Experimental and numerical investigation on failure behavior of ring joints in precast concrete shear walls

Abstract

Precast shear wall structures have been widely used due to their outstanding features, and the joints between precast members play a critical role in complete structures, specifically for vertical joints. The ring joint is a new connection method used for the vertical connection. Few studies and related regulations were traced; therefore more detailed studies are required. In order to study the anchoring performance and failure behavior, an experimental model was designed and tested under monotonic axial loading, taking the composite height of ring rebars, concrete specifications, diameter of the horizontal rebars, relative position of the ring rebars, diameter of the ring rebars, and number of horizontal rebars into consideration. The failure phenomena were observed and the data were collected. The failure pattern, bearing capacity, yield ratio, displacement ductility coefficient, and other performance parameters were analyzed. The study indicated that the failure patterns are divided into ring rebar pull-out and ring rebar fracture. Increasing the composite height of the ring rebar, the concrete specifications and the number of horizontal rebars could improve the bearing performance, and the contribution of the horizontal rebar diameter was limited, and interlocking ring rebars arranged uniformly are not optimal. In the case of joint failure, the yield ratio is relatively small and the displacement ductility coefficient is larger, which shows the bearing capacity reserve is better. A numerical model was established to analyze the internal behavior, and the results were in good agreement with the experimental results, important for us to understand the failure behavior. Design recommendations will promote its application.

Keywords

failure pattern finite element precast buildings pull-out test ring joint shear wall structure

Introduction

Precast shear wall structures, which are environment-friendly structures, have outstanding features for green buildings because of the lower construction costs, shorter construction time, less pollution, as well as better quality control (Lu, 2014). Precast buildings are widely used (Ha et al., 2016; Sun et al., 2018; Zhi and Guo, 2017), and shear walls are essential members for resisting lateral seismic loading in high-rise buildings (Lu and Yang, 2014). Because of the requirement to maintain significant lateral stiffness and resistance, precast shear walls pose specific challenges for assembling techniques. Precast shear walls are typically connected by multiple joints; therefore, the reliability of the joints has a significant influence on the structure, and although many scholars at home and abroad have conducted a lot of research, further detailed research is required.

Numerous pseudo-static experiments and theoretical studies were conducted on different connection methods of precast concrete shear walls, including welded connections, sleeve connections, bolt connections, prestressed connections, and key-tooth connections, to study their seismic performance (Soudki et al., 1995b, 1995a, 1996). A precast concrete wall-slab-wall structure system was proposed and a full-scale two-storey structure was fabricated to evaluate the seismic performance under cyclic loading; research presented a new structural form with good performance (Brunesi et al., 2018a, 2018b). Pull-out and pseudo-static experiments were conducted to evaluate the reliability of the connection method and took grouted sleeve joints (Figure 1(a)), vertical steel bar single-row indirect lap joints, and crush sleeve joints into consideration (Qian et al., 2011). A number of reasonable suggestions to achieve high reliability were proposed. A U-shaped closed rebar horizontal joint method was proposed, and a low-cycle loading test of the shear wall specimen was conducted. It was reported that a reasonably designed connection can achieve the bearing capacity, ductility, and seismic resistance equivalent to cast-in-situ connections (Liu et al., 2013). Besides, a novel grouted rolling pipe splice for precast concrete construction was proposed and numerous studies were conducted for its performance and confining mechanism (Zheng et al., 2016, 2018). A connection technique (Figure 1(b)) and a plug-in filling hole for lap-joint steel bars were tested, taking the reinforced bar diameter, anchorage length, and concrete strength into consideration; Jiang et al. (2011) reported that the plug-in filling hole for lap-joint steel bars was reliable. In traditional connections mentioned above, expensive connecting equipment and complicated joining processes greatly increase the construction cost, and the uncontrollable construction quality is very likely to cause anchor failure.

Figure 1.

Schematic diagram of traditional connection in precast buildings: (a) a plug-in filling hole for lap-joint and (b) sleeve connections.

A pull-out test of U-shaped steel in plain concrete and a pseudo-static test including three shear wall specimens were conducted (Yu et al., 2015), considering horizontal and vertical connections, to study the seismic performance of U-type reinforcement ferrule connections. An innovative joint connecting beam for concrete shear wall structures was proposed (Lu et al., 2016) and conducted studies take height–width ratio, height of joint connecting beam, and longitudinal bars into consideration, with the connecting beam in the middle of the walls. The results indicated that it performed satisfactorily.

