Sudha.K et.

al / International Journal of Engineering and Technology (IJET)

Behaviour of Cold-Formed Steel Built-up I

Section Under Bending
Sudha.K #1, Sukumar. S#2
#
Department of Civil Engineering, Government College of Engineering, Salem, Tamil Nadu, India
1
[email protected]
2
[email protected]
Abstract - This paper presents an experimental and numerical investigation on the bending strength and
behaviour of cold-formed (CF) steel built-up flexural members. Eight specimens in two groups, first
group of four specimens with equal flanges and second group of four specimens with unequal flanges
have been fabricated and experimented. The experimental results show the modes of buckling and their
influence on the bending strength and behaviour of CF built-up I sections. The experimental results are
also verified by simulating finite element models and analysed using FEM software ANSYS. The results
obtained are in good agreement with the experimental results.
Keywords - Cold-formed Built-up section, Thin walled beam, Local buckling, Flexural behaviour,
Distortional buckling, Crippling
I. INTRODUCTION
Cold formed steel products are extremely used in building industry and range from purlins to roof sheeting and
floor decking. These sections are extensively used in various engineering applications because of their high
strength to weight ratio. Thin walled I sections are preferred mostly due to their light weight and full strength
capacity.
Vlassov[1] presented the classical theory in this field excellently . Reviewing the present state of art
one should mention the books presented by Bleich[2],Timoshenko[3], and Murray[4], in addition to a large
number of papers based on linear analysis. Chajes[5], Yu[6], Bulson[7], Hancock[9], Pekoz [10,11] presented
many papers on the stability of open cross-sections with solid webs. Based on the equilibrium at the deflected
position of a structure Ganapathy Chettiar and Varghese[8] have developed a method to find the lateral buckling
load of open-web frames and girders of any geometry and any support conditions. An experimental study on I-
beams with slender sections has been carried out by Kubo and Fukumoto[12] and presented the interaction
behaviour of local and lateral-torsional buckling in the inelastic buckling range. They also found that the beams
with intermediate length undergo combined local and lateral-torsional buckling whereas the long beam fails
only by lateral-torsional buckling.
Put, Pi and Trahair[13] conducted lateral buckling strength tests on C and Z beams and summarized that local or
distortional buckling occurs in the compression element. Yu and Schafer[15] presented the testing plan to
restrict distortional buckling and a simple method to generate local buckling in C and Z beams. Yu and
Schafer[18] presented another method to generate distortional buckling in cold formed steel C and Z beam and
they also found that large strength reduction occurs as the distortional buckling is initiated instead of local
buckling. In case of latticed steel members, as a bar with circular cross-section is very stiff torsionally and
flexurally and as the full section of the bar is continuous at the joints where they are also welded, the system of
web bars provides a high torsional and flexural stiffness at joints. In this, the load transmission relies solely on
the weld connection, resulting in a discontinuous load-transfer mechanism. Heavy stress concentration is
induced at the welded locations.
Though many studies have been performed on buckling of cold formed steel profiles, few studies have been
made on latticed / built-up cold-formed beams and remain largely unexplored.
The literature on built – up CF Sections are very minimum and are available for compression
members[14,16,17]. Whereas the study on cold formed steel built up sections under bending is very rare. Hence
in this investigation an attempt has been made to study the behaviour of CFS built – up flexural members and to
find the feasibility of such sections in the field of construction. Thin walled built-up I section beams are
economical with same flexural rigidity when compared to thin walled solid web sections. Though this type of
sections results in economic type of construction, presently the usage of such members are limited because of
the complexity involved in the analysis and prediction of their behaviour become tedious. As this type of
sections with open cross section are susceptible to multiple modes of buckling when they are connected together
by a system of continuous bent bar with circular cross sections, the behaviour of the beam elements may be
governed by local buckling in the component plates, distortional buckling of the outstanding legs and the local
buckling of web element.

