Behavior and Analysis of RC T-Beams Partially Damaged in Shear and Repaired with
CFRP Laminates
Riadh Al-Mahaidi* , Kuan Lee and Geoff Taplin
Monash University, Australia; ash.edu.au
Abstract
In this study, the experimental results of three shear deficient T-beams strengthened using web-bonded CFRP strips are prented. Non-linear finite element modelling and analys are also ud to investigate the behaviour of the beams assuming plane stress condition and
perfect bond between the concrete surface and the web-bonded CFRP strips. Both numerical and experimental results are then compared to asss the viability of using non-linear finite element modelling in predicting the behaviour of CFRP shear strengthened T-beams.Introduction
In Australia, there are over 1000 reinforced concrete T-beam bridges. Most of the bridges were designed and constructed prior to 1939 when existing bridge design specifications were less stringent and methods of analys less preci. The u of half the level of shear reinforcement of current design standards was permitted. With today’s considerably higher design vehicle loads, some of the bridges are deemed to be deficient in shear requiring them to be strengthened, particularly tho already damaged. With limited financial resources available, bridge authorities are placing more emphasis on the strengthening and rehabilitation, as oppod to replacement, of the ageing bridge infrastructure.
An innovative strengthening technique that has attracted significant attention is the external bonding of fibre-reinforced plastic (FRP) to the structural member in distress. In bridges, the beams may be strengthened in flexure by bonding the FRP to the soffit or strengthened in shear through bonding of FRP to the web.
Sato et al. (1996a) and Taerwe et al. (1997) employed CFRP sheets in the shear strengthening of concrete beams. EMPA (1998a, 1998b) utilid prefabricated L-shaped CFRP laminate plates to strengthen T-beams. Khalifa and Nanni (2000) ud CFRP sheets anchored at the web-flange junction using a rather unique system. The CFRP sheet was bonded to the concrete surface and to t
he walls of a groove made in the flange. The groove was then filled with high viscosity binder and a Glass FRP rod placed into it. Both EMPA and Khalifa and Nanni reported flexural failures in the beams provided with end anchorage of the FRP. Sato et al, Taerwe et al reported premature peeling failure of the FRP sheets.
Little work has been done in investigating the behaviour of partially shear-damaged beams using FRP. In the prent study, the results of three shear-deficient T-beams strengthened using web bonded CFRP laminate strip are prented. This system is similar to that ud in the EMPA testing. The methodology for the finite element modelling of the CFRP-repaired T-beams is outlined as well. Numerical results are then compared with the experimental data.
Beam Specimens & Test Set-up
Three half-scale shear-deficient T-beams were strengthened using prefabricated L-shaped CFRP plates provided by Sika Australia Pty Ltd. The CFRP L-shaped stirrups were bonded to
the soffit and web of the beams extending into the flange, e Figure 1. Prior to bonding, the
D o w n l o a d e d f r o m a s c e l i b r a r y .o r g b y
E a s t C h i n a J i a o t o n g U n i v e r s i t y o n 02/20/14. C o p y r i g h t A S C E .
F o r p e r s o n a l u s e o n l y ; a l l r i g h t s
r e s e r v e d .
concrete surfaces were cleaned using high-pressure water jet to expo the aggregate.
Rectangular slots in the flange for the placement of the CFRP were fashioned using a hammer
drill. The beams were part of 21 T-beams previously tested to failure by Taplin & Al-Mahaidi
(1999) as part of an investigation into the ultimate strength of shear-deficient T-beams. Major
洞庭湖的老麻雀
shear-flexural cracks had developed in the beam at failure but occurred only in one half of the
beam. The partially damaged half that had only minor cracking was strengthened and tested
with the span shortened appropriately so as to exclude the major shear-flexural crack. The
CFRP L-shaped stirrups were placed at a spacing of 150 mm apart or half the overall depth of
the beam, D/2. The details of the three T-beams and test t-up are shown in Figure 2. The
beams were subjected to 3-point loading under displacement control using a Universal Amsler
testing machine.
采购员面试Table 1 shows the material properties of the beams and values ud in the finite element modelling. The location of strain gauges on the CFRP is shown in Figure 3. The
numbers indicate the distance from the bottom of the beam and the terms in brackets indicate
the designation.
