- Email: [email protected]
Experimental research on flexural behaviors of damaged PRC beams strengthened with NSM CFRP strips
Construction and Building Materials 190 (2018) 265–275
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Experimental research on flexural behaviors of damaged PRC beams strengthened with NSM CFRP strips Jia Yang a,b, Lianguang Wang a,⇑ a b
School of Resources and Civil Engineering, Northeastern University, Shenyang 110819, China College of Architectural and Civil Engineering, Shenyang University, Shenyang 110044, China
h i g h l i g h t s PRC beams strengthened with NSM CFRP strips were tested under overloading. Influence of cycle number, overload amplitude, and strengthening under loading was investigated. The above three variables had limited influence on the ultimate bearing capacity. The above three variables affected stiffness and ductility of the strengthened beams.
a r t i c l e
i n f o
Article history: Received 28 November 2017 Received in revised form 17 September 2018 Accepted 17 September 2018
Keywords: Flexural behaviors Experimental research PRC beam NSM CFRP strips Damage
a b s t r a c t This paper investigates the flexural behaviors of the damaged prestressed reinforced concrete (PRC) beams strengthened with near surface mounted (NSM) carbon fiber-reinforced polymers (CFRP) strips under overloading. A bending test was carried out to evaluate the effects of three variables: the cycle number, overload amplitude and strengthening under loading or unloading. Different damage degrees were mainly controlled by cycle number and overload amplitude. The test results showed that strengthening with NSM CFRP strips could effectively inhibit crack development. Compared with the unstrengthened beam, the flexural load-carrying and stiffness of the strengthened beams were enhanced evidently. With the increase of cycle number and overload amplitude, the yield loads increased. The yield load of strengthened beam under unloading was higher than that of strengthened beams under loading. These three variables had limited influence on the ultimate bearing capacity. The deformations of the strengthened beams were smaller than that of the control beam, and decreased with the increase of cycle number, but increased with the increase of overload amplitude. The deformations of the strengthened beams under loading were slightly greater than that of the strengthened beam under unloading in the early phase but smaller in the later phase. The ductility of strengthened beams was evidently reduced compared with that of the control beam. The displacement ductility factors of the strengthened beams under loading decreased with the increase of cycle number and overload amplitude, and were smaller than that of strengthened beam under unloading. Ó 2018 Elsevier Ltd. All rights reserved.
1. Introduction With the increasing scaling-up and heavy-load tendencies of vehicles, the actual traffic volume of a few in-service bridges has exceeded the designed traffic volume. Moreover, the load design standards in new and old bridge design specifications are not uniform, thereby resulting in overload. Operation under an overload situation poses serious harm to the structure of bridge, thus occasionally causing bridge collapse accidents. Moreover, numerous ⇑ Corresponding author. E-mail address: [email protected] (L. Wang). https://doi.org/10.1016/j.conbuildmat.2018.09.109 0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
bridges frequently experience overload [1–3]. Although overloaded bridges do not collapse, they pose severe potential structural hazards. The load-bearing properties of the bridges damaged by overloads have become the focus of researchers and engineering personnel. Strengthening of damaged bridges is necessary to improve the bearing capacities and stiffness, and lengthen their service life. Currently, fiber-reinforced polymers (FRP) are commonly used for strengthening [4–8]. The most common FRP strengthening method of reinforced concrete (RC) beams is to externally bond (EB) FRP laminates onto the soffit of a beam [9]. However, the effectiveness of this system is limited by the most common debonding failure mode, which occurs at an effective
266
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
strain much lower than the ultimate strain achievable by the FRP composite materials [10]. In order to maximize the utilization of the FRP materials, the failure mode caused by debonding should be overcome. At present, strengthening with prestressed FRP material and NSM FRP technique have been introduced. The EB prestressed CFRP strips or sheets using nonmetallic necessary anchor systems can increase the load-carrying capacity [11,12]. The NSM FRP strengthening technique has attracted significant attention worldwide as one of the promising new techniques for structural strengthening [8]. NSM FRP technique involves cutting grooves in the concrete cover, filling the grooves with adhesive and embedding FRP bars or strips into the grooves [13,14]. NSM FRP has several advantages over EB FRP, including better protection of the FRP and a stronger bond between FRP and concrete [15]. The NSM technique can increase load carrying capacity significantly and improve the stiffness of the beams. This technique is more effective than EB CFRP reinforcement, provides higher strength capacity and higher utilization of the strips using the same material with the same axial stiffness [14,16–17]. At present, a few studies conducted in China and abroad deal with the mechanical properties of RC structure under overloading [1–3,18–19]. Numerous studies have been conducted on FRPstrengthened RC and PRC structures [20–21]; a relatively greater number of investigations used the NSM CFRP strips to strengthen RC beams [22–24], but only few studies about NSM CFRP strips strengthened PRC beam. For overload damage, X. Y. Sun and R. X. Wang [25–27] studied the mechanical properties of damaged RC beams strengthened with CFRP strips and steel plates under overloading, which results showed that the strengthening effect was mainly influenced by overload amplitude, cycle number, strengthening method and reinforcement ratio, and overload amplitude and cycle number had considerable effects on the service life of beams. However, studies on the influence of overload on PRC beams are seldom. The most direct approach is to study the flexural behaviors of PRC structure strengthened with CFRP strips under overloading to perform a destructive load test. However, considerable bridge collapse risk is apparent when real vehicles are used to conduct the overload test, and test objects are hardly available (bridges may experience severe structural damage, thereby hindering them from continuing service). As stipulated by the American Association of State Highway and Transportation Officials [28], the internal reinforcement stress of bridge member in service ability limit state shall not exceed 60% of the yield stress of the reinforcement, and any value that exceeds this limit shall be considered overload. Hence, in this study, 0.9 and 1.3 times yield load of the control beam are considered upper limits of overload amplitude, and 0.1 times yield load is the lower limit. Cyclic loading is implemented for test beams to simulate the overload damage state. Normally, the occurrence frequency of overload amplitude is low; hence, the cycle numbers are 1, 50, and 100 times. In addition, components that require strengthening in engineering are generally under the loading state. Therefore, strengthening tests with NSM CFRP strips are conducted for damaged PRC beams under overloading to investigate changes in bearing capacity and stiffness with the increase of cycle number and overload amplitude before and after strengthening and the influence of strengthening under loading on flexural behaviors.
2. Experimental program 2.1. Specimen design and production A total of six specimens, which were all unbonded prestressed concrete T-beams, were designed in the test; one was the control beam, and five were strengthened beams with NSM CFRP strips.
All beams had a length of 4.5 m and a net span of 4.3 m. The measured compressive strength of concrete cube was 32.5 MPa at 28 days. The beams were reinforced by two HRB-335 16 mm diameter longitudinal steel bars, four HPB-300 10 mm diameter steel bars, 6 mm diameter stirrups with spacing of 200 mm in the concrete flange, 8 mm diameter stirrups with the spacing of 120 mm in the concrete web, and two 15.2 mm diameter low relaxation seven-wire prestressing steel strands. The steel reinforcement and the CFRP reinforcement ratios were 1.02% and 0.1% respectively. The mechanical properties of steel reinforcements are listed in Table 1. The mechanical properties of CFRP strips and adhesive are listed in Table 2. Details of the beams are shown in Fig. 1. The specimen manufacturing process included reinforcement cage construction, formwork installation, concrete pouring, specimen maintenance, and prestressed steel strand stretching, as shown in Fig. 2.
2.2. Test scheme A total of three test modes were considered in this process. (1) Flexural capacity test of control beam (PCB0): flexural behaviors of un-strengthened control beam are studied, and yield load Py and ultimate load Pu0 are tested. (2) Flexural capacity test of strengthened beams under loading (MPCB1-MPCB4): to simulate overload damages, the initial damages are performed on the beams by cycle loading through different cycle numbers and overload amplitudes before strengthening. After the initial damage, these beams are strengthened under different sustained loads. They are inversely erected on the loading device and reloaded to different sustained load 0.9 Py or 1.3 Py (MPCB1-MPCB3: 105 kN; MPCB4: 150 kN). The load is kept constant during the CFRP strengthening application. CFRP strips are embedded. Then the load test is continued until the failure after curing period to study the changes in flexural behaviors of beams before and after strengthening with the increase of cycle number and overload amplitude. (3) Flexural capacity test of strengthened beam under unloading (MPCB5): After cycle loading, CFRP strips are embedded for strengthening under unloading, and the load test is implemented after the curing period to investigate the influence of strengthening under loading or unloading on flexural behaviors of beams. The design parameters of the test beams are presented in Table 3. Note: Py is measured yield load value of control beam PCB0. A = Concrete crushing in the compressive zone. B = Wedgeshaped failure of concrete cover in the tensile zone.
2.3. Loading scheme Two-point loading was adopted in this test, and the force was transferred to the test beams through spreader beams. A compression-testing machine was used as loading device in the cycle loading phase, as shown in Fig. 3. Test loading was controlled by the load, and the loading rate was 2 kN/s. Loading device under sustained loading is shown in Fig. 4. The deformation was measured by three linear voltage differential transformers (LVDTs), and the strain was recorded through the DH 3816 static strain test system. The concrete layout of strain gauges is shown in Fig. 5.
267
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275 Table 1 Mechanical properties of steel reinforcement. Type
Diameter (mm)
Yield strength (MPa)
Ultimate strength (MPa)
Elastic modulus (MPa)
Plain bar Plain bar Plain bar Deformed bar Prestressing strand
6 8 10 16 15.2
445.5 406.0 339.2 374.7 –
557.7 551.4 481.8 539.7 1963
2.1 105 2.0 105 2.0 105 2.0 105 2.03 105
Table 2 Mechanical properties of CFRP strips and adhesive. Material
Type specification
Tensile strength (MPa)
Elastic modulus (MPa)
Ductility (%)
CFRP strip adhesive
(1.2 50) mm2 E380
2540 >30
1.65 105 >4.0 103
1.73 >1.3
2.4. Strengthening scheme
Fig. 1. Sectional dimensions and reinforcement of specimens.