ISSN : 0975-4024 Vol 5 No 6 Dec 2013-Jan 2014 4622

 Sudha.K et.al / International Journal of Engineering and Technology (IJET)

The Research activities presented in this current paper are focused on flexural behaviour of cold-formed
steel latticed members under static lateral loading is a part of the ongoing research activity. This paper provides
the outcome of the first experimental phase of the research devoted to study the behaviour of the latticed cold-
formed steel beams.
II. TEST SPECIMEN
Two types of built up CFS I Sections with equal and unequal flanges have been fabricated and
experimented. The specimens are fabricated from 1.2mm thick steel sheets whose yield stress is 230 N/mm2 and
young’s modulus of 1.98 x 105 N/mm2. The specimens are built-up using four numbers of Cold Formed angles
with lip in the flange alone. Out of four angles two are provided in the top flange and two at the bottom flange
which act as the main chords of the member. These main chords are held in position by using a secondary
member of 6mm diameter bar bent in the form of sinusoidal wave form of angle 45°. The main chords and the
secondary members are welded at the point of contact by spot welding. The cross section of the specimen is
shown in the Fig.1 and the geometric properties of the section are tabulated in the Table 1. The series are
designated as UEI – top width – thickness for unequal I sections and EI – top width – thickness for equal I
sections. All the eight specimens were tested under two point loading. UEI – 40 – 1.2 is an unequal I section
fabricated without stiffener under loading points so as to study the behaviour in bearing. EI – Equal flange I
sections were fabricated with 2mm thick stiffener plates to investigate the influence of bearing stiffeners. Other
specimens were fabricated to study the local and distortional buckling of beams by strengthening the stiffeners
against bearing failure. Two stiffener plates one on either side is provided at the points of loading and at the
points of support.

Fig. 1 Cross Section of the Specimen

Table I Geometric Properties of the Sections

S. Description (A) (B) (C) (L) (D) (ld) (tk )

No of Specimen mm mm mm mm mm mm mm
1. UEI – 40 – 1. 2 40 20 30 15 90 6 1.2

2. UEI – 40 – 2. 0 40 20 30 15 90 6 2.0

3. UEI – 35 – 1. 2 35 20 35 15 90 6 1.2

4. UEI – 30 – 2. 0 30 20 40 15 90 6 2.0

5. EI – 40 – 1. 2 40 40 10 15 90 6 1.2

6. EI – 40 – 2. 0 40 40 10 15 90 6 2.0

7. EI – 30 – 1. 2 30 30 30 15 90 6 1.2

8. EI – 30 – 2. 0 30 30 30 15 90 6 2.0

ISSN : 0975-4024 Vol 5 No 6 Dec 2013-Jan 2014 4623

 Sudha.K et.al / International Journal of Engineering and Technology (IJET)

III. EXPERIMENTAL INVESTIGATION

The experimental work was conducted on a 50kN capacity self-straining frame. The Specimens were
experimented under simply supported end condition with two points loading. Deflectometers were placed at
three positions namely 1/3rd distance, mid span and at support. Strain readings were taken in the top
compression flange and in the lip of the compression flange. A Proving ring of 50kN capacity was used to
measure the loading. The load and corresponding deflection was observed. Loading was gradually applied till
the specimen become unstable. The procedure was repeated for all the eight specimens. The experimental set up
is shown in Fig. 2. The modes of failure are noted for every specimen. Some of the tested specimens are shown
from Fig. 3 to Fig. 6. The total critical moments, total ultimate moments and bending stress are presented in
Table II.

