Professional Documents
Culture Documents
Monotonic and Cyclic Loading of New FRP Reinforced Concrete Cantilever Beams
Monotonic and Cyclic Loading of New FRP Reinforced Concrete Cantilever Beams
M.Kazem Sharbatdar
Asst. Prof., Dept. of Civil Engineering, Engineering Faculty, Semnan University, Semnan, Iran
Downloaded from ijce.iust.ac.ir at 10:16 IRDT on Saturday September 1st 2018
Abstract : FRPs (fiber reinforced polymer) possess many favorable characteristics suitable and applicable for
construction industry when compared with steel reinforcement. There are new ideas to use FRPs as
longitudinal or transverse reinforcement for new concrete elements particularly for bridge decks or beams.
Although high tensile strength of FRP is main characteristic for applications at both areas, its weakness to
bending and linear stress-strain behavior with virtually no ductility, makes it vulnerable to probably
premature failures under reversal tension-compression loading during earthquake. A pilot research project
has been conducted to explore the characteristics of large-scale cantilever concrete beams reinforced with
FRP re-bars and grids and were tested under either simulated cyclic loading or monotonically increasing
lateral loading. This paper presents the test parameters and results obtained during research. The analytical
relationships are compared with those recorded experimentally, and test results showed the diagonal cracks
and either rupturing of FRP bars in tension or stability failure in compression bars at long or short shear span
beams. The comparison of nominal moment capacities between analytical and experimental values confirms
that plane section analysis is applicable to FRP reinforced concrete members.
Keywords: Beams, Shear and Flexural Behavior, Monotonic Loading, Fiber reinforced polymers (FRP),
Longitudinal re-bars, Transverse Grids, Ductility, Plane Section Analysis
members so long as the yield strength of be comparer with those recorded experimentally.
reinforcement is replaced by a rational value for
FRP strength.
3. Experimental Research
305 m m
1000 mm
Downloaded from ijce.iust.ac.ir at 10:16 IRDT on Saturday September 1st 2018
63 mm
D iam eter
405 mm
H o le
s /2
405 mm
520 m m
S h o rt B ea m
s
63 mm
Diameter
H ole
s/2
5 20 mm
116 0mm
Long Beam
Figure.1- Geometry of Cantilever Beams
(strong-side) (weak-side)
CFB1 35 10-9.5 mm bar 0.39 0.26 175 1000
(AF=6-9.5 mm at top side & A'F=4-9.5 mm at bottom side of beam) (ρ=AF/bd , ρ’=A'F/bd)
Stress (MPa)
900
Downloaded from ijce.iust.ac.ir at 10:16 IRDT on Saturday September 1st 2018
600
Test#1
300
Test#3
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
Strain(%)
Fig.2 Tensile Stress-Strain Relationship of Pultrall CFRP Bars
the tensile stress-strain relationship of bars tested of hysteretic and monotonic relationships are
at the lab; it shows linear behavior up the rupture presented and discussed in this section. The
load. The tensile strength of these was hysteretic behavior is presented both in terms of
approximately 4 times that of steel bars. The force-displacement, moment-drift, moment-total
transverse reinforcement used consisted of rotation relationships. Rotation readings were
NEFMAC CFRP grids. The grids had a taken within the potential hinge region. The
rectangular configuration with 250 x 350 mm hinging region was defined as the beam segment
out-to-out dimensions. Two types of grids were between the beam-column (footing) interface and
employed; i) 6 x 8 mm rectangular cross-section the section 405 mm away from the interface
forming two equal-size rectangular openings, and (equal to the longer cross-sectional dimension).
ii) 8 x 10 mm rectangular cross-section forming The first set of readings gave total rotation of
two equal-size rectangular openings. assumed hinging region relative to the column.
This set of readings consisted of rotations mostly
The beams were prepared along with attached due to flexure and also due to anchorage slip. The
reinforced concrete elements, representing the rotations due to the anchorage slip were
framing columns. These elements were used to measured and formed the second set of readings.
fix the beams to the laboratory strong floor to These readings were taken as the rotation of a
provide full fixity. The instrumentation consisted beam section near the beam-column interface,
of LVDTs for displacement and rotation relative to the column. Ideally, these readings
measurements, and electric resistant strain should have reflected the rotation of the critical
gauges for reinforcement strain measurements. beam section at the interface. However, the
The strain gauges were placed on longitudinal LVDTs used required some gauge length to be
and transverse reinforcement. A 1000 kN positioned on the beam. Hence, they were
capacity servo controlled MTS actuator was used mounted on a section approximately 25 mm
to apply the lateral load directly on the beams, no above the interface. The difference between the
axial load was applied to the beams. A steel box total and anchorage slip rotations gave those
section was attached to a steel plate by means of caused by flexure.
