RESEARCH PAPERS

BEHAVIOUR OF BOLTED COLD-FORMED STEEL

ANGLE TENSION MEMBERS
By

ABIDA JUSTUS * R. PADMAPRIYA **

* Lecturer, Department of Civil Engineering, Sathyabama University, Chennai, India.
** Associate Professor, Department of Civil Engineering, Sathyabama University, Chennai, India.

ABSTRACT
Tension members are frequently encountered as principal structural members in trusses and lateral bracing system in
general construction and its connections are of significant importance in any steel design. The use of cold-formed steel
tension members in variety of structures has increased these days. The main objective of the study is to investigate the
behavior of cold formed steel single and double angle specimens. Forty eight experiments were conducted on single
and double angles of different cross sections with single and double line connection. The cold formed steel angle
specimens used in this investigation were fabricated from cold formed steel sheets of thickness 2mm by bending and
press breaking operations. Eight single plain angle specimens, eight single lipped angle specimens and thirty two
numbers of double angle specimens connected to the same side and opposite side of the gusset plate were tested in an
Universal Testing Machine using ordinary black bolts of 10mm diameter. All the specimens were tested to failure. Various
types of connection failure, Load vs deflection behavior were studied. Comparison of ultimate load calculated using BS
5950(Part V)-1998, AS/NZS 4600:2005 and experimental loads were also studied.
Keywords: Cold-Formed Angles, Tension Members, Shear lag.

INTRODUCTION node triangular elements. Wu and Kulak (1993) conducted

Cold- formed steel structural elements are widely used as
structural elements in roofs, decks, wall panels, trailer
bodies, agricultural equipments, aircrafts, etc. Angles are
the most basic and widely used sections among the
various forms of all rolled steel sections available.
Practically angles are connected with gusset plates
through one leg and due to this there will be non-uniform
stress distribution due to eccentrically applied load. The
reduction in load carrying capacity occurs due to a
phenomenon as shear lag effect. Chesson and Munse
(1963) carried out the study of shear lag effects on single
and double angles made of hot rolled sections. Their study
included different cross-sectional configurations,
connections, materials and fabrication methods. The
theoretical concept of shear lag and its effect on the angle
members were based on test results of 218 specimens
(among which there were 137 angle specimens) of various
configurations. Rickles and Yura (1983) conducted finite
element analysis to obtain the elastic stress finite element
model of two dimensional 4 node quadrilateral and 3
an experimental investigation of single and double angle
tension members to examine the effect of shear lag on the
net section rupture capacity of the cross section. They
tested 24 specimens (11 single angle members and 13
double angle members) to compare the ultimate loads
with the earlier test results obtained by others. They also
conducted finite element investigation to determine the
stress distribution of the critical cross section at ultimate
load.
Epstein and Chamarajnagar (1996) formulated a 20 node
quadratic brick element model. The material nonlinear
effects were modeled using the Von-Mises yield criterion
and the material stress-strain curve was assumed to be
elastic perfectly plastic. Epstein and McGinnis (2000)
conducted finite element study by including material non-
linearity as represented by a simple elastic-perfectly plastic
yield criterion. The finite element results agree closer with
experimental results. Chi – Ling Pan (2004) conducted tests
on cold formed steel channel sections with different
dimensions to investigate the effect of shear lag. The