Fig. 2 Test Setup

Fig. 3 Distortional Buckling

Fig. 4 Wobbling of Bottom Web

ISSN : 0975-4024 Vol 5 No 6 Dec 2013-Jan 2014 4624

 Sudha.K et.al / International Journal of Engineering and Technology (IJET)

Fig. 5 Local Buckling

Fig. 6 Flexural Buckling

16000
25000

14000

20000 12000

10000
15000
Lo ad (N)
Load (N)

8000

10000 6000

4000
5000
2000

0 0
0.0000 0.0005 0.0010 0.0015 0.0020 0.000 0.001 0.001 0.002 0.002
Strain Strain

Fig. 7(a) UEI – 40 – 2.0 Fig. 7(b) UEI – 35 – 1.2

ISSN : 0975-4024 Vol 5 No 6 Dec 2013-Jan 2014 4625

 Sudha.K et.al / International Journal of Engineering and Technology (IJET)

16000 35000

14000 30000

12000
25000

10000
20000

Load (N)
Load (N)

8000
15000
6000
10000
4000

5000
2000

0 0
0.000 0.001 0.002 0.003 0.004 0.000 0.001 0.002 0.003 0.004 0.005
Strain Strain

Fig. 7(c) UEI – 30 – 2.0 Fig. 7(d) EI – 40 – 2.0

25000 16000

14000
20000
12000

10000
15000
Load (N)
Load (N)

8000

10000
6000

4000
5000
2000

0 0
0.0000 0.0002 0.0004 0.0006 0.0008 0.0010 0.0012 0.0014 0.0000 0.0005 0.0010 0.0015 0.0020 0.0025
Strain Strain

Fig. 7(e) EI – 30 – 2.0 Fig. 7(f) EI – 30 – 1.2

Fig. 7 Load Vs. Strain Curves

10.0 10.0

1.0 1.0
M/Mcr

M/Mcr

0.1 0.1

0.0
0.0
0.0 0.5 1.0 1.5 2.0 2.5
φ/φcr 0.0 0.5 φ/φcr 1.0 1.5

Fig. 8(a) UEI– 40 - 2.0 Fig. 8(b) UEI – 35 – 1.2

ISSN : 0975-4024 Vol 5 No 6 Dec 2013-Jan 2014 4626

 Sudha.K et.al / International Journal of Engineering and Technology (IJET)

10.0 10.0

1.0 1.0
M/Mcr

M/Mcr
0.1 0.1

0.0 0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.0 0.5 1.0 1.5 2.0
φ/φcr φ/φcr

Fig. 8(c) EI – 40 – 2.0 Fig. 8(d) EI – 30 – 2.0

Fig. 8 Moment Vs. Curvature Curves
Table II Experimental Results

S. Description of Critical Section Critical Ultimate Ultimate σcr Mode of

Specimen Moment kNm Modulus Stress σcr Moment kNm Stress σult Failure
No. σult
mm3 N/mm2 N/mm2

1. UEI – 40 – 1. 2 2.76 9394 146.91 3.75 199.60 1.36 B

2. UEI – 40 – 2. 0 5.25 15372 170.76 5.70 185.40 1.09 D

3. UEI – 35 – 1. 2 3.78 9312 202.96 4.20 225.50 1.11 F

4. UEI – 30 – 2. 0 3.96 15291 129.50 4.50 147.10 1.14 D

5. EI – 40 – 1. 2 3.75 13927 134.63 6.60 236.95 1.76 LS

6. EI – 40 – 2. 0 8.25 22962 179.64 9.30 202.51 1.13 F

7. EI – 30 – 1. 2 3.00 11708 128.10 3.90 166.60 1.30 L

8. EI – 30 – 2. 0 5.17 19327 133.90 6.60 170.75 1.28 L

B – Bearing Failure, D- Distortion, LS – Local Buckling of Stiffener, F- Flexural Buckling, L – Local Buckling
IV. FEM ANALYSIS USING ANSYS
An approximate idealised mid-line model with four noded shell – 143 element, for plate and 3D
element for lacing are chosen. A non – linear analysis with large deflections on has been carried out. The model
is force- loaded and the line search option is chosen for the analysis. All the sections fabricated for the
experimental work (Table 1) were modelled and analysed. Failure mode shapes for some specimens are shown
from Fig. 9 to Fig. 12 and the Critical moments and Ultimate moments are presented in Table 3.