F
o
r
p糖浆怎么熬
e
r
s
o
n多糖有哪些
a
l
u
集成吊顶尺寸s
e
o
n
l
y
;
a
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.化合价口诀
Table 1 - Material Properties
Concrete Young’s Modulus 29,240 MPa
Poisson’s ratio
0.2
Compressive Strength 25 MPa
Steel Young’s Modulus 200,000 MPa
Reinforcement Poisson’s ratio 0.3
Yield strength 718 MPa (Main Bars)
232 MPa (Stirrups)
CFRP Strips Young’s Modulus 165,000 MPa
Poisson’s ratio 0.184 (assumed)
小青枣Tensile strength 2800 MPa
Test Results
The beams were tested under a point load to failure. Displacement was measured at the bottom of the beam at the location of the point load. All beams failed by debonding of the main bars at one end or anchorage failure. Figure 4 shows the anchorage failure in beam 1.The maximum shear force achieved for beams 1, 2 and 3 was 153, 137 and 147 kN respectively. The shear strength achieved on the other end of the beams prior to strengthening was 91, 73 and 80 kN for beams 1, 2 and 3 respectively. This corresponds to 68%, 87% and 84% increa in the capacity of the beams
The load displacement respon of Beam 1 is plotted in Figure 6. The failure mechanism for this beam was due to bar anchorage failure at the left end of the beam, e Figure 4. Diagonal tension cracks developed in regions between the CFRP strips. The inclination of the cracks was about 45o . This is quite steeper than the diagonal cracks that developed in the original beams without CFRP strips. The inclination of tho was about 32o with the horizontal.
D o w n l o a d e d f r o m a s c e l i b r a r y .o r g b y
E a s t C h i n a J i a o t o n g U n i v e r s i t y o n 02/20/14. C o p y r i g h t A S C E .
F o r p e r s o n a l u s e o n l y ; a l l r i g h t s
r e s e r v e d .
Figure 4. Anchorage failure in Beam 1
Finite Element Analysis
DIANA (De Witte & Feenstra 1998), a versatile FEA package, was ud for the model analys. Plane stress elements are ud for all the models since out-of-plane forces are negligible. Steel reinforcements are modelled as embedded in concrete elements hence assuming full strain compatibility between the two components. Concrete cracking is bad on the smeared model approach with plasticity bad on the Drucker-Prager yield criterion.Second order mechanisms such as the softening of concrete in compression and tension were taken into consideration. Plasticity for reinforcement was bad on the Von Mis yield criterion. Strain hardening was also accounted for. Loading of the beam is induced by means of a prescribed displacement at the point load. In the FE models, eight-noded quadrilateral isoparametric elements were ud to model the concrete and the carbon fibre strips.
文员的基本要求
In the finite element mesh, 40-mm wide CFRP laminate strips were “bonded” to the web. This was achieved by generating an additional layer of elements over the concrete elements with both CFRP and concrete elements sharing common nodes, hence the
assumption of perfect bond. The finite element mesh is shown in Figure 5.
Figure 5. Finite Element Mesh
CFRP strips are orthotropic in nature with very high elastic modulus in the longitudinal direction compared to the transver direction. Due to modelling limitations, they were assumed to exhibit elastic-perfectly plastic behaviour. The thickness of the strip is 1.2
mm.
D o w n l o a d e d f r o m a s c e l i b r a r y .o r g b y
E a s t C h i n a J i a o t o n g U n i v e r s i t y o n 02/20/14. C o p y r i g h t A S C E .
F o r p e r s o n a l u s e o n l y ; a l l r i g h t s
r e s e r v e d .
Figure 6 shows the predicted and obrved load-deflection curves for Beam 1. It is evident that the CFRP repaired beam exhibited significant increa in load carrying capacity.The test results showed no debonding of the CFRP strips. The analysis didn’t take into account any consideration of debonding of CFRP strips. Failure of the beams was mainly due to bar anchorage failure at the left end of the beam. The apparent ductility in the beam respon was not due to yielding of the steel bars. Rather, it was due to the progressive anchorage failure of the bars at the left end of the beam.
Figure 6. Load-deflection respon of Beam 1
Figure 7 shows plots of the tensile strain in the cond CFRP strip to the left of the applied load. The corresponding predicted strain is also plotted. Initially, the FE model predicted negligibly small strains until diagonal cracks appeared at load level of 85 kN. The strains in strips incread at a fast
rate and began to approach the measured values. The maximum measured strain in this strip was 2342 x 10-6, this corresponds to a stress of 386MPa, which well below the tensile strength of the strip material. The maximum predicted strain in this strip is 1520 x 10-6, which corresponds to a stress of 251 MPa. Figure 8 prents the predicted contour plot of the vertical stress in the CFRP strips near the ultimate load.The maximum stress predicted is 473 MPa. This is well below the tensile strength of strips.
The predicted stress contours in the main steel bars are shown in Figure 9. The maximum predicted strain was 395 MPa at the ction of maximum bending. Although this is below the yield strength of the bars ud, it is well above the yield stress of bars ud in old T-beam bridges. One may conclude from this that had mild steel been ud, the failure would have been initiated by yielding of the main bars resulting in ductile flexural failure.
The stress contours in the steel stirrups are plotted in Figure 10. The maximum stress predicted is 242 MPa, which is in excess of the yield stress of the steel ud for the stirrups.Strain hardening was assumed for the steel stirrups and main bars.
D o w n l o a d e d f r o m a s c e l i b r a r y .o r g b y
E a s t C h i n a J i a o t o n g U n i v e r s i t y o n 02/20/14. C o p y r i g h t A S C E .
F o r p e r s o n a l u s e o n l y ; a l l r i g h t s r e s e r v e d .
