An experimental study on the flexural behavior of local prestressed steel

Abstract

The flexural bearing capacities of three composite continuous box beams with different prestressing degrees were tested and studied to investigate the influences of local prestressing bundles on the deflection, relative slip of interface, strain, and redistribution of the internal force of the steel–concrete composite continuous box beam. Results show that the arrangement of local prestress can not only improve the bending stiffness at the mid-span of the composite continuous box beam, but also significantly enhance the ductility. In the design process of the local prestressed composite continuous box beam, the influence of the slip at the middle support should be fully considered, and the deflection could not be taken as a control factor. The internal force redistribution of the local prestressed continuous box beam is lower than that of the ordinary continuous box beam, but it still has good plastic internal force redistribution. The number of prestressing bundles in the negative moment region is the main factor that affects the internal force redistribution of the middle support.

Keywords

composite continuous box beam deflection local prestressing redistribution of internal forces slip

Introduction

Since the 1950s, as one of the important transverse load-bearing components, steel–concrete composite beams have been widely used in industrial buildings, bridge structures, underground structures, long-span structures, and so on. Because of their wide application in engineering, diverse structure, and complex performance, many scholars have conducted in-depth studies on them (Cheng and Yao, 2016; Hui and Wu, 2001; Xiang et al., 2017; Zhang et al., 2011).

With the development of lightweight and long-span building and bridge structures, common steel–concrete composite beams cannot meet the actual needs of engineering. When the span of a structure is large, the ordinary simply supported composite beam will produce considerable deflection. Meanwhile, when the ordinary composite continuous beam is used, the negative moment zone of the middle support will crack prematurely and affect the normal use. In these cases, prestressed steel–concrete composite continuous beams can not only provide better solutions to the problems of stiffness decline, deformation, and cracking under large span or load conditions caused by the premature cracking of concrete in the negative moment zone, but also can expand the elastic operating range of beams, increase stiffness, and improve the bearing capacity. Relevant research is summarized as follows. Yu and Guo (2004) conducted an experimental study on the generation and development laws of cracks in the negative moment zone of prestressed steel–concrete composite continuous beams. They discussed the major factors that affect crack width and proposed an empirical calculation formula for crack width in the same form as the current norms. Su et al. (2015) performed an experimental study on the structural response of prestressed steel–concrete composite continuous box beams under short-term load. Their results show that the initial cracking load and ultimate load of prestressed concrete slabs are 3.16 and 2.61 times higher than those of ordinary concrete slabs, respectively. Nie and Tao (2009) analyzed the bearing capacity of prestressed steel–concrete composite continuous beams through experiments and proposed accurate and practical analysis methods. Hu and Ye (2013a) conducted theoretical research on the stiffness, deformation, and flexural capacity of prestressed steel–concrete composite continuous beams. The calculation methods for flexural capacity and ultimate flexural capacity in the negative moment zone were derived on the basis of the transformed section method and simplified plasticity theory.

The prestress reinforcement technology can be generally divided into two types: the internal prestress reinforcement method and the external prestress reinforcement method (Chen et al., 2009; Yang et al., 2009). However, most of the research on the prestress reinforcement of composite beams adopts the method of laying prestress reinforcement through the length of the internal body. Zhou et al. (2007) performed an experimental study on the crack resistance and crack width in the negative bending zone of prestressed steel fiber–reinforced concrete (SFRC) and steel composite beams, and put forward an empirical formula for calculating the maximum crack width considering factors such as the degree of prestressing and the force ratio. Wang et al. (2020) studied the mechanical properties of assembled monolithic steel–prestressed concrete composite beams (AMS-PCCB) under negative bending moments. The effects of shear joints, prestress, and other factors on the flexural capacity, interfacial slip, and crack development of composite beams were obtained. Sayyar et al. (2013) studied the variation of the ultimate bending capacity of composite box sections under prestressing effects. The theoretical models of prestressing on the flexural performance of composite materials were established and verified by the flexural test. In most cases, it is only to prevent the occurrence of tensile crack in the negative bending moment area of the steel–concrete composite continuous beam or to solve the problem of the structure’s failure to meet the functional requirements due to the large cracks in local areas. When the span is strengthened using the through-length tensioning method, the structural support will be adversely affected. Similarly, when the support is strengthened, it will considerably affect the span and make the force complex. Although there are relevant studies on the local arrangement of prestressed steel bars, for example, Taoum et al. (2015) carried out experimental studies on the performance of steel beams strengthened using locally prestressed reinforcing steel bars. Picard et al. (2000) proposed the method of applying locally tensioned prestressed bars to concrete beams and discussed the internal force analysis of single-span statically indeterminate beams. However, in practical engineering, bridges are generally continuous beams with equal or unequal spans, and most of them are composite sections. Therefore, related mechanical property tests and research should be conducted on steel–concrete composite continuous box beams with local prestressing bundles.