After the initial damage, CFRP strips were embedded into the damaged beams. Two grooves both were 10 mm width, 20 mm depth and 3000 mm length, which cut in the undersurfaces of damaged beams. CFRP plates were cut into strips with dimensions of 3000 mm 10 mm 1.2 mm, strain gauges were placed in the central position, and principal adhesive and hardening agent were mixed and blended. Three CFRP strips were bonded together to form a whole and embedded into the groove after the adhesive was solidified for 24 h. Adhesive was injected into the groove to half of its depth, and the strip body was lightly pressed into the groove to facilitate sufficient contact of the adhesive to the CFRP strip to discharge bubbles. Finally, the groove was filled with adhesive. After seven days of curing, the tests were implemented. The strengthening process is shown in Fig. 6.
Fig. 2. Production process of specimens.
268
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
Table 3 Design parameters of test beams. Specimen
Upper limits of overload amplitude (kN)
Cycle number (time)
Sustained load (kN)
Strengthening
Loading method
PCB0 MPCB1 MPCB2 MPCB3 MPCB4 MPCB5
– 0.9 0.9 0.9 1.3 0.9
– 1 50 100 50 50
– 0.9 0.9 0.9 1.3 –
Un-strengthened CFRP strips CFRP strips CFRP strips CFRP strips CFRP strips
refer refer refer refer refer refer
Py Py Py Py Py
Py Py Py Py
Fig. 3. Loading device in the cycle loading phase.
Fig. 4. Loading device of sustained load.
Fig. 5. Arrangement of strain gauges.
Fig. 6. Specimen strengthening process.
to to to to to to
method method method method method method
Failure modes (1) (2) (2) (2) (2) (3)
above above above above above above
A A and B B B A and B A
269
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
3. Test results and analysis 3.1. Overload damage analysis Before specimens MPCB1-MPCB5 were strengthened, the first micro crack appeared in the midspan when the specimen was initially loaded to (20–29%) Pui (P ui is the measured ultimate load of each specimen). The cracks continuously widened and extended upward as the load increased. Several major cracks formed inside the pure bending segment when the load reached the upper limit of the overload amplitude, the crack spacing was uniformly distributed and remained stable. During the cycle loading process, the crack spacing remained stable, and only one or two new micro cracks appeared in the bending shear segment of beams. The damage degrees are summarized in Table 4. As shown in Table 4, the maximum crack widths of the last cycle loading increased compared with that of the 1st cycle loading, which showed that the cycle loading aggravated the damage degree. After cycle loading, the crack width slightly increased with the increase of cycle number and obviously increased with the increase of overload amplitude, which indicated that the damage degrees were aggravated as cycle number and overload amplitude increased. 3.2. Failure process and failure modes (1) Control beam PCB0 When the control beam was loaded to approximately 67 kN (30% P u0 , where Pu0 is the measured ultimate load of PCB0), the first micro crack occurred at midspan. Cracks continuously appeared and extended vertically upward, and the crack spacing was initially large with increasing load. The cracks nearly ran through the concrete flange and crack spacing was uniformly distributed and remained stable as the load increased. When the load reached about 120 kN (50% P u0 ), the crack width evidently increased. The crack width widened continuously as the load increased, and deformation increased continuously. When the load reached 229.4 kN, the beam failed by the concrete crushing in the compressive zone with a cracking sound. (2) Strengthened beams MPCB1-MPCB5 The damaged specimens MPCB1-MPCB4 were inversely erected on the loading device. These specimens were then loaded to sustained load. The cracks that width exceeding 0.15 mm were injected with glue for restoration. The surfaces of concrete cracks were sealed, glue injection ports were reserved to form a sealed cavity, a glue injection tool was used to inject resin glue into cracks, and the glue formed a whole with concrete after solidification to achieve the reinforcement objective. After the cracks were injected with glue, CFRP strips were embedded, and then the strengthened beams were loaded continuously up to failure after the curing period. During the process of loading, the original cracks occurred again, and new cracks were generated when a few cracks
that were injected with glue started cracking. New and old cracks extended, widened, and merged toward the concrete flange as the load increased continuously. The deformation increased with the load. The concrete near the loading points in the compressive zones of specimens MPCB1 and MPCB4 were crushed with a cracking sound nearing failure, and the concrete cover in the tensile zone also underwent wedge-shaped failure. The concrete at midspans and loading points of specimens MPCB2 and MPCB3 debonded from the CFRP plate glue, and the concrete cover in the tensile zone exhibited wedge-shaped failure. The concrete near the loading point in the compressive zone of specimen MPCB5 was crushed. Wedge-shaped failure is mainly due to the simultaneous existence of vertical bending cracks of concrete at the maximum bending moment in the tensile zone and nearby horizontal debonding cracks between the concrete and the plate glue. Hence, the concrete cover in the tensile zone was divided into many triangular or quadrangular wedge blocks. Local spalling failure of concrete cover occurred when one or multiple wedge blocks were divided. The failure modes are shown in Fig. 7.
3.3. Crack analysis The crack development status was observed in the test. The cracking conditions after specimen strengthening are summarized in Table 5. Maximum crack widths under the same-level load were measured to further analyze crack development rules after strengthening, as shown in Table 5. The table shows that under the same-level load, the crack widths of the strengthened beams were evidently smaller than that of the control beam. When the load was 150 kN, the crack widths of specimens MPCB1-MPCB3 widened as the cycle number increased. When the load exceeded 170 kN, the crack widths widened irregular with the increase of cycle number. This is mainly because the normal service load was exceeded, and the development of original cracks can effectively be minimized due to glue injection for restoration, consequently, the crack widths did not exhibit obvious rules. The crack width of specimen MPCB4 was smaller than that of MPCB2, and exceeded that of MPCB2 when nearing failure. This is due to the high sustained load of specimen MPCB4 and crack development was delayed after glue injection for restoration. Compared with specimen MPCB5, the crack widths of specimen MPCB2 under the same-level load were wider because the sustained load period was long during the strengthening process and cracks developed due to plastic deformation. Thus, the sustained load period also manifested certain adverse effects on the crack width of strengthened beams. A comparison with Table 4 shows that the crack spacing inside pure bending segment reduced after strengthening. It is observed that individual original cracks would not crack again when the repair quality of the original cracks is satisfactory. Moreover, the concrete near original cracks weakens and generates new cracks. Glue injection could delay development of cracks, and their crack widths were smaller than the cracks that were not injected with glue. When damaged members are strengthened,
Table 4 Specimen cracking conditions before strengthening. Specimen
Cracking load (kN)
Maximum crack width of the 1st cycle loading (mm)
Maximum crack width of the last cycle loading (mm)
Average crack spacing (mm)
PCB0 MPCB1 MPCB2 MPCB3 MPCB4 MPCB5
67 60 67 52 74 59
– 0.31 0.25 0.35 0.95 0.35
– – 0.35 0.45 1.15 0.4
110 122 122 122 110 138
270
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
Fig. 7. Specimen failure modes.
Table 5 Cracking conditions after strengthening. Specimen
PCB0 MPCB1 MPCB2 MPCB3 MPCB4 MPCB5
Maximum crack width (mm)
Average crack spacing (mm)
110 kN
130 kN
150 kN
170 kN
190 kN
210 kN
230 kN
240 kN
0.5 – – – – 0.2
1.5 – – – – 0.3
2.1 0.6 0.7 0.9 – 0.5
2.5 1.1 1.0 1.3 – 0.8
3.8 1.6 1.2 1.4 0.7 1.0
4.0 1.8 1.65 1.6 1.5 1.3
– 2.4 1.8 2.1 1.8 1.5
– – 1.8 – 2.2 –
original cracks should be treated to prevent the excessive development of original cracks from affecting the normal use of the members. This indicated that strengthening with CFRP strips could effectively improve the crack development. Moreover, strengthening under loading would degrade the degree of improvement, but unloading could eliminate this adverse effect. 3.4. Strain analysis Strain relations between load-concrete, load-steel reinforcement, and load-CFRP strips before and after strengthening were obtained through the test, as shown in Figs. 8 and 9. In Fig. 8, the negative value at the left side is concrete compressive strain, and the positive value at the right side is steel reinforcement tensile strain. Under the effect of the cycle loading before strengthening, as shown in Fig. 8, the load-strains curves of concrete and steel reinforcement were divided into two phases. Two-phase divisions of specimens MPCB2-MPCB5 were evident under the effect of the 1st cycle loading. However, those divisions under the 2nd cycle loading were not apparent with a lowered turning point. The strains of concrete and steel reinforcement increased considerably compared with those in the 1st cycle loading, and increased rapidly under the effect of the first ten times cycle loading. Afterwards, the strains of concrete and steel reinforcement increased slowly with the cycle number. The compressive strain of concrete and the ten-
110 100 100 110 69 122
sile strain of steel reinforcement both moved toward the strain increasing direction and exhibited a sparse to dense tendency. The residual strains of concrete and steel reinforcement increased continuously with the increase of cycle number, and increased evidently in the early phase and stabilized in the later phase. This is because the concrete crack widened further with the increase of cycle number, thereby resulting in the increase of strains and residual strains. During the static load to failure phase after strengthening, the load-strain curves of concrete and steel reinforcement showed that the slopes of the curves in the initial phase were all lower than those of slopes of the 1st cycle loading. It is because the cycle loading resulted in damage of specimen and decline of slope. Afterwards, the growth rate of the concrete and steel reinforcement strains accelerated compared with that in the previous phase. After strengthening, the growth rate of the concrete and steel reinforcement strains slowed down with a gradually rising slope because of the effect of CFRP strips. The growth rate of strain accelerated again after the load reached a certain degree. As for specimen MPCB5 which strengthened without loading, the slopes of the curves after the concrete cracked during the static load to failure phase were greater than that of the slopes in the cycle loading phase because CFRP strips participated in load-bearing during the entire process. As shown in Fig. 9, the CFRP strip strains of specimens MPCB1MPCB3 were nearly the same under the same-level load, which indicated that the cycle number had no effect on the strain of CFRP strips. The sustained load of specimen MPCB4 was higher than that
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
271
Fig. 8. Load–strain relationship curves of concrete and steel reinforcement.
