Cyclic tests on two novel welded flange plate connections in precast concrete moment resisting frames

Abstract

Precast concrete moment resistant beam-to-column connections, usually designed in form of semi-monolithic connections which require in-situ concrete casting leading to a complicated assembling process. Additionally, using concrete with different ages in a connection causes cold joints and shrinkage problems. To overcome these issues, in this study, two steel beam-to-column connections (SBCs) were proposed and designed for precast concrete moment resisting frames. Compared to the semi-monolithic connections, proposed connections exclude protruding bars, resulting in a facile and rapid installation. Regarding the inverse T shape, these connections provide a better condition for installing the precast slabs and consequently reduce structural floor thickness. To evaluate the connections, two precast and one in-situ specimens were made and tested on a 2/3 scale and a cyclic load pattern was used to evaluate their performance. Test results indicate that the proposed SBCs have a maximum resistance and ductility until 1.5 and 0.9 times the in-situ concrete beam-to-column connection (ICBC), respectively. ACI374.1-05 criteria including strength and stiffness degradation ratio as well as energy dissipation ratio at drift 4% were employed to evaluate the qualification of the proposed connections. One of the proposed connections met all seismic requirements of ACI374.1-05. The plastic hinge locations were estimated by measuring the strains of rebars and steel parts of the connections. Results revealed that on the contrary to ICBC, the plastic hinge formation in proposed SBCs developed out of the beam-column joint which coincides with the strong-column and weak-beam principle in precast frames and postpones the total fracture mechanism of structure.

Keywords

Flange plate connection precast connection cyclic test precast concrete frame beam-to-column connection

Introduction

Precast concrete columns and beams have better quality than in-situ concrete members and save time and costs for projects. However, using precast concrete frames in high seismic zones is limited due to the unreliable performance of their connections. The criteria of seismic provisions for the structures located in these zones entail high integrity between beam and slabs to create rigid diaphragm at the story level.

Conventional precast connections (dry connections) that usually are used with corbels did not have a successful experience at recent earthquakes and have not indicated enough ductility and energy dissipation. For instance, Rodriguez and Torres (2014) studied conventional precast moment resistant connections that end anchored plates were used to beam-column joint. These plates were joined by welded bars. The results showed that welded bars lead to a decrease in ductility and premature failure in the connection region. Actually, unlike the in-situ connections in which the plastic hinge zone gets larger as the drift ratio increases and provides availability for the members to experience higher deformation, dry connections have damage concentration on the location of plastic hinge formation (Liu et al., 2017). Due to the lack of structural continuity as a result of these connections weakness, progressive collapse is a common failure mechanism for the structures built through them (Elsanadedy et al., 2017). Hence, developing a design approach to transfer the plastic hinge to the outside of the beam-column joint could assist in more structural continuity and ductility. On the other hand, presenting the dry connections, even with comparable performance to in-situ ones, were usually without considering the installation requirements of the precast beam-slab connection (Shufeng et al., 2018). In addition, the application of the stiffeners or corbels in these connections reduces the ceiling height of the story (Aninthaneni et al., 2018). Unfavorable performance of dry connections made several researchers use semi-monolithic connections as an alternative for them. In these connections, beams and columns were fabricated in semi-precast form and after getting reinforced were cast in place. Utilizing these connections is usually carried out without considering the installation process requirements and there is no bearing surface for placing the precast beam (Dong et al., 2018; Yan et al., 2018).

By developing more trust in new materials such as fiber reinforced concrete and ultra-high-performance concrete through the successful experiments on their performance (Dadmand et al., 2020a,b; Pourbaba et al., 2018, 2019), some researchers used these materials in their proposed connections (Deng et al., 2020). In recent studies, the application of post-tension tendons (Guan et al., 2019; Song et al., 2021), headed bars and headed studs (Vella et al., 2018) were also suggested. Most research on semi-monolithic joints indicates their relatively good performance (Esmaeili and Ahooghalandary, 2019). However, the application of structural in-situ concrete to achieve acceptable performance contradicts the primary objective of the precast approach and leads to longer execution time and more costs. Additionally, employing complicated and multi-layer reinforcement at the connection joint makes the installation process difficult (Gou et al., 2018). In most cases, there is not any reference about the connection of precast beams to the precast slabs and their requirements, especially in high seismic zones. Furthermore, in many studies, bereft of information about the precast members’ transportation process and assembling sequence is worth noting. Conversely, much research showed that steel beam-to-column connections (SBCs) in steel MRFs (Zhang et al., 2019) and composite structures (Eghbali and Mirghaderi, 2017) have easy fabrication and installation processes.