ISSN : 0975-4024 Vol 5 No 6 Dec 2013-Jan 2014 4627

 Sudha.K et.al / International Journal of Engineering and Technology (IJET)

Fig. 9 Wobbling in Web

Fig. 10 Local Buckling

Fig. 11 Distortional Buckling

ISSN : 0975-4024 Vol 5 No 6 Dec 2013-Jan 2014 4628

 Sudha.K et.al / International Journal of Engineering and Technology (IJET)

Fig. 12 Distortional Buckling

Table III Comparison of Experimental and ANSYS results

S. Description Total Critical Moment Exp Total Ultimate Moment Exp

No. of Specimen (kNm) ANSYS (kNm) ANSYS
Exp. ANSYS Exp. ANSYS
1. UEI – 40 – 1. 2 2.76 2.82 0.98 3.75 3.56 1.05
2. UEI – 40 – 2. 0 5.25 5.40 0.97 5.70 6.60 0.86
3. UEI – 35 – 1. 2 3.78 3.60 1.05 4.20 4.50 0.93
4. UEI – 30 – 2. 0 3.96 3.75 1.06 4.50 4.80 0.94
5. EI – 40 – 1. 2 3.75 3.90 0.96 6.60 6.75 0.98
6. EI – 40 – 2. 0 8.25 8.10 1.02 9.30 9.00 1.03
7. EI – 30 – 1. 2 3.00 3.23 0.93 3.90 4.20 0.93
8. EI – 30 – 2. 0 5.17 5.4 0.96 6.60 6.75 0.98

V. RESULTS AND DISCUSSION

The failure models are mixed bearing, local, distortional and flexural. The experimental failure modes are in
good agreement with ANSYS results. Load Vs. Deflection curve shows that both exp and ansys are in good
agreement. Bearing failure occurred in UEI – 40 -1.2 under the loading points followed by local buckling of the
elements of top chord members. EI series failed by buckling of stiffeners followed by mixed local and
distortional bucking of compression chords. Wobbling effect as a sine wave is found in all the unstiffened
elements (Lip and web elements). In the compression zone, the individual top chord members were distorted.
UEI series failed by mixed local and distortional buckling of the individual chords. Equal Flange I-section
carried approximately 40% more than that of unequal flange I – section. The comparison shows that member
strength is closely related to the failure pattern. For example the members (EI– 40– 2.0, UEI–35–1.2) failed by
flexural buckling exhibited higher bending stress closer to the yield stress and ultimate stress is reached as soon
as the critical stress is reached, whereas those (EI – 30–1.2, EI–30– 2.0) failed by local buckling in the lip and
flange were unable to reach their yield stress. The members(UEI-40-2.0, UEI-30-2.0) failed by distortional
buckling exhibited very low post buckling reserve strength, whereas those failed by local buckling (EI-40-1.2)
showed high post buckling reserve strength.

ISSN : 0975-4024 Vol 5 No 6 Dec 2013-Jan 2014 4629

 Sudha.K et.al / International Journal of Engineering and Technology (IJET)

Fig. 13 Load Vs Deflection Curve ( UEI 40 - 1.2)

Fig. 14 Load Vs Deflection Curve ( UEI 35 - 1.2)

VI. CONCLUSIONS
The experimental and numerical analyses on the bending strength and behaviour of cold-formed steel built-up I
section flexural members have been investigated. The conclusions from the investigations are as follows
• It is observed that each chord member bend individually. This shows that this type of members does not
bend in a single wave as a whole as in the case of solid cross- section, but the constituent members bend
individually.
• The web element also contributing appreciably to the strength.
• Specimen failed under distortional buckling have little post buckling reserve strength whereas specimen
failed under local or bearing have more post buckling reserve strength.
• To avoid bearing failure vertical stiffeners are required at support and at the loading points.
• By stiffening the web element of each chord member the capacity of the beam is further improved.
• Local buckling, distortional buckling and interaction between local and distortional buckling were
observed. The FEA predictions are generally in good agreement with the experimental buckling modes. The
results show that the buckling mode has greater influence on the strength of the specimens.