Fig. 9. Load–strain relationships of CFRP strips.
of other specimens, and the CFRP strips did not exhibit any effect until strengthening was performed. Hence, the CFRP strip strain was lower than that of other specimens because of the hysteresis effect. The CFRP strip strain of specimen MPCB2 was higher than that of MPCB5 under the same-level load after steel reinforcement yielded, because that specimen MPCB2 was strengthened under loading while specimen MPCB5 was strengthened without loading. Thus it indicated that the beam strengthened without loading proved to be more effective than the beams strengthened under loading. The growth rate of the CFRP strain in the later phase accelerated, which indicated that the CFRP strip mainly bore the
Fig. 10. Load–deformation relation curves of PCB0.
sectional tensile stress after steel reinforcement yielded. The CFRP strip strains ranged from 9000 10 6 to 12300 10 6, thereby confirming that although CFRP strips did not rupture when failure, their strengthening effect was manifested, and their high strength was fully utilized. 3.5. Deformation analysis (1) Control beam The load-deformation curve of the control beam is shown in Fig. 10. The curve can be divided into three phases. The first phase
272
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
is the elastic working phase, in which the curve presents a linear relation. The second phase is the phase in which concrete cracked until the steel reinforcements yielded. Cracks developed continuously with the increase of load, the concrete in the tensile zone exited from the operation gradually causing a reduction in stiffness. The third phase is the phase in which steel reinforcement yielded to failure, the stiffness was reduced continuously, and the deformation in the entire plastic phase grew rapidly. Under the effect of the prestressed reinforcement, the third phase of the curve was long. The specimen finally failed by concrete crushing in the compressive zone. (2) Strengthened beams under loading The load-deformation curves of MPCB1-MPCB4 before and after strengthening are shown in Fig. 11. During the cycle loading phase before strengthening, the loaddeflection curves were measured and shown in the Fig. 11. The curves of the 1st cycle loading presented a linear trend before cracking. They had a great slope represented to the great stiffness of strengthened beams. When the load reached about (25.6–29.1%) Pui (MPCB1: 25.6%; MPCB2: 25.3%; MPCB3: 25.2%; MPCB4: 29.1%), the turning points appeared and caused a significant degradation in stiffness, resulting in the decrease of the curve slope. During the final static load to failure phase after strengthening, the curves of specimens in the initial phase exhibited a linear trend, and the beams were under the elastic working phase. When the load reached about (12.8–17.9%) P ui (MPCB1: 17.9%; MPCB2: 15.3%; MPCB3: 12.8%; MPCB4: unobvious), the growth rate of deformation at midspan was higher than that in the elastic phase, and turning points appeared in the curves, except for specimen MPCB4. Moreover, the slopes in cycle loading phase were higher than those in the static load to failure phase, which indicated that the cycle loading caused a relative degradation in stiffness. After strengthening under sustained load, turning points appeared in the position of sustained load on the curves, where the growth rate
of the deformation at midspan gradually slowed down and the stiffness increased again. This indicates that the strengthening with CFRP strips effectively improves the stiffness. When the load reached about (49.5–52.1%) Pui (MPCB1: 52.1%; MPCB2: 49.5%; MPCB3: 50.9%), the stiffness was decreased compared with that in the previous phase, the growth rate of deformation increased again, and the bearing capacity of the beams could still increase as the deformation rapidly increased in the later loading phase. For specimen MPCB4, the damage was serious after cycle loading due to the high upper limit of overload amplitude, so the division of its curve in the first and second phases was not evident. The steel reinforcement of MPCB4 has been yielded before strengthening, and the stiffness increased again after strengthening. When the load reached 175 kN, an inflection point appeared in the curve, and the deformation growth rate increased again until the specimen failed. (3) Strengthened beam under unloading The load-deformation curves of specimen MPCB5 before and after strengthening are shown in Fig. 12. Fig. 12 shows that the load-deformation curve of the 1st cycle loading presented a linear trend before cracking. A turning point appeared on the curve when the load reached 22% Pu5 (59 kN). During the final static load to failure phase, the curve was evidently divided into three phases. In the first phase, the curve presented a linear trend, and the beam was under the elastic working phase. The slope was slightly higher than that before strengthening. This is because the CFRP strips did not fully exert their effects in the initial phase of load, so the stiffness increased slightly. When the load reached 16% Pu5 (42 kN), the growth rate of the deformation at midspan was higher than that in the elastic phase and the curve turned. Moreover, the stiffness in this phase evidently increased compared with that before strengthening. Stiffness decreased when the load reached 56% P u5 (148 kN), and the curve presented a linear growth tendency again until failure.
Fig. 11. Load–deformation relation curves of MPCB1-MPCB4.
273
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
deformation of MPCB2 which strengthened under loading was slightly greater than that of MPCB5 which strengthened without loading in the early phase, and was smaller when approached failure in the later phase. 3.6. Bearing capacity analysis The measured values of the yield load and ultimate load of test beams are presented in Table 7. The yield loads of the strengthened beams were generally higher than that of the control beam. The yield loads of specimens MPCB1-MPCB5 increased by 10%, 12%, 13%, 18%, and 28% respectively, compared with that of PCB0. The yield load of specimen MPCB4 was 6% higher than that of MPCB2. This indicates that the yield loads increased with the increase of cycle number and overload amplitude because of the cold hardening of steel reinforcements under the effect of the cycle loading. The yield load of specimen MPCB5 increased by 14% compared with that of MPCB2. It served well to illustrate yield load of strengthened beam without loading was higher than that of strengthened beams under loading as the CFRP strips participated in the entire process. Therefore, the strengthening without loading should be comprehensively considered in practical strengthening engineering. Strengthening could effectively increase the ultimate load of the test beams compared with the control beam. The ultimate loads of MPCB1-MPCB5 increased by 7%, 14%, 12%, 11%, and 15% respectively. The ultimate load of MPCB2 was higher than that of MPCB1 because the concrete experienced crushing failure due to stress concentration at the loading point for MPCB1. The ultimate loads of the strengthened beams decreased slightly within 5% with a minor influence with the increase of cycle number and overload amplitude. The ultimate load of MPCB5 increased slightly compared with that of MPCB2. Thus, the cycle number, overload amplitude, and strengthening under loading slightly influenced the ultimate loads of the strengthened beams such that they could be neglected.
Fig. 12. Load-deformation relation curves of MPCB5.
