Experimental study on the shear behavior of joints in precast concrete segmental foundations

Abstract

By adopting the segmental precast post-tensioning method in foundations, the efficiency of constructing precast concrete structures can be significantly enhanced. To study the shear behavior of precast concrete segmental foundation (PCSF) joints, 13 double shear specimens were tested. The failure mode and shear carrying capacity (V_u) of the specimens were investigated. The parameters which were studied included the type of shear key, the confining stress, the type of bond for the internal post-tensioned bars and the application of epoxy resin. The results showed that design parameters changed the crack extensions of the joints. The V_u of the key unreinforced specimens was 29.14% lower than that of the reinforced specimens. After grouting of the post-tensioned ducts, the average V_u increased by 112.87% than that before grouting. With the increased in confining stress and the application of epoxy resin, the average V_u of the specimens increased by 21.9% and 48.45%, respectively. Finally, the test results were compared with the predicted results for the V_u of the joints. In predicting V_u for PCSF joints, the dowel action of the bonded post-tensioned reinforcement and the action of the reinforcement in the key must be considered on the basis of the existing prediction expressions.

Keywords

precast segmental foundation joints shear keys double shear test shear carrying capacity

Introduction

Precast concrete structures are widely used in superstructure construction (Parastesh et al., 2014; Zhou et al., 2022). However, precast structures are basically not fully precast. Usually, the substructures are cast in place. To further improve construction efficiency, precast foundations have been proposed and used in engineering (Nazir et al., 2013; Yu et al., 2008). When the size of the precast foundation is too large, it is necessary to precast the foundation in segments. Thus, the segmental precast posttensioning method (SPPM) are introduced for foundation construction. The SPPM exploits the segmental construction and prestressing technology to connect multiple concrete segments together by posttensioning (Wang et al., 2019; Zhan et al., 2022). In structures formed by SPPM, joints represent the location of discontinuities. Notably, joints are more susceptible to damage than adjacent sections, and excessive stresses lead to unfavorable results (Jiang et al., 2015; Megally et al., 2003; Turmo et al., 2005). Therefore, the proper design of precast concrete segmental foundation (PCSF) joints to achieve acceptable performance is critical.

Basically, shear keys are set at the joints to improve the shear behavior of the interface. Further, the epoxy resin can be applied to the joints to improve water tightness and compensate for irregularities. The application of epoxy changes the shear behavior of the joint compared to dry joints. By comparing specimens with dry and epoxy connections, Allawi et al. (2019) found that the failure modes of the specimens were different for the two connection methods. The failure of the dry joints usually occurred at the segmental interface. Due to the higher tensile strength of the epoxy glue compared to concrete, the failure of the epoxy joints occurred at the concrete near the segmental interface. The same conclusion was obtained by Al-Sherrawi et al. (2018) and Li et al. (2013). Li et al. (2013) also proposed a simplified failure mode for dry and epoxy joints when subjected to combined shear and flexure forces. In addition, Wrayosh and Hashim (2020) concluded that the failure of carbon fiber reinforced joints and steel shear key joints also occurred in the concrete near the segmental interface. Yuan et al. (2019) proposed that the failure mode of the joints was influenced by the geometry, distance, number and reinforcement of the key. Furthermore, Fu et al. (2021) found that the angle of the key also had an effect, with the mechanical properties of the 30° and 60° angle specimens being better than those of the 45° specimens. Therefore, the design of PCSF joints requires a combination of key size, angle and epoxy application quality to achieve the desired failure mode of the joints.

The shear carrying capacity (V_u) of joints is also affected by epoxy and shear key. The presence or absence of key and the presence or absence of epoxy at the joints can be broadly classified into four types of joints. In general, the strength and stiffness of dry joints are always lower than those of epoxy flat or key joints (Saibabu et al., 2013). Compared to monolithic cast joints, epoxy flat joints and epoxy key joints had comparable and higher strengths, respectively (Buyukozturk et al., 1990). Ahmed and Aziz (2019a, 2020) found that epoxy joints had worse ductility than dry joints. However, the brittle behavior could change to a gradual strength degradation pattern with an increasing number of shear keys. Based on the test results of dry joint specimens, Koseki and Breen (1983) suggested that the contributions of the shear friction and number of shear keys to the initial stiffness of the joints were not additive. In addition, Shamass et al. (2015) demonstrated that the contribution of the shear friction to the total V_u decreased as the confining stress increased. Therefore, Shamass et al. (2015) suggested that the friction coefficient used in the American Association of State Highway and Transportation Officials (AASHTO) Load and Resistance Factor Design (LRFD) code equation (American Association of State Highway and Transportation Officials, 2004) could be reduced when applying high confining pressures. AASHTO’s prediction of the V_u of keys takes into account the frictional contribution of the keys surface in addition to the keys themself. Zhang et al. (2022) derived an analytical prediction model based on the Mohr stress circle, which could be used to predict the Vu of epoxy joints. Le et al. (2019) conducted tests on the flexural performance of precast concrete segmental beams (PCSBs). Their results showed that the type of joint had only a slight effect on the beam carrying capacity and ductility and a significant effect on the initial beam stiffness. Tests conducted by Yao et al. (2020), featuring joints with different epoxy glue thicknesses, showed that a thicker epoxy thickness correlated to a higher shear resistance and better plastic deformation capacity of the epoxy joints. However, based on the shear tests conducted by Zhou et al. (2005), the epoxy glue thickness appeared to have a limited positive effect on the joints. Compared to the 3 mm epoxy glue thickness, the 1 or 2 mm epoxy glue thicknesses had greater V_u. In addition, the contribution of epoxy glue to the V_u was also temperature dependent. Issa and Abdalla (2007) found that the V_u of epoxy specimens was higher in hot weather than in cold weather.