i-manager’s Journal on Structural Engineering, Vol. 2 l

No. 2 l
June - August 2013 17
 RESEARCH PAPERS
comparisons were made between the test results and
predictions computed based on several specifications. To
study the stress distribution at the various locations of the
cross-section of specimen, the finite element software
ANSYS was used. Mohan Gupta and L.M.Gupta (2004)
analyzed angles with bolted connections using Finite
Element method giving due considerations to associated
problems such as the shape of the material, stress-strain
curve, the contact between the gusset plate and the
angle, the appropriate failure criteria, the effect of
punching of holes etc. Gupta and Gupta (2005) analysed
angles under tension in the limit state format giving due
considerations to block shear failure and yielding of gross
section. The factor of safety obtained as a result indicated
adequate representation of design strengths. Valdier
Francisco de paula et al (2008) presented experimental
results of 66 specimens carried out on cold-formed steel
angles fastened with bolts under tension. He conducted
multiple linear regression analysis and suggested the
expression for net section efficiency (U) which depended on
the geometrical factors such as connection eccentricity (x)
connection length (L), width of connected leg of the angle
(bc), net width of the angle with connected leg (bcn), width of
unconnected leg (bd), nominal bolt diameter (d) and angle
thickness (t). Prabha et al (2011) investigated the shear lag
phenomenon in cold-formed steel angles under tension
which are connected by one leg. A new expression for shear
lag factor which represents the net section reduction
coefficient has been suggested. They proposed equation for
cold-formed steel tension members which is in the same
format of IS:800-2007 (Indian code for Hot rolled steel
design).The proposed equation for IS:801 presents a good
estimate of the tensile capacity allowing for shear lag effect.
All the above investigations were made for the hot rolled
double angle sections. There were only limited
investigations for cold-formed steel members. The present
investigation aims to study the behavior of cold-formed
steel angle members.
1. Codal Provisions
The existing Indian Standard code of practice for cold-
formed steel IS 801-1975 does not elaborately deal with
the design of tension members. The following codal
18 i-manager’s Journal on Structural Engineering, Vol. 2 l
provisions are used to predict member capacities of the
cold-formed steel angle members.
1.1 Australian/NewZealand Standards: AS/NZS 4600-2005
The nominal section capacity of a member in tension shall
be taken as the lesser of
Nt =Agfy and (1)
Nt = 0.85KtAnfu (2)
Where Ag= gross cross sectional area of the member
fy=yield stress of the material
Kt=correction factor for distribution of forces. for
eccentrically connected single angles and double
angles connected to opposite side of the gusset
plate, the value of Kt = 0.85
for double angles connected to the same side of
the gusset plate the value of Kt = 1.0
An=net area of the cross-section, obtained by
deducting from the gross area of the cross-section,
the sectional area of all penetrations and holes,
including fastener holes.
fu = tensile strength used in the design.
1.2 British Standards: BS:5950 (Part 5)-1998
The tensile capacity Pt, of a member
Pt = Ae * py (3)
Single angles
For single angles connected through one leg only, the
effective area Ae is computed as
Ae = a1(3a1+4a2)/ (3a1+a2)
Double angles
For double angles connected to opposite side of gusset
plate, the effective Area is determined as
Ae = a1(5a1+6a2)/ (5a1+a2)
For double angles connected to the same side of gusset
plate the effective Area can be determined as that of
single angles.
Ae =effective area of the section
a1 =the net sectional area of the connected leg
A2 = the gross sectional area of the unconnected leg
py = the design strength.
No. 2 l
June - August 2013
 RESEARCH PAPERS

2. Experimental investigation with two mild steel gusset plates of thickness 8 mm at ends.
A total of forty eight experiments were conducted on single All the members are connected with gusset plate by
and double angles of different cross sections with single means of bolts. Ordinary block bolts of 10 mm diameter
and double line connection. Eight single plain angle were used in this connection. All the specimens were
specimens, eight single lipped angle specimens and thirty fabricated for length of 500 mm.
two numbers of double angle specimens connected to The length of gusset plate was provided according to the
the same side and opposite side of the gusset plate were pitch requirement and the edge distance as per Indian
tested in an Universal Testing Machine using ordinary black Standards. The gusset plates were not reused for single
bolts of 10mm diameter. The cold formed steel angle angle specimens and were reused for double angle
specimens used in this investigation were fabricated from specimens. All the members were connected with gusset
cold formed steel sheets of thickness 2mm by bending and plate to the larger side by means of bolts.
press breaking operations. Standard tension tests were Figures 2, and 3 presents the details of the fabricated single
conducted on coupons, stress vs strain curve was plotted and double angle specimens. All the specimens were
as shown in Figure 1. The values of yield stress, ultimate tested in a Universal Testing machine of 400kN capacity.
stress, modulus of elasticity and percentage elongation
obtained for these thicknesses of cold formed steel sheets
are presented in Table 1.
The specimens were tested as two different section
configurations as single and double angles. The single
angles were tested as plain angles and lipped angles. The
double angles were tested as two configurations namely
as double angles connected to opposite side of gusset
plate and double angles connected to same side of the
gusset plate.
The single and double angle specimens were connected

Figure 2. Details of single lipped angle Specimen

350

300

250
STRESS IN N/mm2

200

150

100

50

0
0 2000 4000 6000
STRAIN X 10 -6(mm/mm)

Figure 1. Standard tension test

Thickness of Yield Stress Ultimate Stress Modulus of Fu/fy Percentage

steel sheet in MPa in MPa Elasticity elongation
in mm (fy) (fy) in MPa in 200mm
gauge length