It is observed that the curve of the 1st cycle loading demonstrated an obvious turning point. The curve gradually stabilized in the first and second phases with the increase of cycle number. This indicated that the cycle loading evidently influenced the stiffness of beams. The specimen deformation conditions before and after strengthening are presented in Table 6. Table 6 shows that the deformations generated by the last cycle loading in the cycle loading phase were all higher than those generated by the 1st cycle loading, which indicated that the beam stiffness was degraded by the cycle loading (except for specimen MPCB3). The deformations of the strengthened beams under the sustained loading were higher than those of the last cycle loading due to the long period of sustained loading. The sustained loading period also exerted certain adverse effects on the deformations of the strengthened beams. During the final static load to failure phase, the deformations of the strengthened beams were smaller than that of the control beam due to the strengthening effects of CFRP strips under the same load. The ultimate displacements of MPCB1, MPCB2, and MPCB3 were 67.9, 64.69 and 57.98 respectively, which indicated that the deformations decreased with the increase of cycle number. It is likely because that the steel reinforcements were enhanced under the cycle loading because of the cold hardening; consequently, the stiffness was enhanced with the decreasing deformation. The deformation of MPCB4 obviously increased compared with that of MPCB2, which indicated that the deformation increased with the overload amplitude. The
3.7. Ductility analysis As for PRC beams, the displacements that corresponded to the turning points of the second phases on load-deformation curves
Table 6 Specimen deformations before and after strengthening. Specimen
PCB0 MPCB1 MPCB2 MPCB3 MPCB4 MPCB5
In the cycle loading phase when load to upper limits of overload amplitude (mm)
In the sustained loading phase (mm)
In the static load to failure phase (mm)
1st
2nd
10th
20th
30th
40th
50th
100th
105 kN
150 kN
220 kN
250 kN
– 10.23 11.35 12.13 24.02 9.88
– – 10.19 11.0 19.08 9.75
– – 11.27 11.3 23.46 10.06
– – 11.62 11.4 24.61 10.13
– – 11.81 11.44 25.33 10.14
– – 12.0 11.56 26.50 10.16
– – 12.02 11.64 27.20 10.17
– – – 11.8 – –
– 13.03 12.4 13.1 – –
– – – – 37.68 –
98.92 53.3 48.59 37.9 66.6 45.47
– – 59.69 51.04 88.2 60.5
Ultimate displacement (mm)
115.11 67.9 64.69 57.98 94.96 84.38
Table 7 Test results of the specimens. Specimen
Yield load (kN)
Yield displacement (mm)
Ultimate load (kN)
Ultimate displacement (mm)
Ductility factor
PCB0 MPCB1 MPCB2 MPCB3 MPCB4 MPCB5
116 127.8 129.3 131 137 148
13.8 15.01 14.93 15.23 27.6 14.74
229.4 245.8 260.7 257.6 254.6 263.8
115.11 67.9 64.69 57.98 94.96 84.38
8.34 4.52 4.33 3.81 3.44 5.72
274
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
were determined as yield displacements. The corresponding displacements that occurred when the load initially decreased while the displacement increased were determined as ultimate displacements. The ductility factor is determined as the ratio of the ultimate displacements to the yield displacements. Ductility factors of the strengthened beams evidently decreased compared with that of the control beam. The concrete values are shown in Table 7. The ductility factors in Table 7 showed that the ductility factors of the strengthened beams were evidently lower than that of the control beam. This indicated that strengthening by NSM CFRP strips affected the overall structural ductility of the strengthened beam. The ductility factors of the strengthened beams gradually decreased with the increase of the cycle number. The yield and ultimate displacements increased with the increase of overload amplitude thereby resulting in the decrease of the ductility factors. The ductility factor of MPCB4 was 3.44, which was lower than that of MPCB2. The ultimate displacement of MPCB5 strengthened without loading increased compared with that of MPCB2 strengthened under loading. Its ductility factor was 5.72, which was higher than that of MPCB2. This indicated that the beam strengthened without loading proved to be better ductile than the beams strengthened under loading. The cycle number, overload amplitude, and strengthening under loading influenced the ductility factors of the strengthened beams. The displacement ductility factors of the strengthened beams under loading decreased with the increase of cycle number and overload amplitude, and were smaller than that of strengthened beam without loading. Hence, unloading should be comprehensively ensured in practical strengthening engineering to satisfy requirements for structural usability after strengthening. 4. Conclusions The following conclusions were obtained based on the test and analysis results in this paper. (1) The failure modes were observed for the tested beam specimens: crushing failure of concrete in the compressive zone, wedge-shaped failure of the partial protective layer of concrete in the tensile zone, or a combination of both modes. (2) Before strengthening, the cycle loading resulted in specimen damage, and the damage degrees were aggravated with the increase of cycle number and overload amplitude. Strengthening with CFRP strips could effectively inhibit crack development and improve the crack properties. Strengthening under loading could also lower the improvement degree, and this adverse effect could be eliminated through unloading. (3) Compared with the control beam, the bearing capacities of the strengthened beams improved. The yield loads increased by 10–28%, while the ultimate loads increased by 7–15%. The cycle number, overload amplitude, and strengthening under loading influenced the yield loads of the strengthened beams. The yield loads increased with the increase of the cycle number and overload amplitude, and the yield load of the strengthened beam under unloading was higher than that of strengthened beams under loading. The cycle number, overload amplitude, and strengthening under loading had limited influences on the ultimate bearing capacity. (4) Before strengthening, the stiffness of beams degraded under cycle loading; the deformation was affected little by cycle number but affected considerably by overload amplitude. Moreover, a sustained loading period exerted certain
adverse effects on the deformations of the beams. After strengthening, the deformations of all strengthened beams under the same load were smaller than that of the control beam due to the strengthened effect of CFRP strips. Within the cycle number in this test, the deformation decreased gradually with the increase of cycle number, but increased with the overload amplitude. The deformations of the strengthened beams under loading were slightly higher than that of the strengthened beam under unloading in the initial phase but lower when approaching failure in the later phase. (5) Cycle number, overload amplitude, and strengthening under loading all affected the ductility factors of the strengthened beams. The displacement ductility factors of the strengthened beams under loading decreased with the increase of cycle number and overload amplitude, and were smaller than that of strengthened beam without loading. Unloading should be comprehensively ensured in practical strengthening engineering to satisfy requirements for structural usability after strengthening. Conflict of interest The authors declare that there is no conflict of interest. Acknowledgments The authors acknowledge the financial support provided by the Natural Science Foundation of China, and the Science and Technology Agency and Education Department of Liaoning Province. The authors are also grateful for the support provided by colleagues and the School of Resources and Civil Engineering Northeastern University during the experimental program. Finally, the authors would like to gratefully acknowledge their friends and the 211 Test and Research Center of Northeastern University for their assistance in testing the specimens. Funding This work was supported by the Natural Science Foundation of China (grant number 51504125) and the Natural Science Foundation of Liaoning Province (grant number 20170540303). References [1] X.Y. Sun, C.K. Huang, G.F. Zhao, Y.Q. Dou, Experimental study of the influence of truck overloads on the flexural performance of bridge members, China Civ. Eng. J. 38 (6) (2005) 35–40. [2] X.Y. Sun, H.L. Wang, C.K. Huang, Experimental investigation on flexural performance of bridge members under overloading, J. Zhejiang Univ. (Eng. Sci.) 42 (1) (2008) 152–156. 163. [3] S.H. Li, H.W. Jiang, Influence law of different weight limit levels on highway bridge safety, China J. Highway Transp. 29 (3) (2016) 82–88. 115. [4] H. Peng, J.R. Zhang, C.S. Cai, Y. Liu, An experimental study on reinforced concrete beams strengthened with prestressed near surface mounted CFRP strips, Eng. Struct. 79 (2014) 222–233. [5] J.G. Teng, L. De Lorenzis, B. Wang, R. Li, Debonding failures of RC beams strengthened with near surface mounted CFRP strips, J. Compos. Constr., ASCE 10 (2006) 92–105. [6] L. Anania, A. Badala, G. Failla, Increasing the flexural performance of RC beams strengthened with CFRP materials, Constr. Build. Mater. 19 (2005) 55–61. [7] M. Emara, C. Barris, M. Baena, L. Torres, J. Barros, Bond behavior of NSM CFRP laminates in concrete under sustained loading, Constr. Build. Mater. 177 (2018) 237–246. [8] M.I. Kabir, M. Subhani, R. Shrestha, B. Samali, Experimental and theoretical analysis of severely damaged concrete beams strengthened with CFRP, Constr. Build. Mater. 178 (2018) 161–174. [9] S.S. Zhang, T. Yu, Analytical solution for interaction forces in beams strengthened with near-surface mounted round bars, Constr. Build. Mater. 106 (2016) 189–197.
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275 [10] American Concrete Institute (ACI), Guide for the design and construction of externally bonded FRP systems for strengthening concrete structures, ACI-440 2R Farmington Hills, Mich, 2008, pp. 76. [11] J. Michels, M. Staskiewicz, C. Czaderski, R. Kotynia, Y.E. Harmanci, M. Motavalli, Prestressed CFRP strips for concrete bridge girder retrofitting: application and static loading test, J. Bridge Eng. 21 (5) (2016) 04016003. [12] Y.J. Kim, R. Gordon Wight, M.F. Green, Flexural strengthening of RC beams with prestressed CFRP sheets: using nonmetallic anchor systems, J. Compos. Constr. 12 (1) (2008) 44–52. [13] L. De Lorenzis, J.G. Teng, Near-surface mounted FRP reinforcement: an emerging technique for strengthening structures, Compos. Part B 38 (2) (2007) 119–143. [14] R. El-Hacha, S.H. Rizkalla, Near-surface-mounted fiber-reinforced polymer reinforcements for flexural strengthening of concrete structures, Struct. J. 101 (5) (2004) 717–726. [15] S.S. Zhang, J.G. Teng, T. Yu, Bond-slip model for CFRP strips near-surface mounted to concrete, Eng. Struct. 56 (2013) 945–953. [16] J.A.O. Barros, R. Kotynia, Possibilities and challenges of NSM for the flexural strengthening of RC structures, in: Fourth international conference on FRP composites in civil engineering, Zurich, Switzerland, 2008, pp. 6. [17] R. Kotynia, Analysis of reinforced concrete beams strengthened with near surface mounted FRP reinforcement, Arch. Civil Eng. 52 (2) (2006) 305–317. [18] T. Oshiro, S. Hamada, Structural performance and bending test of deteriorated reinforced concrete bridge, ACI SP. 88 (1985) 39–58. [19] E.G. Burdette, D. W. Goodpasture, Correlation of bridge load capacity estimates with test data. National Cooperative Highway Research Program report 306, USA, 1988.
275
[20] R.A. Hawileh, H.A. Rasheed, J.A. Abdalla, A.K. Al-Tamimi, Behavior of reinforced concrete beams strengthened with externally bonded hybrid fiber reinforced polymer systems, Mater. Des. 53 (2014) 972–982. [21] J. Tan, Experiment and analysis on flexural behavior of prestressed concrete beam reinforced with CFRP sheet, Harbin Institute of Technology, 2008. [22] A.M. Khalifa, Flexural performance of RC beams strengthened with near surface mounted CFRP strips, Alexandria Eng. J. 55 (2) (2016) 1497–1505. [23] Y.C. You, K.S. Choi, J.H. Kim, An experimental investigation on flexural behavior of RC beams strengthened with prestressed CFRP strips using a durable anchorage system, Compos. Part B 43 (2012) 3026–3036. [24] M. Badawi, K. Soudki, Flexural strengthening of RC beams with prestressed NSM CFRP rods-Experimental and analytical investigation, Constr. Build. Mater. 23 (10) (2009) 3292–3300. [25] X.Y. Sun, C.K. Huang, Test on flexural performance of RC bridge member after overload reinforced with externally bonded FRP, China J. Highway Transp. 19 (4) (2006) 82–87. [26] X.Y. Sun, C. Xu, H.L. Huang, J.F. Chu, Overloading simulation fatigue experiment of bridge member with and without CFRP reinforcement, J. Zhejiang Univ. (Eng. Sci.) 46 (9) (2012) 1585–1591. [27] R.X. Wang, K. Wang, Z.W. Cui, Mechanical characteristics analysis of overload pre-cracked RC beams strengthened with CFRP and steel sheet, J. Hebei Univ. Technol. 38 (6) (2009) 85–88. [28] L.R.F.D. Aashto, Bridge design specifications, SI Units, third ed., AASHTO, Washington, D.C., 2004.