In addition to the effects of the keys and epoxy, the reinforcement through the joint, the bond of the post-tensioned bars, and the strength of the aggregate have an effect on the transfer of shear forces through the joints. The dowel action through the joint surface reinforcement is an effective shear force transfer mechanism (Dulacska, 1972). Dei Poli et al. (1993) stated that the efficiency of shear transfer in reinforcement depend largely on the constraints imposed by the cover concrete. Takase (2019) studied the mechanical properties of reinforcement under combined shear and tensile stresses. He and Kwan (2001) developed a numerical model for the role of reinforcement dowel action in concrete cracks. Choi et al. (2023) indicated that the V_u of joints decreased by replacing normal coarse aggregates with lightweight aggregates, but the V_u could be improved by adding hook-end steel fibers. Lu et al. (2022) tested the shear behavior of PCSBs considering post-tensioned bars with or without bonding. The results showed that the overall strength and stiffness of PCSBs could be improved by bonding the post-tensioned bars. The same conclusions were obtained from cyclic load tests on precast segmented columns performed by Li et al. (2018). In addition, Li et al. concluded that shear keys set at the joints did not improve significantly in terms of hysteresis behavior. Ahmed and Aziz (2019b) stated that the eccentricity of the prestressing did not affect the elastic behavior of the joints, but its role evidently appeared at the plastic stage.

In summary, the main factors affecting the shear behavior of the joints are the effects of key, epoxy, reinforcement and confining stress. Therefore, it is necessary to consider these factors in the design of PCSF joints. Unlike superstructures, the number of shear keys at PCSF joints is influenced by the foundation dimensions. Usually, the height and width of the joints of PCSF are relatively small. For integrity and safety reasons, PCSF joints are provided with only one shear key and are arranged through the length of the joints in the width direction. Also, the key was equipped with reinforcement to improve the V_u of the key itself. Considering that the PCSF is located underground, epoxy resin was applied at the joints to protect the post-tensioned bars. For PCSF joints, the confining stress comes from the prestressing applied by the post-tensioned bars and the external pressure. The PCSF joints are subjected to bending moments under soil pressure. The confining stress at the joints is not uniform. Thus, despite the extensive research on joint shear behavior of bridges and superstructures in the literature, the joint shear behavior of SPPM forming PCSF has not been fully addressed.

As shown in Figure 1, the three concrete segments were assembled by post-tensioned bars to form a PCSF. In this study, the shear behavior of joints in PCSF was investigated by double shear tests. 13 specimens were tested considering joint type, confining stress, epoxy resin, post-tensioned bar bond type and key reinforcement parameters. The crack development, failure mode and V_u of the specimens were obtained and the effect of different parameters was analyzed. The test data were further used to evaluate several dry or epoxy joints for V_u prediction.

Figure 1.

Schematic diagram and split diagram of the PCSF.

Experimental program

Test specimens

A total of 13 specimens were fabricated and tested. The specimen configuration was designed in such a way that the joint was subjected to a major shear force. Hence, a double shear test setup for studying the shear transfer behavior of concrete was used. The test parameters included the type of joint (i.e., epoxy or dry), the confining stress (i.e., 0.7 MPa, 1.4 MPa, and 0.29 MPa to 1.11 MPa), type of shear key (with a single key or with a flat joint), key reinforcement (reinforced or unreinforced), and post-tensioned bars (bonded or not bonded). The design parameters of each specimen are shown in Table 1. Shear key reinforcement are shown in Figure 2(b).

Table 1.

Design parameters of each specimen.

Specimen		Key type	Bonding type	Joint type	Confining stress (Mpa)
Label	Test name	Key type	Bonding type	Joint type	Confining stress (Mpa)
F-1	F-U-σ1-D	Flat	Unbonded	Dry	0.7
F-2	F-U-σ2-D	Flat	Unbonded	Dry	1.4
F-3	F-U-σ1-E	Flat	Unbonded	Epoxied	0.7
F-4	F-U-σ2-E	Flat	Unbonded	Epoxied	1.4
F-5	F-B-σ1-E	Flat	Bonded	Epoxied	0.7
F-6	F-B-σ2-E	Flat	Bonded	Epoxied	1.4
K-1	K-U-σ2-D	Single	Unbonded	Dry	1.4
K-2	K-U-σ1-E	Single	Unbonded	Epoxied	0.7
K-3	K-U-σ2-E	Single	Unbonded	Epoxied	1.4
K-4	K-U-σ1-E*	Single	Unbonded	Epoxied	0.7
K-5	K-U-σ3-E	Single	Unbonded	Epoxied	0.29–1.11
K-6	K-B-σ1-E	Single	Bonded	Epoxied	0.7
K-7	K-B-σ2-E	Single	Bonded	Epoxied	1.4

Note: F: flat; K: key; U: unbonded post-tensioned bar; B: bonded post-tensioned bar; σ1: confining stress of 0.7 MPa; σ2: confining stress of 1.4 MPa; σ3: uneven confining stress of 0.29 MPa-1.11 MPa (i.e., confining stress of 0.29 MPa at the upper edge of the specimen and 1.11 MPa at the lower edge of the specimen); D: dry; E: epoxy; the symbol *: shear key without reinforcement.

Figure 2.

Specimen dimensions and configurations (unit: mm). (a) Detail of flat joint. (b) Detail of single key joint.

Each specimen was composed of three segments with two post-tensioned bars and with or without a male‒female shear key. Confining stresses were applied by tensioning the post-tensioned bars. The dimensions of each segment were 350 × 250 × 365 mm. Hence, the dimensions of all tested specimens were 1050 × 250 × 365 mm. The trapezoidal shape of the key had a 105 × 250 mm base area and a 35 × 250 mm top area with a 35 mm depth. In order to strengthen the V_u of the key, a longitudinal reinforcement with a diameter of 8 mm was arranged in the male key. At the same time, hoops with a diameter of 6 mm and a spacing of 100 mm were placed along the longitudinal direction of the male key. The centerlines of the post-tensioned ducts used were 20 mm diameter metal sheaths located 102.5 mm from the centerline of the segment. Two post-tensioned ducts were arranged symmetrically above and below the segment. In addition, the shear plane size between the two segments of the tested specimen was 250 × 365 mm. The details of the specimen and the reinforcement arrangement are shown in Figure 2.