Test overview

Specimen details

This experiment designs and manufactures three scale specimens of single-chamber steel–concrete beams which are two-span composite continuous beams with the same size. These models are 1:5 reduced-scale models of an actual bridge and designed according to the Highway Reinforced Concrete and Prestressed Concrete Bridge and Culvert Design Code (JTG D62-2004, 2004). The beams are numbered RC1, PC2, and PC3. RC1 is an ordinary composite continuous box beam. PC2 and PC3 are prestressed composite continuous box beams. A long-term mechanical performance test (455 days) is conducted on the three specimens before the start of this test (Cao et al., 2018). The composite box beam has a total length of 12.00 m, a height of 0.31 m, a concrete wing slab width of 1.20 m, and a steel box beam bottom width of 0.60 m. The upper wing slab is made of C40 concrete, the steel box beam is made of Q235 grade steel, and the longitudinal tensile reinforcement in the wing slab is HPB235, arranged with a 100-mm equal spacing. The material type of the bolt connector is ML15 (GB/T 10433-2002, 2002), the nominal diameter of the round head bolt is 13 mm, and the height after welding is 50 mm. There are 80 shear bolts in a single row and 160 in total. The bolts are arranged at a 150-mm equal spacing on the flange of the steel beam, and all three specimens have partial shear connections. A steel strand with a design strength of 1860 MPa is adopted as the prestressed rebar with a nominal diameter of 15.2 mm. The post-tensioning method is used to apply prestressing force, and the prestressing bundles are only distributed within the 2.5 m range around the middle support to keep the concrete in the negative moment zone under compression. PC2 is arranged with four strands of prestressing bundles with a tensile value of 178 kN per strand. PC3 is arranged with six strands of prestressing bundles with a tensile value of 170 kN per strand. The layout of reinforcement and prestressing bundles and the size of the model section are shown in Figure 1.

Material properties

The production time, batch, and curing conditions of the concrete specimens are the same as those of the model beams, and the mix ratio of the concrete materials is provided in Table 1. The measured mechanical properties of concrete corresponding to the age of the flexural test are listed in Table 2, according to the Standard for Test Method of Mechanical Properties on Ordinary Concrete (GB/T 50081-2002, 2002). The measured mechanical properties of steel are listed in Table 3, according to the Metallic Materials—Tensile Testing—Part 1: Method of Test at Room Temperature (GB/T 228.1-2010, 2010) and the Steel Strand for Prestressed Concrete (GB/T 5224-2003, 2003).

Table 1.

Mixture properties of concrete (kg/m³).

Water	Cement	Coal ash	Fine ore	Sand	Gravel	Pumping agent
168	350	30	100	730	1020	13

Table 2.

Mechanical properties of concrete.

Cube compressive strength (MPa)	Axial compressive strength (MPa)	Tensile strength (MPa)	Elastic modulus (MPa)
54.84	42.24	3.46	3.22 × 10⁴

Table 3.

Mechanical properties of steel.

Type of material	Thickness or diameter (mm)	Tensile strength (MPa)	Yield strength (MPa)	Elongation (%)	Modulus of elasticity (MPa)
Prestressing tendons	15.2	1968.3	1835.0	4.6	1.94 × 10⁵
Common reinforcement	6.0	377.8	266.1	25.4	1.97 × 10⁵
Steel plate	8.0	431.3	291.9	32.0	2.04 × 10⁵
Stud connectors	13.0	453.6	50.4	23.5	1.90 × 10⁵

Test loading and measurement

Since the model beam is made by scaling, in order to simulate the actual dead load, 12 concrete prisms are arranged at each 1/4 span of the test beam according to the principle of stress equivalence, which is converted into the equivalent bending moment of the mid-span section of 27 kN m. The total counterweight of each beam is the same. The flexural capacity test devices of the continuous beam consist of two 1000-kN hydraulic jacks with a stroke of 152 mm and a counterforce frame that applies a concentrated load symmetrically to the span of the specimen. The schematic diagram of the loading device is shown in Figure 2.

