Center of Top Flange (Section- I)

Figure 4 Elastic Critical Moment Ratio (Mcr’/Mcr’0) (As Per IS800 (2007) Clause 8.2.2)
vs. Taper Ratio γ for Various Lengths of Tapered Beams with Concentrated Load applied
at the Center of Top Flange (Section- I)
Figure 5 Elastic Critical Moment Ratio (Mcre/Mcre’0) (As Per IS 800 (2007) Annex-E) vs.
Taper Ratio γ for Various Lengths of Tapered Beams with Concentrated Load Applied at
the Center of Top Flange (Section- I)

Figure 6 Nonlinear Moment Ratio (Mnl/Mnl0) vs. Taper Ratio γ for Various Lengths of
Tapered Beams with Concentrated Load Applied at the Center of Top Flange (Section- II)
 Figure 7 Elastic Critical Moment Ratio (Mcr’/Mcr’0) (As Per IS800 (2007) Clause 8.2.2)
vs. Taper Ratio γ for Various Lengths of Tapered Beams with Concentrated Load Applied
at the Center of Top Flange (Section- II)

Figure 8 Elastic Critical Moment Ratio (Mcre/Mcre’0) (As Per IS 800 (2007) Annex-E) vs.
Taper Ratio γ for Various Lengths of Tapered Beams with Concentrated Load applied at
the Center of Top Flange (Section- II)
 CONCLUSIONS
From the numerical study performed, it is observed that as the taper ratio increases from 0.00
(uniform beam) to 3.00 (tapered beam), there is a drop-in value of the critical load and critical
moment. The influence of the taper ratio on the critical load of web-tapered I-beams is proven
to be very significant and must be taken into account, when designing such members against
buckling. Thus, effect of web tapering ratio on the LTB of web tapered beams needs
consideration and hence, it is essential to find out a factor in the Mcr equation, which will
consider the interactions of these parameters. With the inclusion of this factor, it will be
ensured that the material is used to its fullest capacity by optimizing the steel quantity,
considering all the influencing parameters and thus reducing the steel consumption and
saving the natural resources for future.

REFERENCES
1. INDIAN STANDARD GENERAL CONSTRUCTION IN STEEL-CODE OF
PRACTICE, IS 800, 3rd Edition, Bureau of Indian Standards, 2007.

2. MANUAL OF STEEL CONSTRUCTION-ALLOWABLE STRESS DESIGN, AISC, 9th

Edition, Chicago, IL, 1989,

3. NSC, Lateral torsional buckling and slenderness, 2006, pp 30-34.

4. GIL, H., and YURA, J. A., Bracing requirements of inelastic columns, Journal of
Constructional Steel Research, 51, 1999, pp 1-19.

5. LEE, G. C., MORRELL, M. L., and KETTER, R. L., Design of tapered members, WRC
Bulletin No. 173, Welding Research Council, New York, USA, 1972.

6. MORRELL, M. L., and LEE, G. C., Allowable stress for Web-Tapered beams with lateral
restraints, WRC Bulletin No. 192, Welding Research Council, New York, USA, 1974.