Fabrication of the specimens

The fabrication and assembly process of the specimens are shown in Figure 3. Fabrication of specimens was performed in a precast factory. After the concrete was cured for 28 days, the specimens were transported to the test laboratory for assembly. When assembling, after matching the segments, epoxy glue was applied evenly on the joint surface. Next, the post-tensioned bars were inserted into the aligned specimens and a torque wrench was used to apply tension to the posttensioned bars. Pressure was developed between the joints by post-tensioned bars tension. The compressive stresses between the joints of different specimens were 0.7 MPa, 1.4 MPa, 0.29 MPa to 1.11 MPa. Where 0.29 MPa to 1.11 MPa considered the uneven compressive stresses between the joints of PCSF. Notably, after posttensioning was completed, the bonded specimens needed to be filled with mortar in the post-tensioned ducts for internal reinforcement.

Figure 3.

Shear specimen fabrication and assembly process.

Test setup and instrumentation

Shear specimens were tested under monotonically increasing loads. Static loading was applied to the middle segment by a hydraulic jack, and each increment was 20 kN per stage in 1 min. The confining stresses on the joint surfaces in the PCSF were applied through the tensioning of the post-tensioned bars. Confining stresses were verified by load sensors during the testing process. The detailed setup and layout of the transducers in the test are shown in Figure 4(a)–(c). Two linear variable differential transducers (LVDTs) were placed on both sides of the joint of the specimen to measure the relative displacement of the middle and the side segments.

Figure 4.

Test setup. (a) Schematic diagram of the tested specimen. (b) Photograph of the test. (c) Layout of LVDTs.

Material properties

The compressive cube strength (f_cu) of the concrete obtained at the time of testing of each specimen are shown in Table 2. The f_cu after 28 days was determined from three 150 mm × 150 mm × 150 mm cubes. The mechanical properties for reinforcements are provided in Table 3. The yielding strength and tensile strength of the different diameters of reinforcement were obtained via direct tensile tests.

Table 2.

Mechanical properties of the concrete.

Specimen		The compressive cube strength (f_cu, MPa)
Label	Test name	Sample 1	Sample 2	Sample 3	Average value
F-1	F-U-σ1-D	40.1	36.2	26.8	34.4
F-2	F-U-σ2-D	35.2	34.1	32.7	34.0
F-3	F-U-σ1-E	43.2	33.5	33.1	36.6
F-4	F-U-σ2-E	41.1	31.8	37.7	36.8
F-5	F-B-σ1-E	43.2	33.5	33.1	36.6
F-6	F-B-σ2-E	41.1	31.8	37.7	36.8
K-1	K-U-σ2-D	34.4	40.8	35.7	37.0
K-2	K-U-σ1-E	33.3	36.5	27.8	32.7
K-3	K-U-σ2-E	31.3	31.5	32.6	31.8
K-4	K-U-σ1-E*	40.1	36.2	26.8	34.4
K-5	K-U-σ3-E	36.6	39.2	37.2	37.7
K-6	K-B-σ1-E	33.3	36.5	27.8	32.7
K-7	K-B-σ2-E	31.3	31.5	32.6	31.8

Note: The average axial compressive strength (f_c) is evaluated by assuming f_c = 0.76 f_cu (GB 50010-2010, 2015).

Table 3.

Mechanical properties of the reinforcements.

Diameter (mm)	Yielding strength (MPa)	Tensional strength (MPa)	Elongation (%)
6	464	667	31.7
8	452	663	32.3
B^T15	1016	1129	11.4

Note: B^T: Prestressed bars.

The elongation was defined as the ratio of the difference between the post fracture and original gauge lengths to the original gauge length (GB/T 228.1-2010, 2010).The original gauge length was 5 times the reinforcement diameter.

Crack patterns and failure modes of the tested specimens

Flat joint specimens

Typical failure modes of the flat joint specimens are shown in Figure 5. For the dry flat joint specimens F-1 and F-2, no cracks were observed in the specimens throughout the loading process, and no damage was evident at the joints with only some concrete crushed powder. Their final failure was caused by the shear sliding between the joints. For the epoxy flat joint specimens F-3 and F-6, the first crack usually appeared at the interface between the concrete surface and the epoxy glue surface. Simultaneously, the bonded area between epoxy and concrete appeared partially separated, and small cracks appeared on the concrete surface. The failure of F-3 and F-6 was also caused by shear sliding at the joints. However, unlike F-1 and F-2, the failure of F-3 and F-6 was sudden and violent, with significant relative sliding between the joints. Compared with F-1∼F-6 (bonded epoxy flat joints) had more dense cracks near the joints at the time of failure.

Figure 5.

Typical failure modes of the flat joints.

At the moment when F-1 and F-2 began to slide, the jack applied loads of 80 kN and 166 kN to the middle segment of the specimen, and the corresponding relative displacements were 0.90 mm and 0.48 mm, respectively. The cracking loads and corresponding relative displacements of F-3∼F-6 were 140 kN, 100 kN, 240 kN, 120 kN and 0.70 mm, 0.25 mm, 0.36 mm, 0.21 mm, respectively. The V_u and corresponding relative displacements of F-1∼F-6 were 87 kN, 173 kN, 200 kN, 185 kN, 439 kN, 485 kN and 2.02 mm, 1.34 mm, 1.12 mm, 0.7 mm, 3.84 mm, 1.15 mm, respectively. The test results of all specimens are summarized in Table 4.

Table 4.

Summary of the test results.