2 210 268 2.00 105 1.27 10

Figure 3. Details of double angle specimen connected to the
Table 1. Mechanical Properties of Steel Sheet opposite side of the gusset plate

i-manager’s Journal on Structural Engineering, Vol. 2 l

No. 2 l
June - August 2013 19
 RESEARCH PAPERS

The specimens were fixed vertically by gripping the gusset

plate. The load was applied eccentrically through the
gusset plate. Dial gauge was used for measuring the
elongation. Load was gradually applied with suitable
increments from control panel and for each increment of
load corresponding elongation was taken.
The yield, ultimate and breaking loads were also observed.
The procedure is repeated till the failure stage is reached in
all specimens. During the loading process the failure
pattern was recorded. Figures 4,5 and 6 show the
experimental setup for all single, double angle specimens
tested and Figure 7 shows the gusset plates and bolts used
for the connection.
3. Numerical Investigation
The goal of the finite element analysis is to develop a
model that could study the behavior of bolted cold-
formed steel single and double angle tension members. Figure 5. Experimental setup for double lipped angle specmen
connected to same side of the gusset plate
The behavior observed during the tests was used for
preparing a finite element model. All the one hundred and
twenty specimens were modeled using the finite element
program ANSYS (version 10). The problem was studied as a
nonlinear load vs displacement analysis including plasticity

Figure 6 Experimental setup for double plain angle specimen

connected to opposite side of the gusset plate

and nonlinear effect of geometry. SHELL 63 element type

Figure 4 Experimental setup for single plain angle specimen was used to model the single and double angle

20 i-manager’s Journal on Structural Engineering, Vol. 2 l

No. 2 l
June - August 2013
 RESEARCH PAPERS

Figure 8 Element meshing for single lipped angle 50x25x12x2

Figure 7. Details of the gusset plates and bolts
used for connections.

specimens. It is a 4 noded 3-dimensional quadratic elastic

shell element. It has both bending and membrane
capabilities. This element has six degrees of freedom at
each node: translations in the nodal x,y and z directions
and rotations about the nodal x,y and z axes. A typical
mesh of the model is shown in Figures 8 and 9. In the finite
element models, the shear deformation of the bolts was
ignored. The load was assumed to transfer from gusset
plate to the angle fully by the bearing of the bolts.
4. Results and Discussion
The behaviour of cold-formed steel single and double
angles when subjected to eccentric tension were studied.
Figure 9. Element meshing for double angle connected to
The ultimate-load carrying capacities of the specimens opposite side of the gusset plate (50x50x2)
were compared with the load carrying capacities
and 3. It is observed that in case of single angles the load
predicted using Australian/New Zealand and British
carrying capacity of lipped angle increases by 18.02%
standards. The experimental results were also compared
when compared to that of plain angles. Similarly, in case of
with the numerical results obtained using ANSYS software.
double angles the load carrying capacity increases by 5%
4.1 Experimental investigation
for angles connected to opposite side of gusset plate than
4.1.1 Ultimate load-carrying capacity to the same side of the gusset plate.
The experimental ultimate loads for all the cold-formed 4.1.2 Load vs Deflection
steel single and double angles are presented in Table 2
Figures 10 and 11 show the typical load versus deflection

i-manager’s Journal on Structural Engineering, Vol. 2 l

No. 2 l
June - August 2013 21
 RESEARCH PAPERS

35
S. No. Size of the specimen (mm) Ultimate load carrying capacity in kN
Single line Double line 30
connection connection 25

Load in kN
1 50x50x2 19 16 20
2 60x60x2 20 19 50x50x2 s1
15
3 50x25x2 19 18.4 60x60x2 s2
10
4 60x25x2 20 22
5 50x50x15x2 23 20 5
6 60x60x15x2 0
24 26
7 50x25x15x2 22 20.7 0 2 4 6 8
8 60x25x15x2 26 26 Longitudinal deflection in mm

Figure 11. Load vs Deflection behaviour of double plain angle

Table 2. Ultimate load carrying capacity of the single angles specimen connected to opposite side of the gusset plate
(double line connection)
S. No. Size of the specimen (mm) Ultimate load carrying capacity in kN
Single line Double line
connection connection
4.1.3 Modes of Failure