To address the pre-mentioned setbacks, the current study introduces new SBCs with acceptable seismic performances similar to semi-monolithic ones simultaneous with providing a better condition for beam-slab connection assembling through the geometry of inverse T-shape beam. Considering the premature failure of dry connections in precast structures, the proposed connections were designed in such a way that their steel parts remained elastic and plastic hinges were located in the precast concrete beams. This approach prevents premature failure in the joint area at large drift ratios. In addition, the plastic hinge was transferred into the concrete precast beam that accords with the strong-column and weak-beam principle and postpones the total fracture mechanism of the precast frame. It is similar to the approaches used in the reduced steel beam section (Imanpour et al., 2019; Shin et al., 2017) and double reduced beam section connections (Morshedi et al., 2017) in prefabricated steel structures.

Proposed steel beam-to-column connections for precast concrete frames

To develop new SBCs and ensure their favorable performance under seismic load, these connections were compared with an equivalent in-situ concrete beam-to-column connection (ICBC) in a four story structure. The structure had a story height of 3m and four bays with a length of 6 m at perpendicular directions. It was analyzed based on the seismic requirements of intermediate moment resisting frames that are conventional for ordinary residential buildings in a high seismic zone. For designing purposes, an exterior joint of an interior frame was selected on the third story of the structure. The selected joint was designed in precast and in-situ form according to the regulations of ACI318-14 (2015), PCI design handbook (2010), and AISC 341-10 (2010) with a 2/3 scale.

Figure 1 shows the details and the dimensions of the in-situ (M1) and the precast joints (P2 and P3). Figure 2 exhibits the different parts of precast connections in a 3D view. As it can be seen, the steel tube with the inner stiffener is used in the precast joints. The steel tube is placed in the formwork during manufacturing the column and after casting, the inner stiffener is buried in the column’s concrete. In this way, the column is entirely cast in the factory and does not require cast-in-place concrete.

Figure 1.

Details and dimensions of joints.

Figure 2.

Different parts of joints P2 and P3.

The proposed inverse T-shape precast beam provides better conditions for placing the precast hollow core slabs on the side bearing surfaces. In both proposed precast joints, the end girder is used. The top and the bottom flanges of the end girder are connected to the longitudinal bars of the beam through welded splices and then the beam is cast in the factory. In joint P3, instead of using top longitudinal bars, the top flange is extended as a cover plate on the total length of the beam. This plate is anchored to the concrete through the headed studs at certain intervals. For bending resistance, the top and the seat plates, and for shear resistance, the web angles are used in the proposed SBCs.

Applying the proposed SBCs in the precast structures entails the elimination of cast-in-place concrete and high-speed as well as low-cost construction procedure in the workplace. In addition, making the elements in the prefabricated form in the absence of protruding rebars prevents reinforcement congestion in the beam-column joints and results in easier shipping and assembling in comparison to the semi-monolithic connections. Application of the seat plate leads to an easier installation of the precast beams without any requirement for tight tolerances. Inverse T-shaped geometry of the precast beams provides an appropriate space for the connection of precast slabs across the beam height and increases the ceiling height of the story and reduces the total thickness of the floor.

Experimental program

Material properties

Concrete, reinforcing bars, and steel parts of the experimental joints (specimens) were tested under compression and tensile tests, respectively. The characteristics of the materials are summarised in Table 1.

Table 1.

Material properties.