Specimen		V_in (kN)	V_u (kN)	V_in/V_u	∆_u/∆_in	K_in (kN/mm)	Failure mode
Label	Test name	V_in (kN)	V_u (kN)	V_in/V_u	∆_u/∆_in	K_in (kN/mm)	Failure mode
F-1	F-U-σ1-D	80	87	0.92	2.24	88.89	Shear sliding
F-2	F-U-σ2-D	166	173	0.96	2.79	345.83	Shear sliding
F-3	F-U-σ1-E	140	200	0.70	1.60	200	Shear sliding
F-4	F-U-σ2-E	100	185	0.54	2.80	400	Shear sliding
F-5	F-B-σ1-E	240	439	0.55	10.67	666.67	Shear sliding
F-6	F-B-σ2-E	120	485	0.25	5.48	571.43	Shear sliding
K-1	K-U-σ2-D	240	610	0.39	2.84	324.32	Shear-off
K-2	K-U-σ1-E	160	573	0.28	4.47	213.33	Shear-off
K-3	K-U-σ2-E	280	662	0.42	4.29	400	Shear-off
K-4	K-U-σ1-E*	280	406	0.69	2.50	666.67	Shear-off
K-5	K-U-σ3-E	120	763	0.16	6.93	171.43	Shear-off
K-6	K-B-σ1-E	120	873	0.14	8.06	375	Shear-off
K-7	K-B-σ2-E	320	910	0.35	7.42	744.19	Shear-off

Note: V_in: cracking load or sliding load; V_u: shear carrying capacity; ∆in: displacement corresponding to V_in; ∆u: displacement corresponding to V_u.

The ductility of the specimen is defined as the ratio ∆_u to ∆_in. The initial stiffness (K_in) of the specimen is defined as the ratio of V_in and ∆_in.

Key joint specimens

Typical failure modes of key joint specimens are shown in Figure 6. The failure sequence of the key joint specimen is shown in Figure 7.

Figure 6.

Typical failure modes of the key joints.

Figure 7.

Key joint specimen failure sequence.

For the dry key joint specimen K-1, the initial crack occurred at the top angle of the male key and progressed to an inclined angle of 10°–20° relative to the joint. The crack was formed due to the presence of compressive struts in the shear keys. As the load increased, more diagonal cracks appeared on the top surface of the male key to the bottom corner. Moreover, diagonal cracks occurred in the female keys and extended toward the mid-segment loading point. When K-1 was damaged, main cracks formed at the root of the male keys, while the concrete cover was crushed and spalled off. In addition, the diagonal cracks in the region of the female key and loading point were crushed due to the compression‒shear combination.

For the epoxy key joint specimens K-2∼K-7, the initial crack was formed at the interface between the epoxy glue and the concrete. The subsequent crack development of K-2∼K-7 was similar to that of K-1. Notably, compared with K-1∼K-5, K-6 and K-7 exhibited diagonal cracks on the female keys first rather than on the male keys. In addition, at the time of damage, the cracks in K-6 and K-7 were denser, and the diagonal cracks in the female keys extended longer. This indicated that the failure mode of the specimen was changed after the post-tensioned bar grouting. Compared to K-1, -2, -3, -5, and -6, K-4’s key was fully sheared off when damage. The failure of K-4 was sudden and simultaneously accompanied by a clear failure sound. Thus, the reinforcement of the shear key will change the failure mode of the shear key from key cover crushed to key sheared off.

The cracking loads of K-1∼K-7 were 240 kN, 160 kN, 280 kN, 280 kN, 120 kN, 120 kN, 220 kN, corresponding to relative displacements of 0.74 mm, 0.75 mm, 0.70 mm, 0.42 mm, 0.70 mm, 0.20 mm, 0.43 mm, respectively. The V_u of K-1∼K-7 were 610 kN, 573 kN, 662 kN, 406 kN, 763 kN, 873 kN, 910 kN, corresponding to relative displacements of 2.1 mm, 3.35 mm, 3 mm, 1.05 mm, 4.85 mm, 2.58 mm, 3.19 mm, respectively.

Analysis of the test results

Effect of the shear key type

The vertical load-relative displacement curves of the specimens are shown in Figure 8. For unbonded specimens (post-tensioned bars not grouted), the V_u and ductility of K-1, -2, -3 increased by 252.6% and 1.65%, 186.50% and 179.17%, 257.84% and 53.06%, respectively, compared with F-2, -3, -4. For bonded specimens, compared with specimens F-5, the V_u and ductility of specimens K-6 were increased by 98.86% and 20.94%, respectively. Compared with specimens F-6, the V_u and ductility of specimens K-7 were increased by 87.63% and 35.47%, respectively. In addition, the ratios of cracking load to ultimate load (V_in/V_u) for the key joint specimens were lower than those for the flat joint specimens. According to Figure 9, there was basically no difference between the initial stiffness of flat and key joints. Thus, it can be concluded that the shear key set at the joint could improve the shear carrying capacity (V_u) of the specimen.

Figure 8.

Vertical load versus relative displacement curves for the different key types. (a) Effect of key type on post-tensioned bar unbonded specimens. (b) Effect of key type on post-tensioned bar bonded specimens.

Figure 9.

Vertical load versus relative displacement curves for different confining stresses. (a) Effect of confining stress on flat joint specimens. (b) Effect of confining stress on key joint specimens.

The V_u of K-4 with no reinforcement in the key was the lowest, and the V_in/V_u was the highest, which was a typical brittle failure mode. Compared with K-2 and F-3, the V_u of K-4 decreased and increased by 29.14% and 103%, respectively, and its ductility decreased and increased by 44.03% and 56.25%, respectively. In addition, compared with K-2 and F-3, the V_in/V_u of K-4 increased and decreased by 146.98% and 1.48%, respectively. Thus, the shear key with reinforcement could further improve the V_u and ductility.

Effect of the confining stress

The confining stress of the joint was provided by the post-tensioned bars. The test results of different confining stress specimens were analyzed to determine the effects. As shown in Figure 9, the vertical load on the dry flat joint specimens increased linearly with respect to the load when the joint faces began to slide.