Prior to formal loading, the test beam was preloaded with a load of 20 kN two to three times to eliminate residual strain and make the beam enter the elastic working state. At the same time, the reliability of the middle loading and instrument device was checked. The test is performed via monotonic static loading (10 kN/level), when the load approached the calculated cracking load or ultimate load. It was loaded at 5 kN/level until the test piece exhibited a remarkable failure state or the load cannot be continued. To ensure the stability of the obtained data, the stable loading time should not be less than 5 min after each loading stage, and the stable instrument reading is used as the test data.

Measurement content and point layout

Deflection deformation. The deflection values of a test beam are measured using dial gauges mounted on end supports, L/4, L/2, 3L/4, and the middle support.

Interface relative slip. Nine micrometers were arranged continuously along the length of the beam. The micrometer was fixed between the steel beam and the concrete wing slab through the magnetic stand to measure the relative slip of the interface in each control section of the test beam. The arrangement of measurement points is shown in Figure 2(a).

Strain. During the production process, a vibrating wire extensometer is embedded into the concrete wing slab of the middle span and support sections, and two strain gauges are arranged in each section. A resistive strain gauge is mounted on the surface of the steel beam and the concrete wing slab of the middle span and support sections. Figure 3 shows the specific arrangement of in-body and body-surface strain gauges.

Support reaction. Two pressure sensors, which are 200 and 500 kN, respectively, were arranged at the side and middle supports to measure the branch reaction.

Crack. Prior to loading, crack width is measured using a crack observer. After each stage of loading is completed and the load is held for 5 min, the new cracks are numbered and the widths of all visible cracks are recorded. Crack development is observed and recorded until the specimen is destroyed.

Test results and analysis

Loading process and main test results

The main test results of the failure span of beam specimens are shown in Table 4. In this test, the failure of RC1 and PC2 is bending failure, PC3 shows premature failure under the control of bolts, and its bolts in the negative bending moment area are cut off. Affected by the long-term test, there were cracks of different widths on the top surface of the wing slab in the middle support section of the specimen. The maximum widths of the initial cracks of RC1, PC2, and PC3 were 0.22, 0.12, and 0.08 mm, respectively.

Table 4.

Major test results for failure span.

Specimen	P_cr (kN)	P_y (kN)	P_u (kN)	δ_y (mm)	δ_u (mm)	P_y/P_u	δ_y/δ_u
RC1	–	170	400	11.43	34.56	0.43	0.33
PC2	45	210	460	9.84	46.64	0.48	0.30
PC3	81	240	340	9.35	29.05	0.70	0.32

P_cr is the mid-span load value of concrete cracking on the top surface of the wing slab in the negative bending moment area. P_y is the mid-span load value that the longitudinal tensile reinforcement in the negative moment area reaches the yield strain. P_u is the applied ultimate load of the failure span. δ_y is the mid-span yield deflection at a lower flange yield of the steel beam. δ_u is the mid-span ultimate deflection when the failure span reaches the ultimate bearing capacity state.