dwydag pre-stressing bars and used to apply the
load. The assembly was first secured near the tip The variations of reinforcement strains in
of beams. The detail of test set-up, photographs, longitudinal and transverse reinforcement are
and locations of LVDTs are illustrated in Figure 3. presented in the form of hysteretic force-strain
relationships. Each beam had four longitudinal
3.2. Observed Behavior and Test Results bars instrumented with a total of 6 strain gauges
Observed behavior of beams such as photographs to measure the variation of strain in CFRP bars.
taken during testing and also test results in terms All four corner bars were instrumented at the
4 Dwydag Bars
Steel Plate Actuator Bolts
Downloaded from ijce.iust.ac.ir at 10:16 IRDT on Saturday September 1st 2018
MTS
Pulley Roller Actuator
String C
1000mm
870 mm
Steel Box
Cantilever Beam
405
25mm
6 5 4 3
Temposonic
LVDT(1&2) Rotation Footing
Measurment
LVDT's
Mecano Framing
Fixed to Footing
mm inside the column. The end grids (closer to 0.1 f c' bw d @ Vc 0.035⎜ f c' U w E F d ⎟ bw d @ 0.2 f c' bw d
⎜ M ⎟
the columns) were instrumented with strain ⎝ f ⎠
gauges, as well. Beams with d/2 grid spacing had (2)
the first three grids instrumented, while those
with d/4 spacing had the first four instrumented. Equation (1) is for sections with an effective
The amount and spacing of grid reinforcement as depth greater than 300 mm and with transverse
an important considered parameter varied in test shear reinforcement less than minimum amount
beams. These test parameters are expressed of shear reinforcement, bW and d Equation (2)
relative to those required by CSA S806-02. also is for sections having either at least the
Therefore, a summary of these requirements is minimum amount of transverse reinforcement or
presented below. CSA S806-02 shear design an effective depth not exceeding 300 mm. and
requirements are based on 45-degree truss are effective web width and effective depth of
analogy with contributions from concrete and section, respectively. f Bc and EF are cylinder
FRP reinforcement. Accordingly, the concrete compressive strength of concrete and modulus of
shear resistance, Vc, without considering elasticity of FRP bar in tension. Longitudinal
reduction factors, is calculated at two different reinforcement ratio, UW, is equal to ( A ).VF frp
bw d
Diagonal
CFB1 tension failure
Bar Rupturing
90 L 160
H
M 60
M = F.L 30 M 80
Downloaded from ijce.iust.ac.ir at 10:16 IRDT on Saturday September 1st 2018
M = F.L
0
-4 -3 -2 -1 0 1 2 3 4 5 6 0
-30
-4 -3 -2 -1 0 1 2 3 4 5
Drift(%)
-60 Drift(%)
-80
CFB1 CFB4
-90
H = 1000 mm H = 1900 mm
-120 L = 870 mm -160 s = d/2=180mm L = 1780 mm
M
80
M 80
M = F.L
M = F.L
0
0
-4 -3 -2 -1 0 1 2 3 4 5
-4 -3 -2 -1 0 1 2 3 4 5 6
Drift(%)
Drift(%) -80
-80 CFB5
CFB2
H = 1980 mm
H = 1000 mm
-160 s = d/4=90 mm L = 1870 mm
-160 L = 870 mm
160 160 M
M
M = F.L M = F.L
120 120
80 80
40 40
0 0
0 2 4 6 8 0 2 4 6
Drift(%) Drift(%)
Flex. failure
(bars ruptured)
specified in CSA S806 for seismic design. It also same primary (envelope) force-deformation
meets the maximum spacing requirements for relationship. The beams showed increasing load
shear and stability of longitudinal compression resistance up to 4% drift, followed by subsequent
bars. The reduction in grid spacing increased the strength decay beyond this deformation level due
effectiveness of shear resistance while enhancing to the failure of FRP bars in tension. The strength
concrete confinement, and developed its flexural decay continued gradually up to 6% drift ratio
strength. The hysteresis loops showed near with a clear reduction in load resistance every
elastic behavior with some stiffness degradation time there was bar rupture in tension.
due to concrete cracking up to 3% drift ratios.