The confining stresses of F-1, -3, and -5 (0.7 MPa) were less than those of F-2, -4, and -6 (1.4 MPa). The initial stiffnesses (K_in) of F-2, -4, and -6 were 389.06%, 200% and 85.71% of the K_in values of F-1, -3, and 5, respectively. Similarly, the confining stresses of K-2 and -6 (0.7 MPa) were less than those of K-3 and -7 (1.4 MPa). The K_in values of K-3 and -7 were 187.50% and 85.27% of the K_in values of K-2 and -6, respectively. The V_u values of F-1, -3, -5 at 0.7 MPa were 50.29%, 108.11% and 90.52% of those of F-2, -4, -6 at 1.4 MPa, respectively. The V_u values of K-2 and -6 were 86.56% and 95.93% of those of K-3 and -7, respectively. Hence, the initial stiffness and the V_u generally increased with increasing confining stress. Additionally, dry and post-tensioned bar bonded joints were more effective than the epoxy and post-tensioned bar unbonded joints in improving the shear behavior by confining stresses. The K_in values of K-3 increased by 133.33%, compared with those of K-5 with uneven confining stresses. The reason is that, in a bi-directional stress state, the decrease in lateral pressure leads to a decrease in the shear strength of the specimen. Compared to K-3, the compressive stress applied to the upper part of specimen K-5 was reduced by 79.3% and the upper part cracked earlier during loading.

Effect of the post-tensioned bar bond type

The vertical load and relative displacement curves of the specimens with different bonding types are shown in Figure 10. For the flat joint specimens, compared with F-3, the K_in, V_u and ductility of F-5 were improved by 233.33%, 199.50% and 566.67%, respectively. Compared with F-4, the K_in, V_u and ductility of F-6 were improved by 42.86%, 162.16% and 95.58%, respectively. For the key joint specimens, compared with K-2, the K_in, V_u and ductility of K-6 were improved by 181.25%, 52.36% and 188.81%, respectively. Compared with K-3, the K_in, V_u and ductility of K-7 were improved by 27.91%, 37.46% and 73.10%, respectively.

Figure 10.

Vertical load versus relative displacement curves for different bonding types. (a) Effect of bond type on flat joint specimens. (b) Effect of bond type on key joint specimens.

After grouting, the force transfer between the post-tensioned reinforcement and the concrete was improved, while the prestress loss was reduced. The presence of grout allowed the post-tensioned bars of the bonded specimens to develop stresses more quickly during loading compared to the unbonded specimens. This resulted in a more pronounced dowel action in the bonded specimens. Therefore, the shear behaviors between the joints of the bonded specimens were higher than those of the unbonded specimens.

Effect of the joint type (with or without epoxy)

As shown in Figure 11, the load displacement curves for dry flat joints and epoxy flat joints developed differently. The curve of the dry flat joints had a long development process after the V_u, and the strength remained basically unchanged, while the strength of epoxy flat joints rapidly decreased. When the epoxy adhesive strength of flat joints failed, the inter-joint V_u rapidly degraded to the friction between the epoxy adhesive surface and the concrete surface. Compared with F-1, K_in and V_u of F-3 increased by 125% and 129.89%, respectively, while V_in/V_u of F-3 decreased by 23.88%. Compared with F-2, K_in and V_u of F-4 increased by 15.66% and 6.94%, respectively, while V_in/V_u of F-4 decreased by 43.67%.

Figure 11.

Vertical load versus relative displacement curves for the different joint types. (a) Effect of joint type on flat joint specimens. (b) Effect of joint type on key joint specimens.

The development of load displacement curves for dry key joints and epoxy key joints nearly coincided with each other. Compared with K-1, the K_in and V_u values of K-3 increased by 23.33% and 8.52%, respectively. After the complete failure of the epoxy glue, the V_u of the epoxy key joint was caused by friction and shear keys, and its stress state was similar to that of the dry key joint. Compared with the dry joints, the epoxy glue improved the V_u of the joints. However, this improvement was affected by the thickness and uniformity of the epoxy glue application. The epoxy glue provided unstable V_u.

Shear resistance mechanism discussion of PCSF joints

Composition of the V_u of the PCSF joints

Analysis based on experimental results, the composition of the V_j (the calculated value of the V_u) of the key epoxy joints in this study is shown in eq. (1). Scholars have proposed different equations based on the shear resistance mechanism, as detailed in Section The V_u of key joints compared with existing formulas of this paper. The composition of the V_u of single-side PCSF joint is shown in Figure 12.

Figure 12.

Composition of the V_u of single-side PCSF joint.

The V_u of the key dry joint was mainly derived from the interlocking action of the shear key and the frictional force of the flat part. The V_u of the key epoxy joints was also related to the adhesive strength of the epoxy glue. Compared with the key joints, the V_u of a flat joint did not include the contribution of the shear key, and the friction and adhesive strength were provided by the entire shear surface. According to Moore’s stress circle theory, the V_u of the joint was increased by the confining stress applied through the post-tensioned bars. Moreover, the development of diagonal cracks caused by the compressive strut in the shear keys was also limited by the confining stresses. With increasing relative displacement between the joints, the dowel action and clamping force of post-tensioned bars became more evident, further improving the V_u of the joints.

The joints of PCSF have shear key reinforcement, post-tensioned ducts grouted and the joints are applied with epoxy resin. These parameters increase V_k, F_H, F_dow and V_adh in eq. (1). As a result, the designed PCSF joints achieved the highest V_u in the comparison of the specimens (Chapter 4). According to the failure mode of PCSF joints (Chapter 3), the order of failure is firstly the interface between epoxy and concrete, followed by the shear key. Thus, the contribution of the interface shear component is more active in the early stages with V_adh and V_k. In the later stages the separation of the concrete cover is mainly observed, which is the role of F_dow in action. Among all the specimens only the specimens with post-tensioned ducts grouting had this damage pattern. The post-tensioned ducts grouting increased the force transfer between the post-tensioned bars and the concrete, making the reinforcement dowel action more pronounced.