Before the loading of PC2 and PC3 reached 0.10P_u and 0.24P_u (P_u is the applied ultimate load of the failure span), respectively, the original crack width of the top surface of the specimen did not expand, and the mid-span deflection increased linearly with the loading. As the load continues to increase, transverse cracks appear on the top surface of the wing slab of the middle support section of RC1, and then develop and extend to the edge of the flange. Multiple transverse cracks are uniformly distributed in the negative bending moment area along the beam axis. PC2 and PC3 show a few oblique cracks in addition to the penetrating cracks in the middle support section. When RC1, PC2, and PC3 were loaded to 0.23P_u, 0.28P_u, and 0.44P_u, respectively, the measured bending moment in the mid-span began to increase and exceeded the elastic calculated value, and internal force redistribution gradually started to occur in each beam. RC1, PC2, and PC3 emit abnormal sounds when the loading ranges are (0.25–0.43)P_u, (0.32–0.55)P_u, and (0.47–0.74)P_u, respectively, indicating that the bonding between the wing slab and the steel beam fails. After loading to 0.8P_u, the mid-span deflection of each specimen increased rapidly, cracks on the top surface of the wing slab of the middle support section increased, and the crack width increased. Finally, the measured negative bending moment in the middle support stopped growing, while the measured positive bending moment in the middle span continued to increase. Before the failure of RC1 and PC2, the web of steel beam at the middle support was buckling, forming a plastic hinge in the section of the middle support, and the deflection increased rapidly. Finally, the bottom slab of the steel beam at the middle span of RC1 was subjected to tensile yield, and the concrete wing slab in the middle span of PC2 was crushed. At this time, RC1 and PC2 reached the ultimate bearing capacity. For PC3, due to the shear failure of some bolts in the negative bending moment area of the middle support, the wing slab and the steel beam were detached, the interface slip increased sharply, and the composite beam lost its bearing capacity in advance. Due to the abnormal loading failure of PC3, the actual applied ultimate load is lower than the expected ultimate load. For convenience of unified expression, P_u is still used to represent its applied ultimate load of mid-span.

Load–deflection curve

The load–deflection curves of the mid-span sections of the three composite beams are presented in Figure 4.

Taking PC2 (damaged by normal loading) as an example, the load–deflection characteristic curves of the local prestressed composite continuous box beam can be approximately divided into three stages. Before loading to 0.28P_u (PC3 is loaded to 0.44P_u before), the bending stiffness of the composite beam basically remains unchanged, the deflection–deformation curve develops linearly, and the specimen is in the elastic working condition. It is worth noting that, when the applied load reached 0.5P_u, the deflection of the two spans of PC2 and PC3 does not exceed 1/400 of the single span of the beam (GB 50017-2017, 2017), which is 15 mm, indicating that the flexural rigidity of the specimen is relatively large, and the deflection cannot be considered as a control factor in the design. During the process from 0.28P_u to the ultimate load P_u, the internal force redistribution of the composite beam develops. The negative bending moment of the middle support increases slowly, and the positive bending moment of the middle span increases gradually. The deformation curve of the mid-span deflection develops nonlinearly, and the specimen is in the elastic–plastic working state. After reaching the ultimate load, the failure span deflection increased rapidly, and the load–deflection curve gradually decreased. The composite beam lost its bearing capacity and entered the third stage. The load–deflection curve of specimens RC1 and PC2 decreased more smoothly than that of PC3, indicating that the ductility of PC3 was not fully reflected.

Compared with RC1, the mid-span yield deflections δ_y of PC2 and PC3 decreased by 13.9% and 18.2%, respectively. Due to the premature failure of PC3 after yield, the mid-span ultimate deflection of PC3 is relatively low. Therefore, the analysis only focuses on PC2, and the ultimate deflection δ_u at the mid-span of PC2 is 34.9% higher than that of RC1. The ductility indexes (Jo et al., 2004) of RC1 and PC2 calculated using formula (1) are 3.02 and 4.74, respectively, indicating that the arrangement of local prestress can not only improve the bending stiffness at the mid-span of the composite continuous box beam, but also significantly enhance the ductility

μ_{δ} = \frac{δ_{u}}{δ_{y}}

(1)

where µ_δ is the ductility index of deflection.

Relative slip of interface

Figure 5 shows the relative slip distribution at the interface of the failure span of the composite continuous beam, and the middle support is located at the coordinate origin. As shown in the figure, the slip curve of the local prestressed composite continuous beam is similar to that of the fully prestressed composite continuous beam (Hu and Gong, 2013). The common rules are as follows: (1) the relative slip at the mid-span loading point is basically zero within the elastic working range of a continuous beam; (2) the slip distribution curve has positive and negative maximum values at approximately L/4 of the mid-span, and the maximum slip occurs at a section near the reverse bending point of the section between the loading point and the middle support. The difference is that the slip at the support of PC2 and PC3 is not zero and is kept within 0.2 mm before 0.65P_u of PC2 (0.88P_u of the same level of load in PC3). This is because the stiffness of the upper wing slab is strengthened by local prestress, which causes deformation inconsistency between the concrete slab and the steel box beam, resulting in a deformation difference.