ISSN : 0975-4024 Vol 5 No 6 Dec 2013-Jan 2014 4630

 Sudha.K et.al / International Journal of Engineering and Technology (IJET)

• The comparison between the experimental results and the FEA predictions prove that the finite element
analysis is a reliable tool to get quite accurate results in a reasonable amount of time. So the parametric
study can be performed by the FEA to investigate the behaviour of Cold Formed Steel built up sections.
REFERENCES
[1] Vlasov. V.Z, Thin Walled Elastic Beams, Jerusalem, Israel: Israel Program for Scientific Translation,1961.
[2] Bleich. F., Buckling Strength of Metal Structures, New York : McGraw-Hill Book Co., 1954.
[3] Timoshenko. S.P. and Gere, J.M., Theory of Elastic Stability, New York : McGraw-Hill Book Co.,Inc., 1961.
[4] Murray, N.W., Introduction to the theory of Thin-Walled Structures, London: Oxford University Press,1984.
[5] Alexander Chajes, M., Principles of Structural Stability Theory, New Jersey: Prentice-Hall, Inc., Englewood Cliffs, 1974.
[6] Yu, W.W., Cold-Formed Steel Design, New York : Wiley,2000.
[7] Bulson. P.S., The Stability of Flat Plates,London: Chatto and Windus,1970.
[8] Varghese P.C and Ganapathy Chettiar C. Lateral Stability of Plane Open Web Structures. Journal of Structural Engineering ASCE
1974; 100(ST11): 2223-2233.
[9] Hancock G.J. Local, Distortional and Lateral Buckling of I-Beams. Journal of Structural Engineering ASCE 1978; 104(11):1787 –
1798.
[10] Reynand L.Serette, Teomen Pekoz Distortional Buckling of Thin-Walled Beams/Panels I: Theory. Journal of Structural Engineering
ASCE 1995; 121(4):757 – 766.
[11] Reynand L.Serette, Teomen Pekoz. Distortional Buckling of Thin-Walled Beams/Panels II : Design Methods. Journal of Structural
Engineering ASCE 1995; 121(4): 767 – 776.
[12] Masahiro Kubo and Yuhshi Fukumoto. Lateral – Torsional Buckling of Thin-Walled I Beams. Journal of Structural Engineering
ASCE 1988; 114(4).
[13] Bogdan M. Put, Yong-Lin Pi and Trahair N.S. Lateral Buckling Strength of Cold-formed Channel Section Beams. Journal of
Structural Engineering, ASCE 1999; 125(11): 1277 – 1283.
[14] Sukumar. S, Parameswaran.P, Jayagopal L.S. Behaviour of built up open cross section under axial load. Journal of Structural
Engineering ASCE 2003; 30(4): 89-94.
[15] Cheng Yu and Benjamin W. Schafer Local Buckling Tests on Cold-Formed Steel Beams. Journal of Structural Engineering ASCE
2003; 129(12): 1596 – 1606.
[16] Sukumar S, Parameswaran P and Jayagopal L.S. Torsional flexural buckling of built-up open cross sections under eccentric
load. Journal of Structural Engineering ASCE 2004; 30 (4): 259 – 263.
[17] Sukumar S, Parameswaran P and Jayagopal L.S. Local- Distortional- and Euler- buckling of thin walled built-up open cross
sections under compression. Journal of Structural Engineering ASCE 2006; 32(6): 447-454.
[18] Cheng Yu and Benjamin W. Schafer Distortional Buckling Tests on Cold-Formed Steel Beams. Journal of Structural Engineering
ASCE 2006; 132(4): 505 – 514.

ISSN : 0975-4024 Vol 5 No 6 Dec 2013-Jan 2014 4631