As opposed to the commonly used sleeves, pulp anchors, and mechanical connections in precast shear wall structures, based on the above connecting methods, a new ring joint connection method was proposed for vertical connection in components. It comprised interlocked ring rebars and inserted horizontal rebars, and the connection area was poured on site, then it formed a continuous beam or column to strengthen the joint areas effectively, as shown in Figure 2(a). The method reduces the lap length and the amount of cast-in-situ concrete, and it not only saves the sleeve cost but also has no hidden operation which ensures the quality of the connection. Importantly, the joint realized the connection of all the stressed rebars in shear walls and the connection of regional body rather than the traditional point or line connections, and it improved the security and reliability. A strong connection can be designed depending on the intended purpose, without the possibility of anchorage failure.

Figure 2.

Schematic diagram of (a) ring joint and (b) specimen details.

The joints between precast members are manufactured on site usually, specifically for longitudinal reinforcements. Therefore, the reliability and quality of connections are crucial to the architecture. To study the security and reliability, the pseudo-static in-plane tests (Jiao et al., 2015) and the flexural out-of-plane tests (Gao et al., 2016) of the precast shear wall members with ring joints were studied. Other exploratory experimental studies (Lu et al., 2016; Yu et al., 2015) were conducted. The above research indicated that the mechanical properties of the ring joint could meet seismic requirements in China. However, existing research mainly focused on macroscopic aspects, such as seismic performance evaluation of test pieces under cyclical loading and lacking of research on internal or failure behavior. So, there remained a requirement to conduct the research on ring joint.

This study addresses pull-out tests under monotonic axial loading. A total of 36 specimens with 6 parameters were tested to evaluate bearing capacity, failure patterns, and ductility. A finite element model demonstrated a detailed failure behavior combined with the experimental research, helping us better understand the mechanical mechanism.

Test program

To investigate the bearing performance, experiments were conducted on 36 specimens. This section presents the details of the specimens and the loading protocol used in the study. The tests were used to examine the mechanical properties under loading and validate the bearing capacity performance.

Model design and production

Three basic connection units were selected to form the specimen considering the characteristics of the joint, and the test piece dimensions were 400 × 200 × 400 mm³. A schematic model of Group A3 is used in Figure 2(b) to show the internal details, and this configuration was tested to provide reference data of the bearing capacity performance. The specimen test parameters are presented in Table 1, including composite height (CH), concrete specifications, diameter of the horizontal rebars, relative position of the ring rebars, diameter of the ring rebars, and the number of horizontal rebars.

Table 1.

Design list of specimens.

Group	Concrete	Ring rebar, ϕ (mm)	Relative position	Horizontal rebar, ϕ (mm)	Composite height, H (mm)
A1	C30	10	Adjacent (Figure 2)	8	60
A2	C30	10	Adjacent	8	80
A3	C30	10	Adjacent	8	100
A4	C30	10	Adjacent	8	160
B1	C25	10	Adjacent	8	100
B2	C35	10	Adjacent	8	100
B3	C40	10	Adjacent	8	100
C1	C30	10	Adjacent	10	100
C2	C30	10	Adjacent	12	100
D	C30	10	Interval (Figure 9)	8	100
E	C30	12	Adjacent	8	100
F	C30	10	Adjacent	8	100

Groups A1 and F were single row for horizontal rebars, and the others were double rows.

Properties test

The material tests were conducted on concrete and steel rebars. A cubic test piece was prepared while manufacturing the specimens, and those were cured under the same conditions as the test pieces. On the day of testing, a compressive strength test was conducted on the test pieces and the results are presented in Table 2. Table 3 presents the tensile test results of steel coupons that were manufactured from the same batch of steel as the rebars.

Table 2.

Measured concrete cubic compressive strength.

Specification	C25	C30	C35	C40
Compressive strength, f_c (MPa)	31.94	37.04	39.44	45.93

Table 3.

Mechanical properties of steel rebar.

Diameter, ϕ (mm)	Yield strength, f_y (MPa)	Ultimate strength, f_u (MPa)	Yield strain, με
8	435.6	661.6	2178
10	436.5	668.7	2237
12	457.9	629.3	2255

Test setup

The same test setup was applied for all specimens. A reaction rack loading device was specifically designed for the joint characteristics and to accommodate the test piece, and it was placed on the top of the piece. A hydraulic jack (±300 kN, +100 mm) and force sensor (±300 kN) were located on top of the reaction rack. In order to achieve the simultaneous operation of the two ends of the up-ring rebar, a connecting steel plate was welded to the up-ring rebar at a distance of 100 mm of the concrete in each test piece. A connector comprising two steel plates and four bolts connected the test piece and the steel rod. All of the components were connected by the steel rod in series to apply tension in the work of the jack, as shown in Figure 3.

Figure 3.

Test setup and measuring equipment.