Material properties	For all specimens
Steel parts properties	Longitudinal bars	Transverse bars	Steel plates
Yield strength f_y (MPa)	505	351	301
Ultimate strength f_u (MPa)	623	524	423
Strain in fracture stress ɛ_f (%)	15.58	18.15	24.90
Elastic modulus E_s (MPa)	2.03 × 10⁵	1.71 × 10⁵	1.73 × 10⁵
Strain hardening E_sh (MPa)	1032	1225	656
Concrete property	For M1	For P2	For P3
Compressive strength of cube f_cu^’ (MPa)	49.55	47.70	48.65

The maximum aggregate size used in the concrete was 20 mm. Also, grade 60 longitudinal bars and grade 50 transverse bars (ASTM A615, 2012) and steel parts of type S235JRG2 (Bringas 2016) were used in specimens. The specifications of the beam cross-section are depicted in Table 2 based on the material parameters.

Table 2.

Beam cross-section specifications.

Beam cross-section specifications	Bottom	Top	Bottom	Top	Bottom	Top
Total longitudinal bars area A_s (mm²)	339	942	314	982	314	715^a
Nominal flexural strength^b M_n (kN.m)	53.26	134.17	47.30	137.83	48.91	117.53
Ultimate flexural strength^c M_u (kN.m)	64.15	162.99	57.27	167.14	53.38	162.08
End girder flexural strength^d M_p (kN.m)	—	—	189.03		189.03
Connection flexural strength^e M_c (kN.m)	—	—	260.86		260.86

^aEquivalent bar area based on cross-section of cover plate: (150 × 8) × 301/505.

^bM_n was calculated based on Whitney rectangular stress distribution theory at concrete beam cross-section.

^cM_u was calculated based on ultimate strength of rebars and cover plate at concrete beam cross-section.

^dM_p was calculated based on plastic section modulus (z) of end girder, M_p = zf_y.

^eM_c is related to top and bottom plates flexural capacity and was calculated based on C_prR_yM_P, C_pr = 1.2, R_y = 1.15.

Preparation process

The specimens were tested in the form of an inverse T-shape in a vertical plane and were cast and assembled in a horizontal plane for convenience. Figure 3(a)–(c) shows specimen M1 during and after casting. In this specimen, in contrast to the precast specimens, the beam and column were cast simultaneously. Figure 3(d)–(k) shows the preparation process of the precast specimens. In specimens P2 and P3, before casting the column, the steel tube with its inner stiffener and the reinforcing bars were placed in the column formwork and before casting of the beam, the welded reinforcing bars with the end girder and then the fillers were placed into the beam formwork. Beam P3 was cast inversely in comparison to beam P2 because of using the cover plate. After casting, the fillers were pulled out, and then the end girder was connected to the steel tube of the column by using the seat and the top plates, and the web angles. For assembling, at first, the seat plates were welded to the steel tube and after placement of the precast beam, the web angles and the top plate were welded to the end girder as well as the steel tube. The welds of the top and the seat plates to the steel tube and the flanges were groove and fillet welds, respectively and the welds of the web angles to the steel tube and the end girder were fillet welds. After connecting, the free spaces near the column face were filled with a non-shrink grout. The total height of the end girder was equal to the concrete beam’s height and the width of its top and bottom flanges had the same width as the upper and the lower parts of the concrete beam. Finally, the surfaces of the specimens meshed with 10 cm dimensions.

Figure 3.

Preparation of specimen M1 (a–c), P2 (d–g), P3 (h–k) (a) Beam and column in casting (b) Beam and column after casting (c) Specimen with surface mesh (d) Beam reinforcing skeleton with end girder (e) Column reinforcing skeleton with peripheral steel tube (f) Connecting with seat and top plates and web angles (g) Specimen with surface mesh (h) Beam reinforcing skeleton with cover plate (i) Column reinforcing skeleton with peripheral steel tube (j) Connecting with seat and top plates and web angles (k) Specimen with surface mesh.

Loading pattern

The test setup is shown in Figure 4. Two hinged supports were placed at both ends of the column. At these points, the bending moment in the column is zero. This situation is similar to the column condition affected by lateral seismic loads.

Figure 4.

Test setup.

Two hydraulic jacks were simultaneously used for the tests. The first jack was used to apply an axial force to the column. This applied load was force-control with the value of $0.1 f_{c}^{'} A_{g}$ . In this equation, $f_{c}^{'}$ is compressive concrete strength and $A_{g}$ is the net cross-section of the column. This axial force was considered without eccentricity with respect to the column axis; hence, it did not cause any bending effort. On the opposite side of the column, the hinged support was placed as an axial load-bearing.