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Experimental research on flexural behaviors of damaged PRC beams strengthened with NSM CFRP strips Jia Yang a,b, Lianguang Wang a,⇑ a b
School of Resources and Civil Engineering, Northeastern University, Shenyang 110819, China College of Architectural and Civil Engineering, Shenyang University, Shenyang 110044, China
h i g h l i g h t s PRC beams strengthened with NSM CFRP strips were tested under overloading. Influence of cycle number, overload amplitude, and strengthening under loading was investigated. The above three variables had limited influence on the ultimate bearing capacity. The above three variables affected stiffness and ductility of the strengthened beams.
a r t i c l e
i n f o
Article history: Received 28 November 2017 Received in revised form 17 September 2018 Accepted 17 September 2018
Keywords: Flexural behaviors Experimental research PRC beam NSM CFRP strips Damage
a b s t r a c t This paper investigates the flexural behaviors of the damaged prestressed reinforced concrete (PRC) beams strengthened with near surface mounted (NSM) carbon fiber-reinforced polymers (CFRP) strips under overloading. A bending test was carried out to evaluate the effects of three variables: the cycle number, overload amplitude and strengthening under loading or unloading. Different damage degrees were mainly controlled by cycle number and overload amplitude. The test results showed that strengthening with NSM CFRP strips could effectively inhibit crack development. Compared with the unstrengthened beam, the flexural load-carrying and stiffness of the strengthened beams were enhanced evidently. With the increase of cycle number and overload amplitude, the yield loads increased. The yield load of strengthened beam under unloading was higher than that of strengthened beams under loading. These three variables had limited influence on the ultimate bearing capacity. The deformations of the strengthened beams were smaller than that of the control beam, and decreased with the increase of cycle number, but increased with the increase of overload amplitude. The deformations of the strengthened beams under loading were slightly greater than that of the strengthened beam under unloading in the early phase but smaller in the later phase. The ductility of strengthened beams was evidently reduced compared with that of the control beam. The displacement ductility factors of the strengthened beams under loading decreased with the increase of cycle number and overload amplitude, and were smaller than that of strengthened beam under unloading. Ó 2018 Elsevier Ltd. All rights reserved.
1. Introduction With the increasing scaling-up and heavy-load tendencies of vehicles, the actual traffic volume of a few in-service bridges has exceeded the designed traffic volume. Moreover, the load design standards in new and old bridge design specifications are not uniform, thereby resulting in overload. Operation under an overload situation poses serious harm to the structure of bridge, thus occasionally causing bridge collapse accidents. Moreover, numerous ⇑ Corresponding author. E-mail address: [email protected] (L. Wang). https://doi.org/10.1016/j.conbuildmat.2018.09.109 0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
bridges frequently experience overload [1–3]. Although overloaded bridges do not collapse, they pose severe potential structural hazards. The load-bearing properties of the bridges damaged by overloads have become the focus of researchers and engineering personnel. Strengthening of damaged bridges is necessary to improve the bearing capacities and stiffness, and lengthen their service life. Currently, fiber-reinforced polymers (FRP) are commonly used for strengthening [4–8]. The most common FRP strengthening method of reinforced concrete (RC) beams is to externally bond (EB) FRP laminates onto the soffit of a beam [9]. However, the effectiveness of this system is limited by the most common debonding failure mode, which occurs at an effective
266
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
strain much lower than the ultimate strain achievable by the FRP composite materials [10]. In order to maximize the utilization of the FRP materials, the failure mode caused by debonding should be overcome. At present, strengthening with prestressed FRP material and NSM FRP technique have been introduced. The EB prestressed CFRP strips or sheets using nonmetallic necessary anchor systems can increase the load-carrying capacity [11,12]. The NSM FRP strengthening technique has attracted significant attention worldwide as one of the promising new techniques for structural strengthening [8]. NSM FRP technique involves cutting grooves in the concrete cover, filling the grooves with adhesive and embedding FRP bars or strips into the grooves [13,14]. NSM FRP has several advantages over EB FRP, including better protection of the FRP and a stronger bond between FRP and concrete [15]. The NSM technique can increase load carrying capacity significantly and improve the stiffness of the beams. This technique is more effective than EB CFRP reinforcement, provides higher strength capacity and higher utilization of the strips using the same material with the same axial stiffness [14,16–17]. At present, a few studies conducted in China and abroad deal with the mechanical properties of RC structure under overloading [1–3,18–19]. Numerous studies have been conducted on FRPstrengthened RC and PRC structures [20–21]; a relatively greater number of investigations used the NSM CFRP strips to strengthen RC beams [22–24], but only few studies about NSM CFRP strips strengthened PRC beam. For overload damage, X. Y. Sun and R. X. Wang [25–27] studied the mechanical properties of damaged RC beams strengthened with CFRP strips and steel plates under overloading, which results showed that the strengthening effect was mainly influenced by overload amplitude, cycle number, strengthening method and reinforcement ratio, and overload amplitude and cycle number had considerable effects on the service life of beams. However, studies on the influence of overload on PRC beams are seldom. The most direct approach is to study the flexural behaviors of PRC structure strengthened with CFRP strips under overloading to perform a destructive load test. However, considerable bridge collapse risk is apparent when real vehicles are used to conduct the overload test, and test objects are hardly available (bridges may experience severe structural damage, thereby hindering them from continuing service). As stipulated by the American Association of State Highway and Transportation Officials [28], the internal reinforcement stress of bridge member in service ability limit state shall not exceed 60% of the yield stress of the reinforcement, and any value that exceeds this limit shall be considered overload. Hence, in this study, 0.9 and 1.3 times yield load of the control beam are considered upper limits of overload amplitude, and 0.1 times yield load is the lower limit. Cyclic loading is implemented for test beams to simulate the overload damage state. Normally, the occurrence frequency of overload amplitude is low; hence, the cycle numbers are 1, 50, and 100 times. In addition, components that require strengthening in engineering are generally under the loading state. Therefore, strengthening tests with NSM CFRP strips are conducted for damaged PRC beams under overloading to investigate changes in bearing capacity and stiffness with the increase of cycle number and overload amplitude before and after strengthening and the influence of strengthening under loading on flexural behaviors.
2. Experimental program 2.1. Specimen design and production A total of six specimens, which were all unbonded prestressed concrete T-beams, were designed in the test; one was the control beam, and five were strengthened beams with NSM CFRP strips.
All beams had a length of 4.5 m and a net span of 4.3 m. The measured compressive strength of concrete cube was 32.5 MPa at 28 days. The beams were reinforced by two HRB-335 16 mm diameter longitudinal steel bars, four HPB-300 10 mm diameter steel bars, 6 mm diameter stirrups with spacing of 200 mm in the concrete flange, 8 mm diameter stirrups with the spacing of 120 mm in the concrete web, and two 15.2 mm diameter low relaxation seven-wire prestressing steel strands. The steel reinforcement and the CFRP reinforcement ratios were 1.02% and 0.1% respectively. The mechanical properties of steel reinforcements are listed in Table 1. The mechanical properties of CFRP strips and adhesive are listed in Table 2. Details of the beams are shown in Fig. 1. The specimen manufacturing process included reinforcement cage construction, formwork installation, concrete pouring, specimen maintenance, and prestressed steel strand stretching, as shown in Fig. 2.
2.2. Test scheme A total of three test modes were considered in this process. (1) Flexural capacity test of control beam (PCB0): flexural behaviors of un-strengthened control beam are studied, and yield load Py and ultimate load Pu0 are tested. (2) Flexural capacity test of strengthened beams under loading (MPCB1-MPCB4): to simulate overload damages, the initial damages are performed on the beams by cycle loading through different cycle numbers and overload amplitudes before strengthening. After the initial damage, these beams are strengthened under different sustained loads. They are inversely erected on the loading device and reloaded to different sustained load 0.9 Py or 1.3 Py (MPCB1-MPCB3: 105 kN; MPCB4: 150 kN). The load is kept constant during the CFRP strengthening application. CFRP strips are embedded. Then the load test is continued until the failure after curing period to study the changes in flexural behaviors of beams before and after strengthening with the increase of cycle number and overload amplitude. (3) Flexural capacity test of strengthened beam under unloading (MPCB5): After cycle loading, CFRP strips are embedded for strengthening under unloading, and the load test is implemented after the curing period to investigate the influence of strengthening under loading or unloading on flexural behaviors of beams. The design parameters of the test beams are presented in Table 3. Note: Py is measured yield load value of control beam PCB0. A = Concrete crushing in the compressive zone. B = Wedgeshaped failure of concrete cover in the tensile zone.
2.3. Loading scheme Two-point loading was adopted in this test, and the force was transferred to the test beams through spreader beams. A compression-testing machine was used as loading device in the cycle loading phase, as shown in Fig. 3. Test loading was controlled by the load, and the loading rate was 2 kN/s. Loading device under sustained loading is shown in Fig. 4. The deformation was measured by three linear voltage differential transformers (LVDTs), and the strain was recorded through the DH 3816 static strain test system. The concrete layout of strain gauges is shown in Fig. 5.
267
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275 Table 1 Mechanical properties of steel reinforcement. Type
Diameter (mm)
Yield strength (MPa)
Ultimate strength (MPa)
Elastic modulus (MPa)
Plain bar Plain bar Plain bar Deformed bar Prestressing strand
6 8 10 16 15.2
445.5 406.0 339.2 374.7 –
557.7 551.4 481.8 539.7 1963
2.1 105 2.0 105 2.0 105 2.0 105 2.03 105
Table 2 Mechanical properties of CFRP strips and adhesive. Material
Type specification
Tensile strength (MPa)
Elastic modulus (MPa)
Ductility (%)
CFRP strip adhesive
(1.2 50) mm2 E380
2540 >30
1.65 105 >4.0 103
1.73 >1.3
2.4. Strengthening scheme
Fig. 1. Sectional dimensions and reinforcement of specimens.