Measurement plan

A hydraulic pump was used to control the hydraulic jack for applying a tensile load, and the force control loading procedure was adopted; besides, preloading to 5 kN was performed before the test started, and then the formal loading started until the test piece was destroyed. The displacement measuring points were located on either side of the reaction force rack, and the displacement of the steel plate connected to the up-ring rebar was measured by a linear variable differential transformer (LVDT) during the loading. The force was measured by the force sensor, and the strains of the steel rebar were measured by strain gages attached in critical locations, as shown in Figure 10(a). The displacement, force, and strain were measured simultaneously during the loading.

Phenomena and patterns

During the test, the damage process, bearing capacity, and failure patterns were observed and recorded. All of the specimens exhibited the three stages of elasticity, elastoplasticity, and plastic failure. The cracks in the specimens became more fully developed with increased loading. This phenomenon was easily observed, and the failure modes could be reasonably determined. The failure patterns and crack propagation were similar in all specimens in Groups A1 and F, and the failure patterns were divided into ring rebar pull-out and ring rebar fracture. The observed performance of specimens A11 and A33 are presented in the following sections.

Ring rebar pull-out

All specimens in Group A1 and F11 in Group F suffered up-ring rebar pull-out; the destruction process and modes were essentially identical; and A11 was chosen as being representative and is presented in this section. In the initial stage of loading, the specimen was in the elastic phase with no visible cracks. When loaded to 40 kN, the concrete around the up-ring rebar gradually peeled off, and short small cracks appeared close to the up-ring rebar. In addition, a number of cross cracks were observed on the surface of the concrete between the up-ring rebars. When loaded to 43 kN, “˅”-shaped cracks appeared on both sides of the specimen. With increasing load, a great number of transverse cracks appeared. When loaded to 60 kN, the cracks developed rapidly, and the crack width reached 5 mm, and, at 70 kN, there were more obvious vertical run-through cracks at the lower ring rebar position. When loaded to 81 kN, the bond force of the up-ring rebar failed and pulled out, and the specimen lost its bearing capacity; however, the up-ring rebars did not fall off under the action of the horizontal rebars. Concrete cracks with widths greater than 5 mm and concrete spalling were observed. The crack and failure patterns are shown in Figure 4(a).

Figure 4.

Two typical failure patterns of specimens: (a) crack and failure pattern of A11; (b) crack and failure pattern of A33.

Ring rebar fracture

Up-ring rebar fractures were observed in all other groups, with similar destruction processes and fracture patterns, and A33 was selected as a representative example (Figure 4(b)). The destruction phenomena of the specimens were as follows. The specimen was initially in the elastic phase, as was specimen A11. When loaded to 50 kN, small cracks appeared around the up-ring rebar and developed into transverse cracks, and at 55 kN, “˅”-shaped cracks were observed on both sides. These cracks propagated with increased loading. Horizontal cracks appeared on the down horizontal rebars and propagated at a load of 58 kN, and the horizontal rebar yielded at 63 kN. At a load of 80 kN, cracks developed rapidly, and the up-ring rebars entered the yield phase until one side of rebar fractured; 96 kN was recorded as the ultimate bearing capacity of the specimen. The final failure patterns are shown in Figure 4(b).

Discussion of test results

Force versus displacement relationship

The force versus displacement responses of the specimens are shown in Figure 5. The displacement was taken as the average of the displacement gauges on either side of the up-ring steel connecting plate. The F-Δ curve is the basis for analyzing its mechanical properties and establishing mathematical models. Key features are marked in the figures, such as the yield load, ultimate load, and corresponding displacements.

Figure 5.

Load–displacement curves of specimens.

The curves are initially linear, and the test piece is in a flexible phase. The gradient of the curve decreases, and the test piece enters the elastoplastic stage with increasing loading. In the plastic stage, the curves of the pull-out specimens decreased abruptly, losing bearing capacity. The curves of the rebar fractured specimens showed an increasing trend before the plastic stage fracture. The curves tend to be stable horizontally without abrupt changes until the rebar fractures. Small discrepancies in the curves indicate that the bearing performance is stable and reliable, and it also illustrates the stability of the connection technology.

Bearing capacity and deformability

The detailed results of test piece are presented in Table 4. The yield ratio that reflects the reliability of bearing capacity is the ratio of yield strength to tensile strength (F_y/F_u). The displacement ductility coefficient (μ = Δ_u/Δ_y) that reflects the deformation performance is a critical index for measuring the seismic performance of the concrete structures.

Table 4.

Test analysis results.