Degradation of strength began at the first cycle of Strain gauge readings of beams indicated that all
3% drift, and continued till 4% drift ratio. At this negative FRP bars ruptured in tension at strains
level of deformation, all the 4 positive bars 1.15% to 1.4%, but compressive bars
ruptured in tension causing the beam to loose its experienced 0.15% to 0.3% strain. FRP grids
flexural resistance in the weak direction. The experienced a tensile strain of 0.3% to 0.5%.
beam was able to resist severely reduced loading Figures 7 and 8 show the hysteretic behaviors of
in the strong direction when pushed to 5% drift strains in bars and grids, respectively.
before the rupturing all negative bars and then
ending test. Beam CFB3 was identical to beam 3.3. Effects of Test Parameters
CFB2, except it was tested under monotonically Parameters affecting shear and also effect of
increasing lateral loading. The results showed cyclic loading were as main test parameters. The
that both Beams CFB2 and CFB3 had specimens were designed with two shear spans
approximately the same primary (envelope) and two spacing of transverse shear
force-deformation relationship. The beam reinforcement to investigate the shear behavior of
showed increasing load resistance up to 3% drift, FRP reinforced beams. While beams with short
followed by subsequent strength decay beyond shear span would be subjected to higher shear
this deformation level due to the failure of FRP stresses, those with closely spaced transverse
bars in tension. The strength decay continued grids would have higher shear capacity. Beams
gradually up to 6% drift ratio with a clear CFB1 and CFB2 had short shear spans with two
reduction in load resistance every time there was difference spacing of grid reinforcement. Beam
bar rupture in tension. CFB1 developed a wide diagonal tension crack
due to shear. The spacing of grid reinforcement
Beam CFB4 was companion to CFB1, except for had been reduced by 50% in CFB2, resulted in
its longer length of 1900 mm with promoting the the doubling of the amount of shear
flexural behavior. The hysteretic relationship reinforcement. The beam did not suffer from
indicates that the beam showed stable hysteresis shear failure, and behaved in the flexure mode
loops up to 3.5% drift ratio, but developed despite its short shear span, developing 37%
significant strength decay after the first cycle of 4 higher shear resistance. This is evident in Figure
% drift ratio. The beam failed during the first 9 where the envelopes of hysteretic force-
cycle at 4.5 % drift ratio due to the failure of FRP deformation relationships are compared.
bars in tension. Beam CFB5 was identical to
CFB4, except for the spacing of grids, which was The comparison of long beams CFB4 and CFB5
reduced to 90 mm. Hysteresis loops indicate with different grid spacing does not show any
stable behavior up to 2% drift ratio, for loading in significant difference in behavior. This is
both directions. Figure 6 shows the beams CFB5 expected since these beams behaved
90 90
#6
60 60
4-bar
Downloaded from ijce.iust.ac.ir at 10:16 IRDT on Saturday September 1st 2018
30 30
Strain(%)
0 0
-0.4 -30 0.0 0.4 0.8 1.2 -0.4 -30
0.0 0.4 0.8 1.2
Strain(%)
#4
-60 -60
6-bar sideCFB4
-90 -90 CFB4
H=1980 mm 80
H=1900 mm
S= 90 mm S= 180 mm
0
0
-6 -4 -2 0 2 4 6
-6 -4 -2 0 2 4 6
-80 Drift (%)
-80 Drift (%)
-160 CFB1
-160 CFB5 CFB4
CFB2
-140 CFB2
CFB1 -160 CFB5
CFB4
-160 CFB1
CFB4 - 160 CFB5
CFB2
predominantly in the flexure more because of premature shear failure, whereas CFB4, with the
their longer shear spans. Figure 10 shows same grid spacing was able to develop its flexural
envelope curves of moment-drift relationships strength without shear failure. Figure 11 also
for beams with different shear spans. Of indicates that flexural rotations measured in
significance is the comparison of CFB1 and CFB2 were 70% and 36% higher than those of
CFB4, both with wide grid spacing. The figure CFB1 in the strong and weak directions,
indicates that CFB1, with a short shear span was respectively. Two of the beams, CFB3 and CFB6
subjected to higher shear stresses and suffered were tested under monotonically increasing
CFB3 60
80 CFB6
H=1000 mm, S= 90 mm
Downloaded from ijce.iust.ac.ir at 10:16 IRDT on Saturday September 1st 2018
H=1980 mm, S= 90 mm 30
Monotonic
Monotonic
0
0
-50 -40 -30 -20 -10 0 10 20 30 40 50 60 70
-100 -80 -60 -40 -20 0 20 40 60 80 100 120
-80 Late ral Displaceme nt (mm) -30
Lateral Displace me nt (mm)
-60
-160 CFB3
CFB2
-90 CFB6
CFB5
lateral loading, while all others were tested under against sectional curvatures. The resulting
reversed cyclic loading. Each of these two beams diagram provides a moment-curvature
had a different shear span, but the same relationship, which exhibits all the relevant
reinforcement, including the same spacing of characteristics of a section in terms of strength,
transverse grids. Comparing CFB2 could see the stiffness and deformability. A computer program
effect of cyclic loading and CFB3 with short was used to generate analytical moment-
shear spans and CFB5 and CFB6 with long shear curvature and compared with experimental
spans. The envelope curves for these four beams results. An important step in conducting the
are presented in Figures 12. The results indicate analyses was the consideration of realistic
that the monotonic force-deformation material models for constituent materials. The
relationships are similar to the envelopes of stress-strain relationships for unconfined and
hysteretic relationships obtained under cyclic confined concrete were adopted from earlier
loading, suggesting very little or no effect of research, while the stress-strain relationship for
cyclic loading on beams. FRP reinforcement was established
experimentally.