After the initial damage, CFRP strips were embedded into the damaged beams. Two grooves both were 10 mm width, 20 mm depth and 3000 mm length, which cut in the undersurfaces of damaged beams. CFRP plates were cut into strips with dimensions of 3000 mm 10 mm 1.2 mm, strain gauges were placed in the central position, and principal adhesive and hardening agent were mixed and blended. Three CFRP strips were bonded together to form a whole and embedded into the groove after the adhesive was solidified for 24 h. Adhesive was injected into the groove to half of its depth, and the strip body was lightly pressed into the groove to facilitate sufficient contact of the adhesive to the CFRP strip to discharge bubbles. Finally, the groove was filled with adhesive. After seven days of curing, the tests were implemented. The strengthening process is shown in Fig. 6.
Fig. 2. Production process of specimens.
268
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
Table 3 Design parameters of test beams. Specimen
Upper limits of overload amplitude (kN)
Cycle number (time)
Sustained load (kN)
Strengthening
Loading method
PCB0 MPCB1 MPCB2 MPCB3 MPCB4 MPCB5
– 0.9 0.9 0.9 1.3 0.9
– 1 50 100 50 50
– 0.9 0.9 0.9 1.3 –
Un-strengthened CFRP strips CFRP strips CFRP strips CFRP strips CFRP strips
refer refer refer refer refer refer
Py Py Py Py Py
Py Py Py Py
Fig. 3. Loading device in the cycle loading phase.
Fig. 4. Loading device of sustained load.
Fig. 5. Arrangement of strain gauges.
Fig. 6. Specimen strengthening process.
to to to to to to
method method method method method method
Failure modes (1) (2) (2) (2) (2) (3)
above above above above above above
A A and B B B A and B A
269
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
3. Test results and analysis 3.1. Overload damage analysis Before specimens MPCB1-MPCB5 were strengthened, the first micro crack appeared in the midspan when the specimen was initially loaded to (20–29%) Pui (P ui is the measured ultimate load of each specimen). The cracks continuously widened and extended upward as the load increased. Several major cracks formed inside the pure bending segment when the load reached the upper limit of the overload amplitude, the crack spacing was uniformly distributed and remained stable. During the cycle loading process, the crack spacing remained stable, and only one or two new micro cracks appeared in the bending shear segment of beams. The damage degrees are summarized in Table 4. As shown in Table 4, the maximum crack widths of the last cycle loading increased compared with that of the 1st cycle loading, which showed that the cycle loading aggravated the damage degree. After cycle loading, the crack width slightly increased with the increase of cycle number and obviously increased with the increase of overload amplitude, which indicated that the damage degrees were aggravated as cycle number and overload amplitude increased. 3.2. Failure process and failure modes (1) Control beam PCB0 When the control beam was loaded to approximately 67 kN (30% P u0 , where Pu0 is the measured ultimate load of PCB0), the first micro crack occurred at midspan. Cracks continuously appeared and extended vertically upward, and the crack spacing was initially large with increasing load. The cracks nearly ran through the concrete flange and crack spacing was uniformly distributed and remained stable as the load increased. When the load reached about 120 kN (50% P u0 ), the crack width evidently increased. The crack width widened continuously as the load increased, and deformation increased continuously. When the load reached 229.4 kN, the beam failed by the concrete crushing in the compressive zone with a cracking sound. (2) Strengthened beams MPCB1-MPCB5 The damaged specimens MPCB1-MPCB4 were inversely erected on the loading device. These specimens were then loaded to sustained load. The cracks that width exceeding 0.15 mm were injected with glue for restoration. The surfaces of concrete cracks were sealed, glue injection ports were reserved to form a sealed cavity, a glue injection tool was used to inject resin glue into cracks, and the glue formed a whole with concrete after solidification to achieve the reinforcement objective. After the cracks were injected with glue, CFRP strips were embedded, and then the strengthened beams were loaded continuously up to failure after the curing period. During the process of loading, the original cracks occurred again, and new cracks were generated when a few cracks
that were injected with glue started cracking. New and old cracks extended, widened, and merged toward the concrete flange as the load increased continuously. The deformation increased with the load. The concrete near the loading points in the compressive zones of specimens MPCB1 and MPCB4 were crushed with a cracking sound nearing failure, and the concrete cover in the tensile zone also underwent wedge-shaped failure. The concrete at midspans and loading points of specimens MPCB2 and MPCB3 debonded from the CFRP plate glue, and the concrete cover in the tensile zone exhibited wedge-shaped failure. The concrete near the loading point in the compressive zone of specimen MPCB5 was crushed. Wedge-shaped failure is mainly due to the simultaneous existence of vertical bending cracks of concrete at the maximum bending moment in the tensile zone and nearby horizontal debonding cracks between the concrete and the plate glue. Hence, the concrete cover in the tensile zone was divided into many triangular or quadrangular wedge blocks. Local spalling failure of concrete cover occurred when one or multiple wedge blocks were divided. The failure modes are shown in Fig. 7.
3.3. Crack analysis The crack development status was observed in the test. The cracking conditions after specimen strengthening are summarized in Table 5. Maximum crack widths under the same-level load were measured to further analyze crack development rules after strengthening, as shown in Table 5. The table shows that under the same-level load, the crack widths of the strengthened beams were evidently smaller than that of the control beam. When the load was 150 kN, the crack widths of specimens MPCB1-MPCB3 widened as the cycle number increased. When the load exceeded 170 kN, the crack widths widened irregular with the increase of cycle number. This is mainly because the normal service load was exceeded, and the development of original cracks can effectively be minimized due to glue injection for restoration, consequently, the crack widths did not exhibit obvious rules. The crack width of specimen MPCB4 was smaller than that of MPCB2, and exceeded that of MPCB2 when nearing failure. This is due to the high sustained load of specimen MPCB4 and crack development was delayed after glue injection for restoration. Compared with specimen MPCB5, the crack widths of specimen MPCB2 under the same-level load were wider because the sustained load period was long during the strengthening process and cracks developed due to plastic deformation. Thus, the sustained load period also manifested certain adverse effects on the crack width of strengthened beams. A comparison with Table 4 shows that the crack spacing inside pure bending segment reduced after strengthening. It is observed that individual original cracks would not crack again when the repair quality of the original cracks is satisfactory. Moreover, the concrete near original cracks weakens and generates new cracks. Glue injection could delay development of cracks, and their crack widths were smaller than the cracks that were not injected with glue. When damaged members are strengthened,
Table 4 Specimen cracking conditions before strengthening. Specimen
Cracking load (kN)
Maximum crack width of the 1st cycle loading (mm)
Maximum crack width of the last cycle loading (mm)
Average crack spacing (mm)
PCB0 MPCB1 MPCB2 MPCB3 MPCB4 MPCB5
67 60 67 52 74 59
– 0.31 0.25 0.35 0.95 0.35
– – 0.35 0.45 1.15 0.4
110 122 122 122 110 138
270
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
Fig. 7. Specimen failure modes.
Table 5 Cracking conditions after strengthening. Specimen
PCB0 MPCB1 MPCB2 MPCB3 MPCB4 MPCB5
Maximum crack width (mm)
Average crack spacing (mm)
110 kN
130 kN
150 kN
170 kN
190 kN
210 kN
230 kN
240 kN
0.5 – – – – 0.2
1.5 – – – – 0.3
2.1 0.6 0.7 0.9 – 0.5
2.5 1.1 1.0 1.3 – 0.8
3.8 1.6 1.2 1.4 0.7 1.0
4.0 1.8 1.65 1.6 1.5 1.3
– 2.4 1.8 2.1 1.8 1.5
– – 1.8 – 2.2 –
original cracks should be treated to prevent the excessive development of original cracks from affecting the normal use of the members. This indicated that strengthening with CFRP strips could effectively improve the crack development. Moreover, strengthening under loading would degrade the degree of improvement, but unloading could eliminate this adverse effect. 3.4. Strain analysis Strain relations between load-concrete, load-steel reinforcement, and load-CFRP strips before and after strengthening were obtained through the test, as shown in Figs. 8 and 9. In Fig. 8, the negative value at the left side is concrete compressive strain, and the positive value at the right side is steel reinforcement tensile strain. Under the effect of the cycle loading before strengthening, as shown in Fig. 8, the load-strains curves of concrete and steel reinforcement were divided into two phases. Two-phase divisions of specimens MPCB2-MPCB5 were evident under the effect of the 1st cycle loading. However, those divisions under the 2nd cycle loading were not apparent with a lowered turning point. The strains of concrete and steel reinforcement increased considerably compared with those in the 1st cycle loading, and increased rapidly under the effect of the first ten times cycle loading. Afterwards, the strains of concrete and steel reinforcement increased slowly with the cycle number. The compressive strain of concrete and the ten-
110 100 100 110 69 122
sile strain of steel reinforcement both moved toward the strain increasing direction and exhibited a sparse to dense tendency. The residual strains of concrete and steel reinforcement increased continuously with the increase of cycle number, and increased evidently in the early phase and stabilized in the later phase. This is because the concrete crack widened further with the increase of cycle number, thereby resulting in the increase of strains and residual strains. During the static load to failure phase after strengthening, the load-strain curves of concrete and steel reinforcement showed that the slopes of the curves in the initial phase were all lower than those of slopes of the 1st cycle loading. It is because the cycle loading resulted in damage of specimen and decline of slope. Afterwards, the growth rate of the concrete and steel reinforcement strains accelerated compared with that in the previous phase. After strengthening, the growth rate of the concrete and steel reinforcement strains slowed down with a gradually rising slope because of the effect of CFRP strips. The growth rate of strain accelerated again after the load reached a certain degree. As for specimen MPCB5 which strengthened without loading, the slopes of the curves after the concrete cracked during the static load to failure phase were greater than that of the slopes in the cycle loading phase because CFRP strips participated in load-bearing during the entire process. As shown in Fig. 9, the CFRP strip strains of specimens MPCB1MPCB3 were nearly the same under the same-level load, which indicated that the cycle number had no effect on the strain of CFRP strips. The sustained load of specimen MPCB4 was higher than that
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
271
Fig. 8. Load–strain relationship curves of concrete and steel reinforcement.