No.	F_y	AVG	Δ_y	AVG	F_u	Δ_u	F_y/F_u	AVG	Δ_u/Δ_y	AVG	P/F
A11	38.06	37.71	0.82	0.83	90.64	12.81	0.42	0.46	15.62	15.77	P
A12	38.36		0.83		81.06	18.90	0.47		22.77		P
A13	36.71		0.84		76.41	7.50	0.48		8.93		P
A21	50.26	49.58	0.87	0.82	96.16	18.83	0.52	0.51	21.64	22.95	F
A22	50.26		0.77		97.04	17.79	0.52		23.10		F
A23	48.23		0.82		97.04	19.76	0.50		24.10		F
A31	50.55	51.42	0.75	0.81	92.39	18.28	0.55	0.54	24.37	23.40	F
A32	50.84		0.95		94.42	20.07	0.54		21.13		F
A33	52.86		0.72		96.45	17.79	0.55		24.71		F
A41	53.46	54.62	1.04	0.92	95.58	21.74	0.56	0.58	20.90	23.70	F
A42	54.62		0.86		95.01	22.12	0.57		25.72		F
A43	55.78		0.86		94.13	21.05	0.59		24.48		F
B11	40.67	41.74	0.45	0.61	96.16	21.48	0.42	0.43	47.73	36.72	F
B12	43.87		0.6		95.87	19.78	0.46		32.97		F
B13	40.67		0.78		95.87	22.99	0.42		29.47		F
B21	51.71	49.48	0.73	0.81	95.58	21.61	0.54	0.51	29.60	26.28	F
B22	50.26		0.76		96.45	21.16	0.52		27.84		F
B23	46.48		0.95		96.45	20.32	0.48		21.39		F
B31	50.55	53.17	0.88	0.99	97.04	19.58	0.52	0.55	22.25	23.14	F
B32	54.91		0.72		97.91	22.45	0.56		31.18		F
B33	54.04		1.36		94.71	21.73	0.57		15.98		F
C11	46.19	51.52	0.72	0.74	95.87	20.51	0.48	0.54	28.49	28.34	F
C12	58.69		0.78		95.29	18.64	0.62		23.90		F
C13	49.68		0.72		96.45	23.49	0.52		32.63		F
C21	47.07	50.70	0.48	0.61	93.26	15.64	0.50	0.54	32.58	31.76	F
C22	54.33		0.73		95.29	22.78	0.57		31.21		F
C23	50.71		0.61		94.13	19.21	0.54		31.49		F
D11	41.55	46.00	0.52	0.59	94.13	19.74	0.44	0.48	37.96	35.82	F
D12	56.36		0.74		96.45	17.96	0.58		24.27		F
D13	40.09		0.50		93.84	22.62	0.43		45.24		F
E11	62.75	66.10	0.8	0.93	132.19	28.37	0.47	0.50	35.46	30.22	F
E12	69.44		1.05		133.64	26.81	0.52		25.53		F
E13	66.10		0.93		132.92	27.59	0.50		29.67		F
F11	39.22	40.68	0.37	0.47	89.77	10.47	0.44	0.44	28.30	32.70	P
F12	42.13		0.56		96.16	20.59	0.44		36.77		F
F13	40.68		0.47		92.97	15.53	0.44		33.04		F

P: ring rebar pulled-out; F: ring rebar failure.

Table 4 presents the detailed test data at the main stages. The F_y of the fracture specimens ranging from 40.68 to 69.44 MPa is better than those of the ring pull-out specimens that range from 36.71 to 39.22 MPa. The F_u of the fracture specimens ranging from 92.97 to 133.64 MPa is also better than those of the ring pull-out specimens that range from 81.06 to 90.64 MPa. The displacement ductility factor, u, ranges from 22.95 to 36.72, except for A1 that is 19.12.

Parametric analysis

As shown in Table 1, the experimental performance studies were carried out considering six parameters, and the influence was analyzed from the yield strength, yield ratio, and ductility coefficient, respectively, and the results are shown in this section.

Composite height

Four levels were used to study the effects of CH on the bearing performance: 60, 80, 100, and 160 mm, as shown in Groups A1–A4 in Table 1. With a single-step increase in CH, the yield load increased from 4% to 31% (Figure 6(a)). It can be seen in Figure 6(b) that the yield ratio increases linearly with increasing CH, with increases of 6%–16% per stage. The displacement ductility coefficient increases with increasing CH, with increases of 1%–20% per stage, as can be seen in Figure 6(c). The damage process was observed, and the crack development was significantly reduced with increasing CH. Therefore, increasing the CH could improve the anchoring performance of the ring rebars to concrete, thereby improving the bearing performance of the connection.

Figure 6.

Result analysis of composite height tests: (a) yield strength analysis, (b) yield ratio analysis, (c) and ductility analysis.

Concrete specification

In order to study the effect of the concrete specification on the bearing performance, four groups were tested: C25, C30, C35, and C40, as shown in Groups A3 and B1–B3 in Table 1. With a single-step increase in concrete specification level, the yield load increased by 3%–19%, as shown in Figure 7(a). It can be seen in Figure 7(b) that the yield ratio increases by 1%–18% per stage. The displacement ductility coefficient shows a decreasing trend with a decrease of 1%–28% per stage (Figure 7(c)). Improving the concrete specification is an enhancing factor for the bearing capacity, but the ductility coefficient of the test pieces decreased because of the increase in stiffness.