V_{j} = 2 μ (F_{H} + F_{cla}) + V_{k} + 2 F_{dow} + V_{adh}

(1)

where V_j is the calculated value of the joint’s V_u, μ is the coefficient of friction, F_H is the prestress of the post-tensioned bar, F_cla is the post-tensioned bar clamping stress, V_k is the contribution of the root of the shear key and the shear force provided by the support pressure, F_dow is the dowel action of the post-tensioned bar, and V_adh is the adhesion contribution of the epoxy resin.

The V_u of key joints compared with existing formulas

In the prediction model, the dry joint V_u usually consists of two main components: the contribution of keys and the friction between different segment surfaces. Hence, the V_u of the dry flat joints is calculated using eq. (2). In addition, the V_u prediction model for epoxy joints includes the contribution of the epoxy layer.

V_{j} = μ σ_{n} A_{j}

(2)

where V_j is the calculated value of the joint’s V_u (kN) and μ is the friction coefficient. σ_n is the confining stress (MPa), and A_j = total area of the shear plane (mm²).

Typical formulas for calculating the V_u of dry and epoxy joints are shown in Table 5. The test data from our study were substituted into the equations in Table 5 to assess the reliability of the prediction models in the table. A comparison of the predicted results with the test results is shown in Table 6. The prediction accuracy was assessed based on analyzing the average of the test results and prediction model (V_t/V_p) and the corresponding coefficient of variation (COV).

Table 5.

Typical formula for calculating the V_u of the joints.

Literature	Expression	Joint type
Buyukozturk and Bakhoum	$V_{j} = A_{j} (0.647 \sqrt{f_{c}^{'}} + 1.36 σ_{n})$	Key dry
	$V_{j} = A_{j} (0.921 \sqrt{f_{c}^{'}} + 1.20 σ_{n})$	Key epoxy
	$V_{j} = A_{j} (0.85 \sqrt{f_{c}^{'}} + 0.98 σ_{n})$	Flat epoxy
AASHTO	$V_{j} = A_{k} \sqrt{f_{c}^{'}} (0.2048 σ_{n} + 0.9961) + 0.6 A_{s m} σ_{n}$	Key dry
AASHTO	$V_{j} = A_{j} (4.17 + 1.06 σ_{n})$	Key epoxy
Lu et al.	$V_{j} = A_{j} (0.55 \sqrt{f_{c}^{'}} + 1.0 σ_{n})$	Key dry
	$V_{j} = A_{j} (0.55 \sqrt{f_{c}^{'}} + 1.4 σ_{n})$	Key epoxy
	$V_{j} = A_{j} (0.6 \sqrt{f_{c}^{'}} + 0.83 σ_{n})$	Flat epoxy

Note: V_j: the calculated value of the joint’s V_u (kN); A_j: total area of the shear plane (mm²); σ_n: confining stress (MPa); f_c^′: compressive strength of the concrete cylinders; A_k: area of all key roots in the shear plane (mm²); A_sm: area of all flat surfaces in the shear plane (mm²); f_ck: standard value of the axial compressive strength of concrete (MPa); f_cu,k: standard value of the concrete cubic compressive strength (MPa).

Table 6.

Comparison of the predicted results with the test results.

	Joint type	V _t	V _Buyukozturk	V_t/V_Buyukozturk	V _AASHTO	V_t/V_AASHTO	V _Lu	V_t/V_Lu
F-U-σ1-E	Flat epoxy	200	479.66	0.42	—	—	347.42	0.58
F-U-σ2-E	Flat epoxy	185	543.40	0.34	—	—	401.24	0.46
F-B-σ1-E	Flat epoxy	439	479.66	0.92	—	—	347.42	1.26
F-B-σ2-E	Flat epoxy	485	543.40	0.89	—	—	401.24	1.21
K-U-σ2-D	Key dry	610	319.19	1.91	236.66	2.58	399.09	1.53
K-U-σ1-E	Key epoxy	573	503.80	1.14	448.22	1.28	344.51	1.66
K-U-σ2-E	Key epoxy	662	574.53	1.15	515.93	1.28	430.40	1.54
K-U-σ1-E*	Key epoxy	406	514.76	0.79	448.22	0.91	351.06	1.16
K-U-σ3-E	Key epoxy	763	611.95	1.25	515.93	1.48	452.74	1.69
K-B-σ1-E	Key epoxy	873	503.80	1.73	448.22	1.95	344.51	2.53
K-B-σ2-E	Key epoxy	910	574.53	1.58	515.93	1.76	430.40	2.11
Average (flat epoxy)	—	—	—	0.643	—	—	—	0.878
COV (flat epoxy)	—	—	—	0.411	—	—	—	0.411
Average (key epoxy)	—	—	—	1.273	—	1.443	—	1.782
COV (key epoxy)	—	—	—	0.242	—	0.236	—	0.244

The friction coefficients for the dry flat joints were calculated via eq. (2). The friction coefficients of F-U-σ1-D and F-U-σ2-D were 0.681 and 0.677 at confining stresses of 0.7 MPa and 1.4 MPa, respectively. When the concrete joint surface was not roughly treated, 0.7 was used as the friction coefficient of the joint and was generally consistent with ACI 318-19 (2019) and AASHTO (μ = 0.6). However, when the confining stress was larger, the friction coefficient could decrease due to the polishing effect (Turmo et al., 2006).