Fig. 9. Load–strain relationships of CFRP strips.
of other specimens, and the CFRP strips did not exhibit any effect until strengthening was performed. Hence, the CFRP strip strain was lower than that of other specimens because of the hysteresis effect. The CFRP strip strain of specimen MPCB2 was higher than that of MPCB5 under the same-level load after steel reinforcement yielded, because that specimen MPCB2 was strengthened under loading while specimen MPCB5 was strengthened without loading. Thus it indicated that the beam strengthened without loading proved to be more effective than the beams strengthened under loading. The growth rate of the CFRP strain in the later phase accelerated, which indicated that the CFRP strip mainly bore the
Fig. 10. Load–deformation relation curves of PCB0.
sectional tensile stress after steel reinforcement yielded. The CFRP strip strains ranged from 9000 10 6 to 12300 10 6, thereby confirming that although CFRP strips did not rupture when failure, their strengthening effect was manifested, and their high strength was fully utilized. 3.5. Deformation analysis (1) Control beam The load-deformation curve of the control beam is shown in Fig. 10. The curve can be divided into three phases. The first phase
272
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
is the elastic working phase, in which the curve presents a linear relation. The second phase is the phase in which concrete cracked until the steel reinforcements yielded. Cracks developed continuously with the increase of load, the concrete in the tensile zone exited from the operation gradually causing a reduction in stiffness. The third phase is the phase in which steel reinforcement yielded to failure, the stiffness was reduced continuously, and the deformation in the entire plastic phase grew rapidly. Under the effect of the prestressed reinforcement, the third phase of the curve was long. The specimen finally failed by concrete crushing in the compressive zone. (2) Strengthened beams under loading The load-deformation curves of MPCB1-MPCB4 before and after strengthening are shown in Fig. 11. During the cycle loading phase before strengthening, the loaddeflection curves were measured and shown in the Fig. 11. The curves of the 1st cycle loading presented a linear trend before cracking. They had a great slope represented to the great stiffness of strengthened beams. When the load reached about (25.6–29.1%) Pui (MPCB1: 25.6%; MPCB2: 25.3%; MPCB3: 25.2%; MPCB4: 29.1%), the turning points appeared and caused a significant degradation in stiffness, resulting in the decrease of the curve slope. During the final static load to failure phase after strengthening, the curves of specimens in the initial phase exhibited a linear trend, and the beams were under the elastic working phase. When the load reached about (12.8–17.9%) P ui (MPCB1: 17.9%; MPCB2: 15.3%; MPCB3: 12.8%; MPCB4: unobvious), the growth rate of deformation at midspan was higher than that in the elastic phase, and turning points appeared in the curves, except for specimen MPCB4. Moreover, the slopes in cycle loading phase were higher than those in the static load to failure phase, which indicated that the cycle loading caused a relative degradation in stiffness. After strengthening under sustained load, turning points appeared in the position of sustained load on the curves, where the growth rate
of the deformation at midspan gradually slowed down and the stiffness increased again. This indicates that the strengthening with CFRP strips effectively improves the stiffness. When the load reached about (49.5–52.1%) Pui (MPCB1: 52.1%; MPCB2: 49.5%; MPCB3: 50.9%), the stiffness was decreased compared with that in the previous phase, the growth rate of deformation increased again, and the bearing capacity of the beams could still increase as the deformation rapidly increased in the later loading phase. For specimen MPCB4, the damage was serious after cycle loading due to the high upper limit of overload amplitude, so the division of its curve in the first and second phases was not evident. The steel reinforcement of MPCB4 has been yielded before strengthening, and the stiffness increased again after strengthening. When the load reached 175 kN, an inflection point appeared in the curve, and the deformation growth rate increased again until the specimen failed. (3) Strengthened beam under unloading The load-deformation curves of specimen MPCB5 before and after strengthening are shown in Fig. 12. Fig. 12 shows that the load-deformation curve of the 1st cycle loading presented a linear trend before cracking. A turning point appeared on the curve when the load reached 22% Pu5 (59 kN). During the final static load to failure phase, the curve was evidently divided into three phases. In the first phase, the curve presented a linear trend, and the beam was under the elastic working phase. The slope was slightly higher than that before strengthening. This is because the CFRP strips did not fully exert their effects in the initial phase of load, so the stiffness increased slightly. When the load reached 16% Pu5 (42 kN), the growth rate of the deformation at midspan was higher than that in the elastic phase and the curve turned. Moreover, the stiffness in this phase evidently increased compared with that before strengthening. Stiffness decreased when the load reached 56% P u5 (148 kN), and the curve presented a linear growth tendency again until failure.
Fig. 11. Load–deformation relation curves of MPCB1-MPCB4.
273
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
deformation of MPCB2 which strengthened under loading was slightly greater than that of MPCB5 which strengthened without loading in the early phase, and was smaller when approached failure in the later phase. 3.6. Bearing capacity analysis The measured values of the yield load and ultimate load of test beams are presented in Table 7. The yield loads of the strengthened beams were generally higher than that of the control beam. The yield loads of specimens MPCB1-MPCB5 increased by 10%, 12%, 13%, 18%, and 28% respectively, compared with that of PCB0. The yield load of specimen MPCB4 was 6% higher than that of MPCB2. This indicates that the yield loads increased with the increase of cycle number and overload amplitude because of the cold hardening of steel reinforcements under the effect of the cycle loading. The yield load of specimen MPCB5 increased by 14% compared with that of MPCB2. It served well to illustrate yield load of strengthened beam without loading was higher than that of strengthened beams under loading as the CFRP strips participated in the entire process. Therefore, the strengthening without loading should be comprehensively considered in practical strengthening engineering. Strengthening could effectively increase the ultimate load of the test beams compared with the control beam. The ultimate loads of MPCB1-MPCB5 increased by 7%, 14%, 12%, 11%, and 15% respectively. The ultimate load of MPCB2 was higher than that of MPCB1 because the concrete experienced crushing failure due to stress concentration at the loading point for MPCB1. The ultimate loads of the strengthened beams decreased slightly within 5% with a minor influence with the increase of cycle number and overload amplitude. The ultimate load of MPCB5 increased slightly compared with that of MPCB2. Thus, the cycle number, overload amplitude, and strengthening under loading slightly influenced the ultimate loads of the strengthened beams such that they could be neglected.
Fig. 12. Load-deformation relation curves of MPCB5.