After the RC1 composite beam loaded to 0.45P_u, due to the increased crack depth, the anchoring effect between the concrete and the bolt connector decreases, resulting in rapid slip development. Due to the mutual influence between cracking and sliding, on the top surface of the wing plate at the middle support section of RC1 there exist relatively wide (0.22 mm) initial cracks after the long-term test was finished. With the increase of load, the crack gradually penetrated through the section and formed a penetrating crack, which caused an abnormal phenomenon that the slip at the middle support was larger than that at L/4 of the mid-span when the load reached P/P_u = 0.75. By comparing RC1 and PC2, it can be seen that the slip development of the prestressed composite continuous box beam is significantly delayed and the slip can be effectively restrained. By comparing PC2 and PC3, it can be found that, under the same load, the slippage of PC2 at 0.39P_u, 0.53P_u, and 0.65P_u (PC3 is 0.53P_u, 0.71P_u, and 0.88P_u, respectively) is 0.049, 0.122, and 0.178, while the slippage of PC3 is 0.138 and 0.151, which are 0.191, 2.82, 1.24, and 1.07 times the slippage of PC2, respectively. It means that the interface slip of the local partial prestressed composite continuous box beam is obviously affected by the number of prestressing bundles in the negative bending moment region. The deformation difference caused by local prestress is also verified. In the design process of the local prestressed composite continuous box beam, more consideration should be given to the influence of slippage at the middle support.

The slip distribution diagram of PC3 clearly shows that the slip on the right side of the support increases sharply when it is loaded to 0.71P_u. The slip reached 1.78 mm when it is loaded to P_u, and the shear failure of the interface bolt on the right side of the support in PC3 is observed in the test. By analyzing the data in Table 5, the measured ultimate bending moment values at the middle support of RC1, PC2, and PC3 were 188.1, 232.3, and 265.5, respectively. Therefore, it can be inferred that the stiffness of the upper wing slab of PC3 is enhanced due to the deployment of more prestressing bundles, which leads to a greater negative bending moment in the support, resulting in shear failure. After the shear failure of the bolt, PC3 loses its bearing capacity in advance under the control of the bolt.

Table 5.

Test values of the ultimate bending moment amplitude modulation coefficient of the middle support.

Specimen	M_sec (kN m)	M_bu (kN m)	P_u (kN)	M_eb (kN m)	α (%)
RC1	–	188.1	400	442.5	57.5
PC2	43.0	232.3	460	508.9	45.9
PC3	61.7	265.5	340	376.1	13.0

In vivo and surface strains

The concrete strain curves of the middle span and support sections are shown in Figure 6. In the figure, L, M, and R represent the middle span section of the left span, the middle support section, and the right span, respectively. L₁ and L₂ are the vibrating wire extensometers embedded in both sides of the upper wing slab in the middle of the left span, as shown in Figure 3. Similarly, M₁ and M₂ are the vibrating wire extensometers embedded in both sides of the upper wing slab in the middle support section, while R₁ and R₂ are their counterparts in the right span.

The test results indicate that the strains on the left and right sides of the mid-span cross section of the three beams are insignificantly different. By contrast, the in-body strains on the left and right sides of the mid-support cross section gradually vary with loading, with RC1 being the most significant. The reason for such a relationship is that the concrete slab appears to be asymmetrically and laterally cracking after loading, and the concrete on the side with a large cracking degree produces a large strain. Hence, the strains on both sides exhibit the asymmetrical phenomenon. However, cracks appear on the top surface of the upper wing of RC1 before loading. Therefore, the internal strains on both sides of RC1 exhibit a considerable difference from the beginning of loading. As shown in Figure 8, the cracking load of the supports in the prestressed composite continuous box beams is significantly increased. Consequently, the elastic working performance of the composite beams and the stress distribution in the concrete body are improved. At the mid-span section, the ultimate strain of PC2 is 16.2% higher than that of RC1, indicating that the local prestressed composite continuous box beam still demonstrates better plastic internal force redistribution performance.

Figure 7 shows the average strain distribution of the body surface of the middle span and support sections of the failure span. As the surface concrete strain gauge of the support section wing slab in RC1 is damaged, relevant data are missing.