Figure 7.

Result analysis of concrete specification tests: (a) yield strength analysis, (b) yield ratio analysis, and (c) ductility analysis.

Horizontal rebar diameter

To investigate the effect of the horizontal rebar diameter on the bearing performance, three diameters were tested: 8, 10, and 12 mm, as shown in Group A3 and C1–C2 in Table 1. As can be seen in Figure 8(a) and (b), the change in the diameter has little effect on the yield load and the yield ratio, with essentially horizontal linear relationships. Increases in the diameter made the joint rigid and the yield displacement decreased, and Figure 8(c) shows that the displacement ductility coefficient increases from 12% to 21% per stage. It can be concluded that the horizontal rebar diameter has a negligible effect on the improvement of bearing capacity as the “core” area formed by the horizontal rebars already meets the shear force requirements, indicating that the ring joint can be designed reliably.

Figure 8.

Result analysis of horizontal rebar diameter tests: (a) yield strength analysis, (b) yield ratio analysis, and (c) ductility analysis.

Ring rebar position

Different ring rebar positions were tested, with adjacent (Figure 2(b)) and interval arrangements (Figure 9(a)), as shown in Groups A3 and D in Table 1. The interval arrangement resulted in an 11% decrease in the yield ratio and a 53% increase in the displacement ductility factor. The interval arrangement also exhibited an 11% decrease in the yield load, as shown in Table 4. As can be seen in Figure 9(b) and (c), increasing horizontal cracks in the lower horizontal rebars and vertical cracks in the down-ring rebars were observed on both sides of the specimens. Therefore, it can be concluded that the interval arrangement does not significantly increase the bearing capacity, and interval arrangement undergoes a greater deformation during the loading process, which is an unfavorable factor.

Figure 9.

Crack patterns of uniform arrangement specimen: (a) interval arrangement, (b) D11 specimen, and (c) D13 specimen.

Ring rebar diameter

A comparison Group E with a diameter of 12 mm was designed, related to the baseline Group A3. As the ring rebar diameter increased, the yield load, yield ratio, and ductility coefficient increased by 29%, 11%, and 29%, respectively. A load increase with increasing diameter resulted in a greater crack development. In summary, increasing the ring rebar diameter could effectively increase the bearing capacity to some extent, which indicates that the joint could withstand greater force.

Horizontal rebar number

The effects of single-row and double-row horizontal rebars on the bearing performance were studied, as shown in Groups A3 and F in Table 1. Single-row rebars were arranged in the middle of the CH, and double-row rebars in the four corners. The yield load and yield ratio decreased by 21% and 20%, respectively, for the single-row arrangement. Two failure patterns were observed in this group indicating that the performance was unstable and uncontrollable. The single-row arrangement did not exhibit satisfactory bearing capacity performance; therefore, double-row designs are recommended.

Failure behavior analysis

Model verification

Finite element models were constructed using the commercial program ABAQUS to investigate the internal behavior during loading. The three-dimensional, eight-node solid elements with reduced integration (C3D8R; Nascimbene, 2014) and concrete damaged plasticity material were used for concrete (Li and Wu, 2005). Two-node Timoshenko beam elements with 6 DOFs at each node (B31) for rebars and the parameters of concrete are shown in Table 5. Additional related material parameters were obtained through experiments. Non-linearities, including materials and geometry, were taken into consideration. Considering the complexity of the connection technology, the slip model will greatly increase the analysis difficulty and the contact is more likely to cause unit distortion and non-convergence calculation. Finally, the contact relationship between rebars and concrete was realized by the embedded method. The boundary conditions were consistent with the actual tests, and the constraint was imposed on the contact portion between the top of the specimen and the reaction rack. The models were subjected to constant vertical tension under load control in ABAQUS/Standard.

Table 5.

Parameters of concrete damaged plasticity material.

Compressive behavior		Tensile behavior		Plasticity
Inelastic strain (ε)	Yield stress (MPa)	Cracking strain (ε)	Yield stress (MPa)	Content	Value
0	29.03	0	2.59	Dilation angle	30
0.00002	29.44	0.00012	2.01	Eccentricity	0.1
0.00033	34.34	0.00024	1.58	fb₀/fc₀	1.16666
0.00081	36.78	0.00036	1.37	K	0.66666
0.00121	37.04	0.00048	1.27	Viscosity parameter	0
0.00165	36.81	0.00060	1.26
0.00236	35.42
0.00399	29.71
0.00637	21.86
0.00785	18.50
0.00994	15.28
0.01129	13.85

Figure 10(b)–(d) shows the validation of the strain values from the A, B, and P positions shown in Figure 10(a) by comparing them to the results from benchmark Group A33 specimen. The measured forces were distributed to ring rebars according to the strain ratio. The curves were essentially consistent with the experimental results, specifically in the initial phase, which means the connection can transfer the load effectively. The simulation results were in good agreement, and the simulation can be used for further analysis, which could help in better understanding the internal failure behavior visually.