Double angles connected to opposite side of gusset plate The mode of failure of all single and double angle
1 50x50x2 35.8 46 specimens were noticed during testing. Generally tearing
2 60x60x2 42.4 41
3 50x25x2 42 37.2 failure, block shear failure, net section fracture failure were
4 60x25x2 45 45
5 50x50x15x2 47.2 35
observed as in Figures 12, and 13. The failure modes are
6 60x60x15x2 51 56 different for single and double angle sections. The mode of
7 50x25x15x2 46 48.1
8 60x25x15x2 54 54.4 failure depends upon the cross section and rigidity of
Double angles connected to same side of gusset plate connection.
1 50x50x2 34.6 31
2 60x60x2 39 40 During the loading process, the gusset plates of double
3 50x25x2 43.5 34.5
4 60x25x2 48.5 42.5 angle members remained straight. However, in the case of
5 50x50x15x2 45 35
6 60x60x15x2 49 45 single angles the gusset plate and the angles bend during
7 50x25x15x2 52 44 loading. This is due to eccentrically applied load. As the
8 60x25x15x2 55 50
load was being applied, the corners of the angle at the two
Table 3. Ultimate load carrying capacity of the double angles
ends gradually separated from the gusset plates for both
behaviour for single angles with and without lips and single and double angle members. Thus, a gap was
double angles. From the graphs, it is observed that the formed between the corner of the connected leg and the
ultimate load carrying capacity increases as the cross-
sectional area in the connection increases. It is also
observed that when the rigidity of the connection increases
the stiffness of the member also increases.
9
8
7
6
Load in kN

5
4 60x25x12x2
3 50x25x12x2
2
1
0
0 5 10 15 20
Deflection in mm
Figure 10. Load vs Deflection behaviour of single lipped angle Figure 12 Mode of failure for single lipped angle specimen
specimen(singleline connection) (Block shear failure) (double line connection)

22 i-manager’s Journal on Structural Engineering, Vol. 2 l

No. 2 l
June - August 2013
 RESEARCH PAPERS

The comparison of predicted ultimate loads by the various

Figure 13. Mode of failure for double plain angle specimen
connected to opposite side of gusset plate
(Net section failure) (single line connection)
gusset plate. The visible length of gap was usually from the
edge of the angle to the innermost bolt. The width of the
gap varied from one specimen to another, with a
maximum observed value of 10mm. Generally larger gaps
were associated with the cases of greater eccentricity of
codes for single and double angles tested is shown in
Figures 14 and 15. The tensile capacity equations of the
international codes take into account the effect of shear
lag and incorporate the capacity reduction factor in
addition to net effective area of the section. In case of
single line connection and double line connection the
values predicted by AS/NZS are higher than the
experimental loads and loads calculated by BS. Provision
of lip increases the load carrying capacity of the angles by
18.02% .
4.3 Numerical investigation
To perform the non-linear analysis, the angle specimens
are modeled based on the experimental set up
incorporating geometric imperfections. The geometric
imperfections included the thickness of the section, width
of the connected leg, width of unconnected leg in case of
single plain angles and it includes width of lip in case of
45
40
35
30
Load in kN

the cross-section, smaller angle thicknesses and shorter 25 Experimental

load
20
connection lengths. AS/NZS
15
There was no major slip of the connections during the tests. BS
10
All the specimens failed at the critical cross-section (inner 5
0
most bolt hole) as the ultimate load was reached. After 50x25x2 60x25x2 50x25x12x2 60x25x12x2
Specimens
necking, the critical cross-section was torn out from the
edge of the connected leg to the hole then to the corner of Figure 14. Comparison of ultimate loads with loads based on
codal provisions for single plain angles (single line connection)
the angle. The specimens carried some amount of load
80
beyond the ultimate load and until failure. It was noted that
70
all the bolts were still tight after completion of the tests. This
60
indicates that the bolts were not highly stressed during the
50
Load in kN

tests. The outstanding leg which is subjected to

40 Experimental load
compression experience local buckling nearer to the AS/NZS
30
supports. BS
20
4.2 Comparison of experimental and predicted ultimate 10
loads 0
A comparative study between the experimentally 50x25x2 60x25x2 50x25x12x2 60x25x12x2
Specimens
observed ultimate loads of the specimen tested with the
Figure 15. Comparison of ultimate loads with loads based on
tensile load carrying capacity of equations of the following
codal provisions for double angles connected to same side
codes AS/NZS:4600-2005, BS:5950 (Part 5 )-1998 is made. of the gusset plate (double line connection)