The section strain under initial loading is approximately distributed linearly along the beam height, which conforms to the assumption of the flat section. A comparison of the strain distribution of the mid-span section of the three beams indicates that RC1 is characterized by a large and rapid development of the steel beam strain, while PC2 and PC3 are characterized by a large upper wing strain. The neutral axis heights of the mid-span section of RC1, PC2, and PC3 are 180, 155, and 130 mm, respectively. The neutral axis height of the mid-span section decreases with the increase of the prestressing degree. The neutral axis of the support section of RC1 moves down to approximately 45 mm from the bottom of the steel beam due to the cracking of the upper concrete. By contrast, PC2 and PC3 exhibit good stress conditions of the sections under the prestressing effect. The height of the neutral axis of PC2 and PC3 remains at approximately 75 and 80 mm after a small amount of downshifting, respectively.

The neutral axis of the middle support section increases with the prestressing degree, while the neutral axis of the middle span section decreases with the increase of the prestressing degree, indicating that local prestressing not only exerts a beneficial effect on improving the stress distribution of the middle support of the composite continuous box beam, but also has a good effect on the mid-span section.

Internal force redistribution and bending moment amplitude modulation

The bending moment–load relationship of each composite continuous beam calculated according to the elastic theory, based on the measured support reaction, is shown in Figure 8. In the figure, M_m is the measured positive bending moment in the mid-span and M_b is the measured negative bending moment in the middle support. M_em and M_eb are the calculated elastic bending moments of the mid-span and middle supports, respectively.

As can be seen from Figure 8, in the elastic working stage, except for RC1, the measured positive bending moment M_m at the mid-span and the negative bending moment M_b at the middle support of PC2 and PC3 are basically consistent with the calculated values of elasticity, which is mainly caused by the more severe initial damage of RC1. With the increase of load, the increase of the negative bending moment of the middle support gradually slows down and decreases compared with the elastic calculation results, while the positive bending moment of the middle span increases greatly and gradually exceeds the elastic calculation value, that is, internal force redistribution. By comparison, it can be seen that with the increase of the number of prestressing bundles in the negative bending moment area, the redistribution degree of the bending moment decreases, the middle support section forms a plastic hinge later, and the actual bending moment distribution is closer to the elastic calculation results.

During practical use, the internal force redistribution phenomenon caused by the stiffness change along the entire length of a composite beam leads to the differences between the calculation results of the internal force and the elastic analysis results (Hu and Ye, 2013b, Nie et al., 2003). Hence, the bending moment amplitude modulation method is used to consider the internal force redistribution of continuous beams. Table 5 lists the ultimate moment modulation coefficients of the supports in each composite continuous beam.

The measured ultimate bending moment amplitude modulation coefficient of the middle support of the ordinary composite continuous beam is calculated as follows

α = \frac{M_{eb} - M_{bu}}{M_{eb}} \times 100 %

(2)

where M_eb is the bending moment of the middle support calculated on the basis of the elastic theory, $M_{eb} = - (3 / 16) P_{u} l$ , l is the span length, and M_bu is the measured ultimate bending moment of the middle support.

The measured ultimate bending moment amplitude modulation coefficient of the middle support of the prestressed composite continuous beam is calculated as follows

α^{P} = \frac{(M_{eb} - M_{\sec}) - M_{bu}}{M_{eb}} \times 100 %

(3)

where M_sec is the secondary bending moment value generated by the tensioned prestressing tendon in the middle support section. $M_{\sec} = (- 3 c + 1.5 c^{2}) N_{pe} e_{u}$ , where N_pe is the effective pre-pressure after tensioning the prestressing tendon, e_u is the eccentricity from the prestressing tendon to the section centroid of the composite beam, and c is the distance from the middle support to the end of the prestressing tendon (Wang and Cao, 2004). The increment of the internal force of the prestressing tendon is less than 30% of the effective prestressing degree and smaller than the elastic limit state. Thus, it is assumed that the second moment of the prestressing tendon remains unchanged during the entire loading process.

It can be seen that the bending moment amplitude modulation coefficient of the middle support of the local prestressed composite continuous box beam is smaller than that of the ordinary composite continuous box beam, and it decreases significantly with the increase of prestress. This is because the stiffness of the negative bending moment section is larger, and thus the redistribution of the bending moment of the whole beam is reduced. For the composite continuous beams with local prestressing bundles, the number of prestressing bundles in the negative moment zone is the primary factor that affects the redistribution of internal forces in the middle supports.