It is observed that the curve of the 1st cycle loading demonstrated an obvious turning point. The curve gradually stabilized in the first and second phases with the increase of cycle number. This indicated that the cycle loading evidently influenced the stiffness of beams. The specimen deformation conditions before and after strengthening are presented in Table 6. Table 6 shows that the deformations generated by the last cycle loading in the cycle loading phase were all higher than those generated by the 1st cycle loading, which indicated that the beam stiffness was degraded by the cycle loading (except for specimen MPCB3). The deformations of the strengthened beams under the sustained loading were higher than those of the last cycle loading due to the long period of sustained loading. The sustained loading period also exerted certain adverse effects on the deformations of the strengthened beams. During the final static load to failure phase, the deformations of the strengthened beams were smaller than that of the control beam due to the strengthening effects of CFRP strips under the same load. The ultimate displacements of MPCB1, MPCB2, and MPCB3 were 67.9, 64.69 and 57.98 respectively, which indicated that the deformations decreased with the increase of cycle number. It is likely because that the steel reinforcements were enhanced under the cycle loading because of the cold hardening; consequently, the stiffness was enhanced with the decreasing deformation. The deformation of MPCB4 obviously increased compared with that of MPCB2, which indicated that the deformation increased with the overload amplitude. The
3.7. Ductility analysis As for PRC beams, the displacements that corresponded to the turning points of the second phases on load-deformation curves
Table 6 Specimen deformations before and after strengthening. Specimen
PCB0 MPCB1 MPCB2 MPCB3 MPCB4 MPCB5
In the cycle loading phase when load to upper limits of overload amplitude (mm)
In the sustained loading phase (mm)
In the static load to failure phase (mm)
1st
2nd
10th
20th
30th
40th
50th
100th
105 kN
150 kN
220 kN
250 kN
– 10.23 11.35 12.13 24.02 9.88
– – 10.19 11.0 19.08 9.75
– – 11.27 11.3 23.46 10.06
– – 11.62 11.4 24.61 10.13
– – 11.81 11.44 25.33 10.14
– – 12.0 11.56 26.50 10.16
– – 12.02 11.64 27.20 10.17
– – – 11.8 – –
– 13.03 12.4 13.1 – –
– – – – 37.68 –
98.92 53.3 48.59 37.9 66.6 45.47
– – 59.69 51.04 88.2 60.5
Ultimate displacement (mm)
115.11 67.9 64.69 57.98 94.96 84.38
Table 7 Test results of the specimens. Specimen
Yield load (kN)
Yield displacement (mm)
Ultimate load (kN)
Ultimate displacement (mm)
Ductility factor
PCB0 MPCB1 MPCB2 MPCB3 MPCB4 MPCB5
116 127.8 129.3 131 137 148
13.8 15.01 14.93 15.23 27.6 14.74
229.4 245.8 260.7 257.6 254.6 263.8
115.11 67.9 64.69 57.98 94.96 84.38
8.34 4.52 4.33 3.81 3.44 5.72
274
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275
were determined as yield displacements. The corresponding displacements that occurred when the load initially decreased while the displacement increased were determined as ultimate displacements. The ductility factor is determined as the ratio of the ultimate displacements to the yield displacements. Ductility factors of the strengthened beams evidently decreased compared with that of the control beam. The concrete values are shown in Table 7. The ductility factors in Table 7 showed that the ductility factors of the strengthened beams were evidently lower than that of the control beam. This indicated that strengthening by NSM CFRP strips affected the overall structural ductility of the strengthened beam. The ductility factors of the strengthened beams gradually decreased with the increase of the cycle number. The yield and ultimate displacements increased with the increase of overload amplitude thereby resulting in the decrease of the ductility factors. The ductility factor of MPCB4 was 3.44, which was lower than that of MPCB2. The ultimate displacement of MPCB5 strengthened without loading increased compared with that of MPCB2 strengthened under loading. Its ductility factor was 5.72, which was higher than that of MPCB2. This indicated that the beam strengthened without loading proved to be better ductile than the beams strengthened under loading. The cycle number, overload amplitude, and strengthening under loading influenced the ductility factors of the strengthened beams. The displacement ductility factors of the strengthened beams under loading decreased with the increase of cycle number and overload amplitude, and were smaller than that of strengthened beam without loading. Hence, unloading should be comprehensively ensured in practical strengthening engineering to satisfy requirements for structural usability after strengthening. 4. Conclusions The following conclusions were obtained based on the test and analysis results in this paper. (1) The failure modes were observed for the tested beam specimens: crushing failure of concrete in the compressive zone, wedge-shaped failure of the partial protective layer of concrete in the tensile zone, or a combination of both modes. (2) Before strengthening, the cycle loading resulted in specimen damage, and the damage degrees were aggravated with the increase of cycle number and overload amplitude. Strengthening with CFRP strips could effectively inhibit crack development and improve the crack properties. Strengthening under loading could also lower the improvement degree, and this adverse effect could be eliminated through unloading. (3) Compared with the control beam, the bearing capacities of the strengthened beams improved. The yield loads increased by 10–28%, while the ultimate loads increased by 7–15%. The cycle number, overload amplitude, and strengthening under loading influenced the yield loads of the strengthened beams. The yield loads increased with the increase of the cycle number and overload amplitude, and the yield load of the strengthened beam under unloading was higher than that of strengthened beams under loading. The cycle number, overload amplitude, and strengthening under loading had limited influences on the ultimate bearing capacity. (4) Before strengthening, the stiffness of beams degraded under cycle loading; the deformation was affected little by cycle number but affected considerably by overload amplitude. Moreover, a sustained loading period exerted certain
adverse effects on the deformations of the beams. After strengthening, the deformations of all strengthened beams under the same load were smaller than that of the control beam due to the strengthened effect of CFRP strips. Within the cycle number in this test, the deformation decreased gradually with the increase of cycle number, but increased with the overload amplitude. The deformations of the strengthened beams under loading were slightly higher than that of the strengthened beam under unloading in the initial phase but lower when approaching failure in the later phase. (5) Cycle number, overload amplitude, and strengthening under loading all affected the ductility factors of the strengthened beams. The displacement ductility factors of the strengthened beams under loading decreased with the increase of cycle number and overload amplitude, and were smaller than that of strengthened beam without loading. Unloading should be comprehensively ensured in practical strengthening engineering to satisfy requirements for structural usability after strengthening. Conflict of interest The authors declare that there is no conflict of interest. Acknowledgments The authors acknowledge the financial support provided by the Natural Science Foundation of China, and the Science and Technology Agency and Education Department of Liaoning Province. The authors are also grateful for the support provided by colleagues and the School of Resources and Civil Engineering Northeastern University during the experimental program. Finally, the authors would like to gratefully acknowledge their friends and the 211 Test and Research Center of Northeastern University for their assistance in testing the specimens. Funding This work was supported by the Natural Science Foundation of China (grant number 51504125) and the Natural Science Foundation of Liaoning Province (grant number 20170540303). References [1] X.Y. Sun, C.K. Huang, G.F. Zhao, Y.Q. Dou, Experimental study of the influence of truck overloads on the flexural performance of bridge members, China Civ. Eng. J. 38 (6) (2005) 35–40. [2] X.Y. Sun, H.L. Wang, C.K. Huang, Experimental investigation on flexural performance of bridge members under overloading, J. Zhejiang Univ. (Eng. Sci.) 42 (1) (2008) 152–156. 163. [3] S.H. Li, H.W. Jiang, Influence law of different weight limit levels on highway bridge safety, China J. Highway Transp. 29 (3) (2016) 82–88. 115. [4] H. Peng, J.R. Zhang, C.S. Cai, Y. Liu, An experimental study on reinforced concrete beams strengthened with prestressed near surface mounted CFRP strips, Eng. Struct. 79 (2014) 222–233. [5] J.G. Teng, L. De Lorenzis, B. Wang, R. Li, Debonding failures of RC beams strengthened with near surface mounted CFRP strips, J. Compos. Constr., ASCE 10 (2006) 92–105. [6] L. Anania, A. Badala, G. Failla, Increasing the flexural performance of RC beams strengthened with CFRP materials, Constr. Build. Mater. 19 (2005) 55–61. [7] M. Emara, C. Barris, M. Baena, L. Torres, J. Barros, Bond behavior of NSM CFRP laminates in concrete under sustained loading, Constr. Build. Mater. 177 (2018) 237–246. [8] M.I. Kabir, M. Subhani, R. Shrestha, B. Samali, Experimental and theoretical analysis of severely damaged concrete beams strengthened with CFRP, Constr. Build. Mater. 178 (2018) 161–174. [9] S.S. Zhang, T. Yu, Analytical solution for interaction forces in beams strengthened with near-surface mounted round bars, Constr. Build. Mater. 106 (2016) 189–197.
J. Yang, L. Wang / Construction and Building Materials 190 (2018) 265–275 [10] American Concrete Institute (ACI), Guide for the design and construction of externally bonded FRP systems for strengthening concrete structures, ACI-440 2R Farmington Hills, Mich, 2008, pp. 76. [11] J. Michels, M. Staskiewicz, C. Czaderski, R. Kotynia, Y.E. Harmanci, M. Motavalli, Prestressed CFRP strips for concrete bridge girder retrofitting: application and static loading test, J. Bridge Eng. 21 (5) (2016) 04016003. [12] Y.J. Kim, R. Gordon Wight, M.F. Green, Flexural strengthening of RC beams with prestressed CFRP sheets: using nonmetallic anchor systems, J. Compos. Constr. 12 (1) (2008) 44–52. [13] L. De Lorenzis, J.G. Teng, Near-surface mounted FRP reinforcement: an emerging technique for strengthening structures, Compos. Part B 38 (2) (2007) 119–143. [14] R. El-Hacha, S.H. Rizkalla, Near-surface-mounted fiber-reinforced polymer reinforcements for flexural strengthening of concrete structures, Struct. J. 101 (5) (2004) 717–726. [15] S.S. Zhang, J.G. Teng, T. Yu, Bond-slip model for CFRP strips near-surface mounted to concrete, Eng. Struct. 56 (2013) 945–953. [16] J.A.O. Barros, R. Kotynia, Possibilities and challenges of NSM for the flexural strengthening of RC structures, in: Fourth international conference on FRP composites in civil engineering, Zurich, Switzerland, 2008, pp. 6. [17] R. Kotynia, Analysis of reinforced concrete beams strengthened with near surface mounted FRP reinforcement, Arch. Civil Eng. 52 (2) (2006) 305–317. [18] T. Oshiro, S. Hamada, Structural performance and bending test of deteriorated reinforced concrete bridge, ACI SP. 88 (1985) 39–58. [19] E.G. Burdette, D. W. Goodpasture, Correlation of bridge load capacity estimates with test data. National Cooperative Highway Research Program report 306, USA, 1988.
275
[20] R.A. Hawileh, H.A. Rasheed, J.A. Abdalla, A.K. Al-Tamimi, Behavior of reinforced concrete beams strengthened with externally bonded hybrid fiber reinforced polymer systems, Mater. Des. 53 (2014) 972–982. [21] J. Tan, Experiment and analysis on flexural behavior of prestressed concrete beam reinforced with CFRP sheet, Harbin Institute of Technology, 2008. [22] A.M. Khalifa, Flexural performance of RC beams strengthened with near surface mounted CFRP strips, Alexandria Eng. J. 55 (2) (2016) 1497–1505. [23] Y.C. You, K.S. Choi, J.H. Kim, An experimental investigation on flexural behavior of RC beams strengthened with prestressed CFRP strips using a durable anchorage system, Compos. Part B 43 (2012) 3026–3036. [24] M. Badawi, K. Soudki, Flexural strengthening of RC beams with prestressed NSM CFRP rods-Experimental and analytical investigation, Constr. Build. Mater. 23 (10) (2009) 3292–3300. [25] X.Y. Sun, C.K. Huang, Test on flexural performance of RC bridge member after overload reinforced with externally bonded FRP, China J. Highway Transp. 19 (4) (2006) 82–87. [26] X.Y. Sun, C. Xu, H.L. Huang, J.F. Chu, Overloading simulation fatigue experiment of bridge member with and without CFRP reinforcement, J. Zhejiang Univ. (Eng. Sci.) 46 (9) (2012) 1585–1591. [27] R.X. Wang, K. Wang, Z.W. Cui, Mechanical characteristics analysis of overload pre-cracked RC beams strengthened with CFRP and steel sheet, J. Hebei Univ. Technol. 38 (6) (2009) 85–88. [28] L.R.F.D. Aashto, Bridge design specifications, SI Units, third ed., AASHTO, Washington, D.C., 2004.