i-manager’s Journal on Structural Engineering, Vol. 2 l

No. 2 l
June - August 2013 23
 RESEARCH PAPERS
lipped angles. As the nonlinear problem is
dependant, the solution process requires a step by step
load incremental analysis. In the analysis, the solution
usually converged very slowly after yielding, and the
increment for each load step had to be made very small.
Yielding is determined using von-Mises yield criteria. At the
completion of each incremental solution, the program
adjusts the stiffness matrix to reflect the nonlinear changes
in structural stiffness before proceeding to the next load
increment. ANSYS employs Newton-Raphson equilibrium
iterations. The general post processor in ANSYS is used to
review results at each load increments. Figures 16 and 17
shows the stress distribution.
Figure 16. Element stress for double lipped angle connected to
same side of the gusset plate (60x60x15x2)
Figure 17. Element stress for single plain angle 50x25x2
24 i-manager’s Journal on Structural Engineering, Vol. 2 l
path Conclusions
Based on the experimental, theoretical and numerical
investigations, the following conclusions are made.
·The ultimate load carrying capacity increases as the
cross – sectional area increases. Provision of more
number of bolts improves the connection rigidity which
also contributes to increase in load carrying capacity.
·The presence of lip increases the load carrying
capacity of single angles by 18.02%. The load carrying
capacity increases by 5% for double angles
connected to the opposite side of the gusset than the
connected to same side of gusset plate.
·Cold-formed steel angles with larger outstanding legs
experiences local buckling under eccentric tensile
loading.
·In case of single line connection and double line
connection, the values predicted by AS/NZS are higher
than the experimental loads and loads calculated by
BS.
·The stress contours obtained in the finite element
analysis indicates that maximum stresses occur in the
innermost bolt holes from which the experimental
failures were initiated.
Ackowledgement
The authors express their heartfelt thanks to the P.G. Students
Mohamed Riyadh, James, Karthiga , Arun and also Non
teaching staff of Strength of Materials laboratory for the
help rendered by them for carrying out the experimental
work.
References
[1]. AISI commentary (2007). 'Design of Cold-formed Steel
Structural Members “, North American Specification.
[2]. AS/NZS: 4600 (2005). 'Cold-formed Steel Structures',
Australia / New Zealand Standard.
[3]. BS:5950-Part 5 (1998),.'Structural Use of Steelwork in
Building-Code of practice for design of cold-formed thin
gauge sections', British Standards Institution.
[4]. Chesson EFjr. & Munse W.H (1963). “Riveted and bolted
joints truss type tensile connections”, J.Struct Div., ASCE,
Vol. 89(1), pp 67 – 106.
No. 2 l
June - August 2013
 RESEARCH PAPERS

[5]. Chi – Ling pan. (2004). “Prediction of the strength of
bolted cold – formed channel sections in tension”, Thin
walled structures , Vol. 42 , pp 1177 – 1198.
[6]. Epstein, H.I., & Chamarajanagar, R.,(1996). “Finite
element studies for correlation with block shear tests,
“Computers and Structures”, Vol.61, No.5, pp 967-974.
[7]. Eric Yue Wu & Geoffrey L.Kulak (1997), “Shear lag in
bolted angle tension members”, Journal of Structural
Engineering, Vol.123, No.9 pp 1144-1152.
[8]. Gupta, L.M., & Mohan Gupta (2004). “Evaluation of
stress distribution in bolted steel angles under tension”,
Electronic Journal of Structural Engineering.
[9]. Gupta L.M., & Mohan Gupta., (2005). “Limit state
design of bolted steel angles under tension”, Journal of
Structural Engineering, Vol.31, No.4, pp265-274.
[10]. LaBoube R.A., & Yu W.W. (1995). 'Tensile and bearing
capacities of bolted connections', Final summary Report,
Civil Engineering study 95-6, University of Missouri-Rolla.
[11]. Prabha.P., ArulJayachandran.S., Saravanan.M., &
Marimuthu.V (2011). 'Prediction of the tensile capacity of
cold formed angles experiencing shear lag', Thin-walled
structures, Vol. 49, pp.1348-1358.
[12]. Valdier Francisco de Paula., Luciano Mendes
Bezerra., & William Taylor Matias (2008). 'Efficiency
reduction due to shear lag on bolted cold-formed steel
angles', Journal of Constructional Steel Research, Vol.64,
pp.571-583.

ABOUT THE AUTHORS

Mrs. Abida Justus is presently working as Lecturer, Department of Civil Engineering, Sathyabama University, Chennai. She
obtained her Master's degree from Anna University.

Dr.R.Padmapriya M.E, Ph.D is working as Associate Professor, Department of Civil Engineering Sathyabama University, Chennai.
She has obtained her Ph.D degree from Anna University and has 11 years of teaching experience.

i-manager’s Journal on Structural Engineering, Vol. 2 l

No. 2 l
June - August 2013 25