Fracture development

The distribution of cracks on the top surface of the negative bending moment area in the continuous beams is shown in Figure 9, where “1” and “2” marked in the figure represent the first and second cracks with a width exceeding 0.1 mm measured on the concrete roof of each specimen, respectively. Before loading, RC1 has more cracks on the surface of the concrete flange than on the top surface, and cracks are primarily distributed around 450 mm on both sides of the middle support. The original cracks continue to extend to both sides along the direction perpendicular to the beam axis until they penetrate the section and develop into main cracks.

Before the loading process of PC2 and PC3, no evident crack appears on the top surface of the upper wing slab at the middle support, only a few minor cracks exist on the surface of the two wings, and the cracks on the surface of the flange of PC3 are less than those of PC2. The distribution pattern of cracks in local prestressed composite continuous beams is similar; in addition to transverse through-cracks in the middle support, there are also a few oblique cracks with asymmetric bias. The development of subsequent fractures is deepening and expanding along with the initial damage fractures, but the morphology and distribution of the initial damage fractures and the final damage fractures are basically the same. At the end of the experiment, the maximum crack width of RC1 was 4.14 mm and the average crack spacing was about 103.2 mm. The maximum crack width of PC2 was 3.12 mm and the average crack spacing was about 90.8 mm. The maximum crack width of PC3 was 2.98 mm, and the average crack spacing was about 82.2 mm. The crack width at the support of the local prestressed composite continuous beams is small and concentrated, while that of the ordinary composite continuous beams is large and mostly through-crack.

Conclusion

This work investigates the influences of local prestressing bundles on the deflection, relative slip of interface, strain, and redistribution of the internal force of the steel–concrete composite continuous box beam and compares its performance with that of the ordinary composite continuous box beam. The main conclusion of this research is as follows:

The arrangement of local prestress can not only improve the bending stiffness at the mid-span of the composite continuous box beam, but also significantly enhance the ductility.

Under the same load, the slip in the middle support of PC3 is 2.82, 1.24, and 1.07 times that of PC2, respectively. The interface slip of the local prestressed composite continuous box beam is significantly affected by the number of prestressed bundles configured in the negative moment zone. In the design process of the local prestressed composite continuous box beam, the influence of the slip at the middle support should be fully considered, and the deflection could not be taken as a control factor.

Although the internal force redistribution degree of the local prestressed composite continuous box beam is lower than that of the ordinary composite continuous box beam, it still has good plastic internal force redistribution characteristics. The local prestress has a good effect on improving the stress distribution of the support and the mid-span section of the composite continuous box beam.

The ultimate bending moment amplitude modulation coefficients in the middle support of RC1, PC2, and PC3 are 57.5%, 45.9%, and 13.0%, respectively. The bending moment amplitude modulation coefficient of the middle support of the local prestressed composite continuous box beam is smaller than that of the ordinary composite continuous box beam, and it decreases significantly with the increase of prestress. The number of prestressing bundles in the negative moment zone is the primary factor that affects the redistribution of internal forces in the middle supports.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the National Natural Science Foundation of China (Grant No. 51551801), the Construction Special Foundation of Hunan Innovative Province (Grant No. 2019RS1059), and the Scientific Research Initiation Project of Xiangtan University (Grant No. 16QDZ03).

References

Cao

Han

Dai

, et al. (2018) Long-term experimental study on prestressed steel-concrete composite continuous box beams. Journal of Bridge Engineering 23(9): 04018067.

Chen

Wang

Jia

(2009) Comparative study of continuous steel-concrete composite beams prestressed with external tendons: experimental investigation. Journal of Constructional Steel Research 65(7): 1480–1489.

Cheng

Yao

(2016) Simplified method for predicting the deflections of composite box beams. Engineering Structures 12(128): 256–264.

GB 50017-2017 (2017) Standard for Design of Steel Structures. Beijing, China: China Planning Press.

GB/T 10433-2002 (2002) Cheese Head Studs for Arc Stud Welding. Beijing, China: China Standards Press.