Based on Table 6, for the flat epoxy joints, the predicted V_u obtained by both Bakhoum (1990); Lu (2004) were higher than our test V_u, with average V_t/V_p values of 0.643 and 0.878 and COV values of 0.411 and 0.411, respectively. The average value of V_u predicted by Buyukozturk et al. was higher than 36.5% of the V_u predicted by Lu. Compared with the results from Lu, the results from Buyukozturk showed greater cohesion and a higher coefficient of internal friction, resulting in a higher V_u for the flat epoxy joints. For the key epoxy joints, the average values for V_t/V_p and COV predicted by Buyukozturk, AASHTO and Lu were 1.273 and 0.242, 1.443 and 0.236, and 1.782 and 0.244, respectively. Compared with AASHTO and Lu, the results predicted by Buyukozturk were closer to our test results. Additionally, the predicted V_u obtained by Buyukozturk was greater than those predicted by AASHTO and Lu by 13.4% and 40.0%, respectively. These results were caused by the maximum cohesion of the Buyukozturk prediction model, while the internal friction coefficient was lower than that of Lu and greater than that of AASHTO. Notably, the cohesion in the AASHTO prediction model was independent of the concrete strength and produced a constant value of 0.9961. Thus, this model was only applicable to the adhesive failure mode. When damage occurred in the concrete, Buyukozturk and Lu’s prediction models were more suitable. For the key dry joints, Lu’s predictions were more consistent with our test results compared with those from Buyukozturk and AASHTO.

Notably, according to Table 6, each prediction model generally underestimated the V_u of the bonded and reinforced key specimens. These results occurred because the positive contributions of the dowel action of the reinforcement and the shear key reinforcement to the joint carrying capacity were not considered in each prediction model. Typically, the V_u of concrete shear keys are considered in the equations. When predicting the V_u of reinforced shear key joints, the coefficients of the shear key contribution portion should be adjusted upwards accordingly. In addition, the contribution of the dowel action of the reinforcement at the joints was not considered in any of the prediction models. Meanwhile, only the effect of specimen concrete strength on the V_u was considered in the prediction model. When the strength of the grout is higher than the concrete strength class, the dowel action of the post-tensioned bars will obviously be stronger. Therefore, further research is needed to reflect the above parameters in the equations.

Conclusions

This paper presented an experimental study on the shear behavior of precast concrete segmental foundation (PCSF) joints. 13 double shear specimens were examined by double shear tests, and the effects of the key type, confining stress, bond type of post-tensioned bars and joint type (with or without epoxy) on the shear behavior of the joints were analyzed. Several current expressions for predicting the shear carrying capacity (V_u) of joints were further discussed, and their predictions were compared with test results. Some conclusions are listed as follows:

(1) The joint failure mode was influenced by the design parameters of the joints. Flat joints, plain concrete key joints, and reinforced shear key joints exhibited shear sliding, shear key cover crushing, and shear keys completely shearing off, respectively. Compared with flat dry joints, flat epoxy joints exhibited significant brittle damage. Compared with post-tensioned unbonded joints, the cracks near the joints were denser after post-tensioned grouting.

(2) The V_u of the joints was significantly increased by the reinforcement shear key and grouting of the post-tensioned bars. The increased confining stress and the application of epoxy glue could further improve the V_u of the joints.

(3) For the prediction of the V_u values of the flat epoxy joints and key dry joints, Lu’s expression was more accurate than the Buyukozturk and AASHTO expressions. For the prediction of the V_u value of the key epoxy joints, Buyukozturk’s expression was more accurate than Lu’s and AASHTO’s expressions.

(4) When the expressions were applied to PCSF joints, the contribution of the reinforcement in the key and the dowel action of the post-tensioned bars needed to be considered in the design relationship; otherwise, the value of V_u will be underestimated.

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: The research in this paper was financially supported by the National Natural Science Foundation of China (No. 52178472), the Anhui Provincial Key Research and Development Plan (No. 202104a07020022), Fundamental Research Funds for the Central Universities (No. JZ2023HGTB0257), Xinjiang Agricultural University’s 2022 University-level Graduate Research Innovation Program (No. XJAUGRI2022019). Their support is gratefully acknowledged.

ORCID iDs

Qing Jiang

Xun Chong

Junqi Huang

References

ACI 318-19 (2019) Building Code Requirements for Structural Concrete and Commentary. Farmington Hills, MI: American Concrete Institute.

Ahmed

Aziz

(2019a) Shear behavior of dry and epoxied joints in precast concrete segmental box girder bridges under direct shear loading. Engineering Structures 182: 89–100.

Ahmed

Aziz

(2019b) Influence of intensity & eccentricity of posttensioning force and concrete strength on shear behavior of epoxied joints in segmental box girder bridges. Construction and Building Materials 197: 117–129.

Ahmed

Aziz

(2020) Stresses, deformations and damages of various joints in precast concrete segmental box girder bridges subjected to direct shear loading. Engineering Structures 206: 110151.

Al-Sherrawi

Allawi

Al-Bayati

, et al. (2018) Behavior of precast prestressed concrete segmental beams. Civil Engineering Journal 4(3): 488–496.

Allawi

Al-Sherrawi

Al-Bayati

, et al. (2019) Experimental investigation of segmental post-tensioned girders. In: Challenges in Mechanics of Time-dependent Materials, Volume 2. Berlin, Germany: Springer International Publishing, 87–94.

American Association of State Highway and Transportation Officials (2004) AASHTO LRFD Bridge Specification. 3rd edition. Washington, DC: American Association of State Highway and Transportation Officials.

Bakhoum

(1990) Shear Behavior and Design of Joints in Precast Concrete Segmental Bridges. Cambridge, MA: Massachusetts Institute of Technology.

Buyukozturk

Bakhoum

Michael Beattie

(1990) Shear behavior of joints in precast concrete segmental bridges. Journal of Structural Engineering 116(12): 3380–3401.

10.

Choi

Lee

Yuan

, et al. (2023) Shear strength of steel fiber reinforced lightweight self-consolidating concrete joints under monotonic and cyclic loading. Construction and Building Materials 363: 129829.

11.

Dei Poli

Di Prisco

Gambarova

(1993) Cover and stirrup effects on the shear response of dowel bar embedded in concrete. Structural Journal 90(4): 441–450.

12.

Dulacska

(1972) Dowel action of reinforcement crossing cracks in concrete. Journal Proceedings 69(12): 754–757.

13.

Yuan

Wan

(2021) Behavior of fiber reinforced key joints in precast concrete segmental bridge: experimental and numerical analysis. Journal of Bridge Engineering 26(8): 04021053.

14.