S=90 mm Cyclic
Moment, M (kN.m)
Monotonic 160 Loading
160
Loading
Downloaded from ijce.iust.ac.ir at 10:16 IRDT on Saturday September 1st 2018
80 80
Model
Model
Experimental-LVDT
Experimental-LVDT
------ Experimental-Strain Gauges
0 ------ Experimental-Strain Gauges
0
0 20 40 60 0 10 20 30 40 50 60
-6 -6
Curvature(1 / mm) x1 0 Curvature(1 / mm) x1 0
Moment, M (kN.m)
H=1980 mm
Moment, M (kN.m)
S=90 mm S=90 mm
160 Cyclic Loading 160 Monotonic
Loading
80 80
Model Model
Experimental-LVDT Experimental-LVDT
------ Experimental-Strain Gauges ------ Experimental-Strain Gauges
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
-6 -6
Curvature(1 /mm) x10 Curvatur e(1 /mm) x 1 0
difference between measured total rotation and strength. The agreement between analytical and
anchorage slip gave flexural rotation within 355 experimental values is reasonably good, once
mm of column segment. The flexural rotation again confirming the applicability of plane
divided by the segment length provided section analysis to FRP reinforced concrete
experimental average curvature to be plotted beams.
against average moment within the 355 mm
segment for comparison with analytical moment-
curvature relationship. The comparison for each 5. Summary And Conclusion
beam is illustrated in Figures 13. The figure also
includes a second set of relationships for Experimental and analytical results of beam tests
experimental moment-curvature relationships, conducted in this paper present several
this time obtained from the strain gauges located conclusions. Tension FRP reinforcement would
on longitudinal FRP bars at column-footing rupture prior to significant distress in concrete at
interface. These relationships show curvatures all beams except for CFB1, which was critical in
and corresponding moments at exactly the beam- shear. And also all the flexure dominant beams
column interface. developed their flexural capacities computed on
the basis of plane-strain analysis. Beams with
The results indicate approximately linear short shear span and wide spacing of transverse
moment-curvature relationship for both reinforcement developed a wide diagonal tension
analytical and experimental values because of the crack due to shear, resulting in diagonal tension
failure mode of beams. All beams failed by failure in tensile bars, while the compression bars
rupturing of FRP bars in tension before crushing were subjected increased dowel action in addition
of concrete. Maximum compressive fiber strain to compression, and experienced stability failure
recorded was approximately 0.003, which is only and then was broke. Beams with the same grid
slightly above the strain at unconfined concrete spacing but long shear span were able to develop
monotonically increasing lateral force are similar [3] ACI Committee 440 (1996), "State-of-Art
to the envelopes of hysteretic relationships report on Fiber Reinforced Plastic
obtained under reversed cyclic loading, Reinforcement for Concrete Structures,"
suggesting very little or no effect of cyclic American Concrete Institute, USA,
loading on beams and this implies that a 1996,68 pp.
monotonic curve would provide sufficient
strength envelope for hysteretic relationships of [4] ACI Committee 440.1R-01(2001), "Guide
FRP reinforced beams. The hysteretic for the Design and Construction of
relationship of flexure dominant FRP reinforced Concrete Reinforced with FRP Bars",
beams indicate stiffness degradation under cyclic American Concrete Institute, USA, 2001,
loading, due to progressive cracking of concrete 41 pp.
and associated softening in the member,
developing approximately 3% lateral drift ratio [5] ACMBS III, Advanced Composite
prior to failure. This level of lateral drift may be Materials in Bridges and
considered to be sufficient for earthquake Structures.(2000). "Proceedings of the
resistant construction. Moreover the results Third International Conference (Ottawa,
indicate approximately linear moment-curvature Canada)". Humar, J., and Razaqpur, G.