GB/T 228.1-2010 (2010) Metallic Materials—Tensile Testing—Part 1: Method of Test at Room Temperature. Beijing, China: China Architecture & Building Press.

GB/T 50081-2002 (2002) Standard for Test Method of Mechanical Properties on Ordinary Concrete. Beijing, China: China Architecture & Building Press.

GB/T 5224-2003 (2003) Steel Strand for Prestressed Concrete. Beijing, China: China Standards Press.

Gong

(2013) Experimental study on flexural behavior of prestressed composite continuous beams with partial shear connections. Building Structure 43(6): 64–67.

10.

(2013a) Analysis of flexural capacity of prestressed composite continuous beams in negative moment zone. Engineering Mechanics 30(11): 160–165.

11.

(2013b) Study on internal force redistribution of prestressed steel-concrete composite continuous beams. Journal of Disaster Prevention and Mitigation Engineering 33(2): 231–234.

12.

Hui

(2001) The moment amplitude modulation coefficient and internal force redistribution of composite continuous beams. Non-ferrous Mining and Metallurgy 17(5): 41–43.

13.

Tae

Kwon

(2004) Ductility evaluation of prestressed concrete beams with CFRP tendons. Journal of Reinforced Plastics and Composites 23(8): 843–859.

14.

JTG D62-2004 (2004) Code for Design of Highway Reinforced Concrete and Prestressed Concrete Bridges and Culverts. Beijing, China: China Communications Press.

15.

Nie

, et al. (2003) Experimental study on internal force redistribution of prestressed steel-concrete composite continuous beams. Industrial Construction 33(12): 12–14.

16.

Nie

Tao

(2009) Analysis of bearing capacity of prestressed steel-concrete composite continuous beams. China Civil Engineering Journal 42(4): 38–47.

17.

Picard

Bastien

Massicotte

, et al. (2000) Analysis of continuous beams prestressed with localized tendons. Journal of Structural Engineering 126(2): 262–265.

18.

Sayyar

Soroushian

Weerasiri

, et al. (2013) Prestressing effects on the ultimate flexural strength of composite box sections. Composite Structures 106: 806–814.

19.

Yang

Bradford

(2015) Behavior of a composite continuous box beam with a prefabricated prestressed-concrete slab in its hogging-moment region. Journal of Bridge Engineering 20(8): B4014004.

20.

Taoum

Jiao

Holloway

, et al. (2015) Upgrading steel I-beams using local post-tensioning. Journal of Constructional Steel Research 113: 127–134.

21.

Wang

Cao

(2004) Calculation of internal forces of local tensioned prestressed tendons to reinforce statically indeterminate concrete beams. Journal of Southeast University (Natural Science Edition) 34(1): 89–91.

22.

Wang

Liu

, et al. (2020) Experimental study on assembled monolithic steel-prestressed concrete composite beam in negative moment. Journal of Constructional Steel Research 167: 105667.

23.

Xiang

Qiu

(2017) Calculation of long-term slippage for externally prestressing steel-concrete composite beam. Journal of Zhejiang University (Engineering Science) 51(4): 739–744.

24.

Yang

Park

Neale

(2009) Flexural behaviour of reinforced concrete beams strengthened with prestressed carbon composites. Composite Structures 88(4): 497–508.

25.

Guo

(2004) Experimental study on crack width in negative moment zone of local prestressed steel-concrete composite continuous beams. Journal of Building Structures 25(4): 55–59.

26.

Zhang

Fan

(2011) Characteristics of the redistribution of composite continuous beam and its variation with load. Journal of Central South University (Science and Technology) 42(2): 449–455.

27.

Zhou

Dai

Liu

(2007) Experimental study of crack-resistance and crack width in negative bending zone of interior prestressed SFRC-steel composite beams. Journal of Building Structures 28(3): 82–90.

An experimental study on the flexural behavior of local prestressed steel–concrete composite continuous box beams

Abstract

Keywords

Introduction

Test overview

Specimen details

Material properties

Test loading and measurement

Measurement content and point layout

Test results and analysis

Loading process and main test results

Load–deflection curve

Relative slip of interface

In vivo and surface strains

Internal force redistribution and bending moment amplitude modulation

Fracture development

Conclusion

Footnotes

Declaration of Conflicting Interests

Funding

References