Figure 10.

Strain value verification of simulation results: (a) strain gauges, (b) strain of A1 and A2, (c) strain of B1 and B2, and (d) strain of P1 and P2.

Failure behavior

Detailed changes of the concrete and the failure behavior of rebars could be clearly observed during the simulation combining the simulation analysis results with the experimental results. This discussion focuses on the damage process of two failure patterns giving the equivalent plastic strain (PEEQ) of concrete and the von Mises stress of rebars. Specimens A11 and A33 were taken as representative specimens.

Ring rebar pull-out

Specimen A11 was used as an example in this section. In the initial stage, the up-ring rebars bore the tensile force, and the rebars and the concrete were co-stressed. At this point, no obvious cracks were visible. With increasing loading, the concrete around the up-ring rebars tended to peel off, and a number of cracks gradually appeared between the rebars at the top of the concrete. When the load reached a certain value, more cracks were observed at the top of the specimen. In addition, because of the overlap action of the ring rebars and the dowel effect of the horizontal rebars, the middle section (position B in Figure 10(a)) of the up-ring rebars transferred a portion of the force to the down-ring rebars (position C in Figure 10(a)), as shown in Figure 11(d). A “˅”-shaped deformation of the horizontal rebars was observed in Figure 11(b) and (d), and an increased strain in the horizontal rebars was observed with increasing loading. The up-ring rebars then yielded and the specimen entered the plastic phase, as shown in Figure 11(a) and (c). Finally, the specimen lost all bearing capacity and a large deformation was observed; a “˅”-shaped concrete damage occurred at the top of the concrete, and cracks appeared on both sides.

Figure 11.

Failure pattern and simulation results of specimen A11: (a) pull-out damage, (b) rebar deformation, (c) PEEQ of plasticity, and (d) von Mises stress.

Ring rebar fracture

Specimen A33 was used as a representative specimen in this section. In the initial stage, the specimen was in the flexible phase. A number of cracks around the up-ring rebars indicated the concrete peel off from the rebars. With increasing loading, the middle section (position B in Figure 10(a)) of the up-ring rebars transferred a portion of the force to the down-ring rebars (position C in Figure 10(a)) by the overlap action of the ring rebars and the dowel effect of the horizontal rebars. Following this, the strain on the up-ring rebars increased abruptly, and the strain on the down-ring rebars began to increase, with the load gradually being transmitted downward. A “˅”-shaped damage was observed in the concrete at the top of the test piece during loading (Figure 4(b)). More importantly, the four horizontal rebars formed a “core” zone bearing the shear action in the direction of the ring rebars, and an increasing strain in the horizontal rebars observed with increasing loading. The up-ring rebars yielded and then entered the strengthening phase. Finally, the up-ring rebars fractured and were destroyed (Figure 12(a)).

Figure 12.

Failure pattern and simulation results of specimen A33: (a) rebar fracture of ring rebar, (b) PEEQ of plasticity, and (c) strain of interval rebars.

The concrete damage was primarily around the ring rebars, and a number of “˅”-shaped cracks appeared on the sides as shown in Figures 6 and 12(b). The horizontal rebars in the “core” area were still in a flexible state (Figure 12(c)). According to the analysis, it could be concluded that the failure behavior of ring joint was obvious, safe, and reliable.

Design suggestion

Based on the above research, it can be found that the specimens had two failure patterns as follows: the ring rebars were pulled out and fractured. Obviously, the latter was a desired pattern, showing better anchoring performance, and the ring rebars can play the role of distribution reinforcement in the shear wall. Different from the traditional anchorage of steel rebars, the anchoring performance of the ring rebars studied in this article is completely different, which is also the focus of this section.

Anchorage length in current specification

Both Chinese and American specifications gave the form and technical requirements for hooks and mechanical anchoring in concrete structures. In GB 50010, the anchorage length can be calculated according to the following formulas, in which different forms have different correction factors

l_{a b} = α \frac{f_{y}}{f_{t}} d

(1)

l_{a} = ζ_{a} l_{a b}

(2)

In ACI 318 (ACI, 2014), the anchorage length can be calculated according to the following formula

l_{a} = (\frac{f_{y} ψ_{t} ψ_{e}}{25 λ \sqrt{{f^{'}}_{c}}}) d_{b}

(3)

In ordinary concrete structure, λ = 0; in uncoated reinforcement structure, ψ_e = 1.0; if the steel rebars fully exert tensile properties, then ψ_t = 1.0.