relationship for both analytical and experimental editors. The Canadian Society for Civil
values because of the failure mode of beams by Engineering, Montreal, Canada.
rupturing of FRP bars in tension before crushing
of concrete. The agreement between analytical [6] Alsayed, S.H., Al-Salloum, Y. A.,
and experimental values is reasonably good, Almusallam, T.H., and Amjad, M. A.
confirming the applicability of plane section (1999). "Concrete Columns Reinforced by
analysis to FRP reinforced concrete beams. Glass Fiber Reinforced Polymer Rods,"
American Concrete Institute, Special
Publication, SP-188, Farmington Hills,
6. Acknowledge Michigan, p.103-112.
The author would like to acknowledge Professor [7] Canadian Standards Association, S806-02,
Murat Saatcioglu (U of Ottawa) for his guidance, May 2002, " Design and Construction of
valuable suggestions, encouragement, and Building Components with Fibre-
funding and also Technical officers of the Reinforced Polymers", CSA, Ontario.
Structural Laboratory at the University of Ottawa
in Canada for their assistance and continuous [8] FRPRCS-3, Non-Metallic Reinforcement
support during the course of this investigation. for Concrete Structures. (1997).
"Proceedings of the Third International
Symposium on Non-metallic (FRP)
7. Reference reinforcement for Concrete Structures
(Sapporo, Japan)". Japan Concrete
[1] ACI Committee 318, 1999. Building code Institute, Tokyo, Japan.
requirements for reinforced concrete and
commentary (ACI 318-95). American [9] FRPRCS-4, Fiber Reinforced Polymer
Concrete Institute, Farmington Hills, MI. Reinforcement for Concrete Structures.
369 pp. (1999). "Proceedings of the Fourth
[10] Fukuyama, H. and Masuda, Y. (1995). Reinforced with FRP Stirrups," 1993, ACI,
"Structural Performances of Concrete SP-138-47, p.789-811.
Frame Reinforced with FRP
Reinforcement," Non-metallic (FRP) [18] Nanni, A.(2001)," North America design
Reinfocement for Concrete Structures. Guidelines for Concrete Reinforcement
Edited by Taerwe. E&FN Spon, London, and Strengthening Using FRP: Principles,
p.275-286. Applications, and Unresolved Issues" FRP
Composites in Civil Engineering,Vol1,
[11] ISIS Canada, ISIS-M04-00 Draft, 2000, " Conference proceeding, Hong Kong,2001.
Reinforcing Concrete Structures With
Fibre Reinforced Polymers", Manitoba, [19] Pauly, T., Priestly, M.J.N., " Seismic
Winnipeg, Canada. Design of Reinforced Concrete and
Masonry Buildings", Jihn Willey & Sons,
[12] Japanese Society of Civil Engineers New York, 1991, 744 pp.
(JSCE). (1997). "Recommendation for
Design and Construction of Concrete [20] Saatcioglu, M., and Salamat, A.H., and
Structures Using Continuous Fiber razvi, S.R., "Confined Columns Under
Reinforcing Materials." Concrete Eccentric Loading," Journal of Structural
Engineering Series 23.Tokyo. Engineering, ASCE, 1995, 121(11),
p.1547-1556.
[13] J.G.Teng, J.F.Chen, S.T.Smith, L.Lam
(2002), " FRP Strengthened RC [21] Sameh R. Salib et al (2001)," Strength of
Structures", John Wiley&Sons Ltd. Concrete Beams Reinforced and/or
Prestressed With FRP Bars", FRP
[14] Kent D.C. and Park R. (1971), " Flexural Composites in Civil Engineering,Vol1,
Members with Confined Concere," Journal Conference proceeding, Hong Kong,2001.
of Structural Division, ASCE, 97(ST7),
p.1969-1990. [22] Sonobe, Y., Fukuyama, H., Okamoto, T.,
Kani, N., Kimura, K., Kobayashi, K.,
[15] Kobayashi, K. and Fujisaki, T. (1995). Masuda, Y., Matsuzaki, Y., Mochizuki, S.,
"Compressive Behavior of FRP Nagasaka, T., Shimizu, A., Tanao, H.,
Reinforcement in Non-Prestressed Tanigaki, M., Teshigawara, M. (1997).
Concrete Members," Non-metallic (FRP) "Design Guidelines of FRP Reinforced
Reinfocement for Concrete Structures. Concrete Building Structures." Journal of
Edited by Taerwe. E&FN Spon, London, Composites for Construction, ASCE,
p.267-274. Vol. 1, No.3,90-114.