GB 50010-2010 (2015) Code for Design of Concrete Structures (2015 Version). Beijing, China: China Building Industry Press (in Chinese).

15.

GB/T 228.1-2010 (2010) Metallic Materials-Tensile Testing-Part 1 Method of Test at Room Temperature. Beijing, China: China Standards Press (in Chinese).

16.

Kwan

AKH

(2001) Modeling dowel action of reinforcement bars for finite element analysis of concrete structures. Computers & Structures 79(6): 595–604.

17.

Issa

Abdalla

(2007) Structural behavior of single key joints in precast concrete segmental bridges. Journal of Bridge Engineering 12(3): 315–324.

18.

Jiang

Chen

, et al. (2015) Shear behavior of dry joints with castellated keys in precast concrete segmental bridges. Journal of Bridge Engineering 20(2): 04014062.

19.

Koseki

Breen

(1983) Exploratory study of shear strength of joints for precast segmental bridges. Research Report 248-1, Center for Transportation Research, The University of Texas at Austin, Austin, Tex., Sept. 1–83.

20.

Pham

Hao

, et al. (2019) Performance of precast segmental concrete beams posttensioned with carbon fiber-reinforced polymer (CFRP) tendons. Composite Structures 208: 56–69.

21.

Yang

Lei

(2013) Combined shear and bending behavior of joints in precast concrete segmental beams with external tendons. Journal of Bridge Engineering 18(10): 1042–1052.

22.

Wang

, et al. (2018) Seismic performance of precast concrete bridge columns with quasi-static cyclic shear test for high seismic zones. Engineering Structures 166: 441–453.

23.

(2004) Study on the Design Theory of Precast Segmental Prestressed Concrete Girders with External Tendons. Ph.D. thesis, School of Civil Engineering, Beijing Jiaotong University, Beijing, China.

24.

Zhu

, et al. (2022) Experimental study on the shear behavior of precast concrete segmental simply supported beam with internal tendons and glue joints. Case Studies in Construction Materials 17: e01545.

25.

Megally

Seible

Dowell

(2003) Seismic performance of precast segmental bridges: segment-to-segment joints subjected to high flexural moments and low shears. PCI Journal 48(2): 80–97.

26.

Nazir

Momeni

Marsono

, et al. (2013) Precast spread foundation in industrialized building sys-tem. In: Proceedings of the 3rd international conference on geotechnique, construction materials and environment, Nagoya, Japan, 13–15 November 2013, 13–15.

27.

Parastesh

Hajirasouliha

Ramezani

(2014) A new ductile moment-resisting connection for precast concrete frames in seismic regions: an experimental investigation. Engineering Structures 70: 144–157.

28.

Saibabu

Srinivas

Sasmal

, et al. (2013) Performance evaluation of dry and epoxy jointed segmental prestressed box girders under monotonic and cyclic loading. Construction and Building Materials 38: 931–940.

29.

Shamass

Zhou

Alfano

(2015) Finite-element analysis of shear-off failure of keyed dry joints in precast concrete segmental bridges. Journal of Bridge Engineering 20(6): 04014084.

30.

Takase

(2019) Testing and modeling of dowel action for a post-installed anchor subjected to combined shear force and tensile force. Engineering Structures 195: 551–558.

31.

Turmo

Ramos

Aparicio

ÁC

(2005) Shear strength of joints in precast concrete segmental bridges. ACI Structural Journal 102(6): 901.

32.

Turmo

Ramos

Aparicio

ÁC

(2006) Shear behavior of unbonded post-tensioned segmental beams with dry joints. ACI Materials Journal 103(3): 409.

33.

Wang

Zhao

, et al. (2019) Design criterion for the self-centering capacity of precast segmental UHPC bridge columns with unbonded post-tensioning tendons. Engineering Structures 200: 109706.

34.

Wrayosh

Hashim

(2020) Flexural behaviours of box segmental beams with internal tendons subjected to repeated and static loads. IOP Conference Series: Materials Science and Engineering 671(1): 012109.

35.

Yao

Cheng

(2020) Experiment of shear performance of epoxy resin joints with single keys in precast concrete segmental bridges. IOP Conference Series: Earth and Environmental Science 525(1): 012152.

36.

Al-Hussein

Nasseri

, et al. (2008) Sustainable precast concrete foundation system for residential construction. Canadian Journal of Civil Engineering 35(2): 190–199.

37.

Yuan

Yang

Wang

, et al. (2019) Shear behavior of epoxy resin joints in precast concrete segmental bridges. Journal of Bridge Engineering 24(4): 04019009.

38.

Zhan

Chen

, et al. (2022) Experimental and numerical investigations on shear performance of key tooth joints of precast concrete segmental bridge under repeated loading. Construction and Building Materials 351: 128794.

39.

Zhang

, et al. (2022) Full-scale experimental study on shear behavior of multiple-keyed epoxy joints in precast concrete segmental bridges. Structures 45: 437–447.

40.

Zhou

Mickleborough

(2005) Shear strength of joints in precast concrete segmental bridges. ACI Structural Journal 102(1): 3.

41.

Zhou

Zhang

Luo

, et al. (2022) Reliability of fully assembled precast concrete frame structures against progressive collapse. Journal of Building Engineering 51: 104362.

Experimental study on the shear behavior of joints in precast concrete segmental foundations

Abstract

Keywords

Introduction

Experimental program

Test specimens

Fabrication of the specimens

Test setup and instrumentation

Material properties

Crack patterns and failure modes of the tested specimens

Flat joint specimens

Key joint specimens

Analysis of the test results

Effect of the shear key type

Effect of the confining stress

Effect of the post-tensioned bar bond type

Effect of the joint type (with or without epoxy)

Shear resistance mechanism discussion of PCSF joints

Composition of the Vu of the PCSF joints

The Vu of key joints compared with existing formulas

Conclusions

Footnotes

Declaration of conflicting interests

Funding

ORCID iDs

References

Composition of the V_u of the PCSF joints

The V_u of key joints compared with existing formulas