Now that two countries have no relevant regulations on the form of the ring rebar anchorage, and the anchorage length can be calculated with reference to the provisions of the 90° hook mechanical form in the specification. The ring joint diagram is presented in Figure 13.

Figure 13.

Ring joint diagram: (a) vertical connection and (b) anchorage length.

Taking Group A3 as an example, the anchorage length discussion is shown in Table 6; obviously, the anchorage length did not meet the requirements simultaneously. However, the experimental and numerical results showed that the ring joint had good anchoring performance for ring rebars, which indicated that the anchoring mechanism and failure mode of the joint were different from the existing straight and mechanical anchorage forms.

Table 6.

Anchorage length discussion.

Content	Concrete specifications		Note
Content	China concrete code	ACI 318-11	Note
Basic anchor length	36d	42d	d = 10 mm
Mirrored anchor length	16.5d	16.5d	l_a₁ = l_v + 0.5l_h
Ultimate anchor length	22d	29d	$l_{a}^{'}$ = ζ l_a
Whole anchor length	24d	24d	l_a₂ = l_v + l_h
Satisfy or not	l_a₁ < $l_{a}^{'}$ , l_a₂ < $2 l_{a}^{'}$	l_a₁ < $l_{a}^{'}$ , l_a2 < $2 l_{a}^{'}$

ACI: American Concrete Institute.

Design recommendations

Drawing on the research results of bent rebars (Shao et al., 1987), research shows that the bent rebar can meet the requirements of anchoring performance under reasonable design when the bent section meets 0.5l_h ≥ 7d, the bond–slip curve rises and is stable, and the stiffness design requirements are met. It is recommended that the horizontal part of ring joint, l_h, should meet 0.40l_a. For ring joint, the horizontal rebars yielded after the ring rebars yielded, and the yield load increased with the increase in the CH; when the CH was greater than 16d, the horizontal rebars no longer yielded, as shown in Figure 14. Therefore, the vertical length should meet 0.45l_a to ensure the strength and stiffness design requirements.

Figure 14.

Load curves of horizontal rebars.

Suggestions are proposed for application and practice. For ring joint, a ring rebar can be divided into two bent rebars symmetrically, and the free end of the bent rebar becomes the restrained end; besides, the dowel action of the horizontal rebars can significantly improve the anchoring performance (Li et al., 2017). Based on the above research, the recommendations are as follows: for anchor length, the vertical length l_v ≥ 0.45l_a, the horizontal length l_h ≥ 0.40l_a, and the total anchorage length 2l_v + l_h ≥ 1.30l_a; the interlocking rebars should be arranged adjacent to each other within 25 mm, and the CH ≥ 0.45l_a; the horizontal rebars should be the same to the ring rebars, and at least four rebars should be arranged at corners in the ring-shaped closed area; the concrete specification should be greater than C25, and its strength level should be one level higher than that of the walls. All these could ensure the reliability of the connection.

Conclusion

This study presented monotonic axial pull-out tests of ring joints in concrete shear walls. A total of 36 specimens were tested to evaluate the bearing capacity behavior. The following conclusions can be drawn from the results and analysis:

Typical failure patterns, including up-ring rebar pull-out and rebar fracture, and a significant number of horizontal and oblique cracks, were observed on the specimens. The fracture specimens exhibited ideal destruction patterns, and the horizontal rebars in the joints remained in the flexible stage.

All of the specimens exhibited essentially similar hysteretic curves, which indicated the stages of elasticity, elastoplasticity, and plasticity. The specimens had a smaller yield strength, a greater displacement ductility coefficient, and a better bearing capacity reserve. More importantly, smaller dispersions indicate that the joints are stable and reliable.

A parameter analysis showed that increasing the CH, concrete specification, ring rebar diameter, and horizontal rebar rows could improve the bearing performance of the ring joint. Increases in the horizontal rebar diameter did not improve the performance. In addition, an adjacent arrangement is recommended instead of an interval arrangement.

The simulation results were in good agreement with the experimental results, which means the finite element model and method were feasible and reasonable. More importantly, the method can more intuitively reflect the internal failure behavior of the ring joint.

Suggestions including anchorage length, relative position of the interlocking rebars and ring units, and other content were proposed for practice, which could guarantee the anchorage performance and ensure the reliability of the connection.

Thus, the aim of the study was to explore the influencing factors and the internal failure behavior of the ring joint in precast shear walls and propose design suggestions for practice. The study found that the ring joint can transfer the load from the up-ring to down-ring, and the joint is stable and reliable.

Footnotes

Appendix 1 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 study was financially supported by the Special Project of National Key Research and Development Plan (No. 2017YFC0703804). The help of China Construction Seventh Engineering Division Corp. Ltd is sincerely appreciated.

ORCID iD

Jian Zhou

References

American Concrete Institute (ACI) (2014) 318-11 Building Code Requirements for Structural Concrete and Commentary. Farmington Hills, MI: American Concrete Institute.

Brunesi

Peloso

Pinho

, et al. (2018a) Cyclic testing of a full-scale two-storey reinforced precast concrete wall-slab-wall structure. Bulletin of Earthquake Engineering 16(11): 5309–5339.

Brunesi

Peloso

Pinho

, et al. (2018b) Cyclic testing and analysis of a full-scale cast-in-place reinforced concrete wall-slab-wall structure. Bulletin of Earthquake Engineering 16(10): 4761–4796.

Gao

Jiao

Zhang

, et al. (2016) Tests on out-of-plane bending of precast shear walls with annular closed reinforcements. Building Structure (S2): 439–442. (in Chinese). Available at: http://en.cnki.com.cn/Article_en/CJFDTotal-JCJG2016S2094.htm

GB 50010:2010 (2010) Code for design of concrete structures (in Chinese).

Kim

Lee

, et al. (2016) Performance evaluation of semi precast concrete beam-column connections with U-shaped strands. Advances in Structural Engineering 17(11): 1585–1600.

Jiang

Zhang

Liu

, et al. (2011) Experimental study on plug-in filling hole for steel bar lapping of precast concrete structure. Journal of Harbin Institute of Technology 43(10): 18–23. (in Chinese)

Jiao

Zhang

, et al. (2015) Tests on seismic behavior of pre-cast shear walls with annular closed reinforcements. Journal of Building Structures 36(5): 103–109. (in Chinese)

(2005) Elastoplastic damage constitutive model for concrete based on damage energy release rates, part I: basic formulations. China Civil Engineering Journal 38(9): 14–20. (in Chinese)

10.

, et al. (2017) Three-dimensional bond model considering the coupled damage effect for dowel action. Magazine of Concrete Research 69(14): 728–744.

11.

Liu

Chen

Guo

, et al. (2013) Test on seismic performance of precast concrete shear wall with U-shaped closed reinforcements connected in horizontal joints. Journal of Southeast University 43(3): 565–570. (in Chinese)

12.

(2014) Precast concrete structures in the future. Structural Concrete 15(1): 1–2.

13.

Yang

(2014) Seismic behavior of T-shaped steel reinforced concrete shear walls in tall buildings under cyclic loading. The Structural Design of Tall and Special Buildings 24(2): 141–157.

14.

Wang

, et al. (2016) An innovative joint connecting beam for precast concrete shear wall structures. Structural Concrete 17(6): 972–986.

15.

Nascimbene

(2014) Towards non-standard numerical modeling of thin-shell structures: geometrically linear formulation. International Journal for Computational Methods in Engineering Science and Mechanics 15(2): 126–141.

16.

Qian

Yang

Qin

, et al. (2011) Test on seismic behavior of pre-cast shear walls with various methods of vertical rein-forcement splicing. Journal of Building Structures 32(6): 51–59. (in Chinese)

17.

Shao

Shen

(1987) Anchorage reliability and anchoring design of reinforced concrete. Journal of Building Structures 8(4): 36–49. (in Chinese)

18.

Soudki

Rizkalla

Dalkiw

(1995b) Horizontal connections for precast concrete shear walls subjected to cyclic deformations part 2: prestressed connections. PCI Journal 40(5): 82–96.

19.

Soudki

Rizkalla

Leblanc

(1995a) Horizontal connections for precast concrete shear wall subjected to cyclic deformations part 1: mild steel connections. PCI Journal 40(4): 78–96.

20.

Soudki

West

Rizkalla

, et al. (1996) Horizontal connections for precast concrete shear wall panels under cyclic shear loading. PCI Journal 41(3): 64–80.

21.

Sun

Qiu

Jiang

(2018) Lateral load behaviour of a rectangular precast shear wall involving vertical bolted connections. Advances in Structural Engineering 22(5): 1211–1224.

22.

Peng

Guo

, et al. (2015) Seismic performance of precast concrete shear wall with U-type reinforcements ferrule connection. Journal of Zhejiang University 49(5): 975–984. (in Chinese)

23.

Zheng

Guo

Guan

, et al. (2018) Parametric study on a novel grouted rolling pipe splice for precast concrete construction. Construction and Building Materials 166: 452–463.

24.

Zheng

Guo

Liu

, et al. (2016) Performance and confining mechanism of grouted deformed pipe splice under tensile load. Advances in Structural Engineering 19(1): 86–103.

25.

Zhi

Guo

(2017) Experimental evaluation of precast concrete sandwich wall panels with steel-glass fiber-reinforced polymer shear connectors. Advances in Structural Engineering 20(10): 1476–1492.