The second jack was used to apply cyclic forces at the top of the beam and perpendicular to it. The applied force was displacement-control and its load pattern is depicted in Figure 4. This jack was fastened to the top of the beam using the threaded dowels and the stiffened plate. To increase accuracy and avoid the effect of jack weight on the test results, the jack was connected to the strong frame with a chain.

The selected hysteresis load pattern was similar to the recommended loading pattern by ACI374.1-05 (2005). According to this code, the initial drift ratio should be within an elastic range and subsequent drift ratios to the previous one are suggested to be about 1.25–1.5. Moreover, the test should continue at least up to a 3.5% drift ratio. As shown in Figure 4, the selected experimental loading protocol was relatively severe compared to the ACI374.1-05 (2005) criteria, because the maximum mentioned ratios were chosen to be 2 and the test was continued up to the minimum 5% drift ratio to observe the failure mechanism. Evidently, each drift ratio included three load cycles. In the absence of a 3.5% drift ratio in the load pattern, the performance of the test specimens was evaluated conservatively, not less than the 4% drift ratio.

Test results

Hysteresis curve parameters

The hysteresis curves of the in-situ and the precast specimens are shown in Figure 5. The drift ratio in the graphs was obtained from the following equation

θ = Δ / (L_{b} + 0.5 h_{c})

(1)

Where,

Δ

is beam end displacement,

L_{b}

is clear vertical distance from the column face to the jack centerline and

h_{c}

is column depth.

Figure 5.

Hysteresis curve of specimens and their parameters (a) Specimen M1 (b) Specimen P2 (c) Specimen P3 (d) Deformation parameters definition based on equal energy principle.

The parameters of the hysteresis curves can be seen in Table 3 in three parts: the load-carrying capacity, the deformation capacity, and the stiffness. As it is evident, the maximum resistance (

P_{\max}

) of specimens P2 and P3 are 1.37 and 1.25 times more than specimen M1, respectively, in the negative moment. These ratios are about 1.41 and 1.53, respectively, in the positive moment. The maximum capacity was obtained in M1, P2, and P3 under the negative moment in 4%, 4%, and 2% drift ratios, and under the positive moment, was in 5%, 4%, and 4% drift ratios, respectively. The nominal resistance (

P_{n}

) shown in Table 3 is based on the flexural strength divided by the vertical distance from the plastic hinge region to the jack centerline (185 cm for specimen M1 and 135 cm for specimens P2 and P3). The results imply a ratio of 0.91–1.24 times the maximum resistance to the nominal resistance. The ratios less and more than 1 are related to the negative and positive moments, respectively. The collapse of the concrete in the lower part of the beam in the plastic hinge region caused a shorter bending lever arm in the beam cross-section. It led to fewer resistance ratios under the negative moment. The maximum joint shear values (

V_{j \max}

) in Table 3 were obtained from the following equation

V_{j \max} = P_{\max} \times (L_{b} / j d - (L_{b} + 0.5 h_{c}) / L_{c})

(2)

Table 3.

Parameters of hysteresis curves.

Connection	Load-carrying capacity						Deformation capacity			Stiffness
Connection	Load direction^a	Maximum resistance P_max (kN)	Nominal resistance P_n (kN)	P _max /P _n	Maximum joint shear V_jmax (kN)	V _jmax /V _n	Yielding displacementδ_y (mm)	Maximum displacementδ_u (mm)	Ductility δ_u/δ_y	k_y (kN/mm)	k_p (kN/mm)	k_p/k_y (%)
M1	−	72.10	72.52	0.99	398.39	0.74	36.40 (1.82)	108.60 (5.43)	2.98	1.98	−0.19	10
M1	+	29.43	28.79	1.02	153.58	0.29	23.60 (1.18)	108.60 (5.43)	4.60	1.25	−0.86	69
P2	−	99.08	102.09	0.97	399.69	0.93	36.00 (1.80)	98.60 (4.93)	2.74	2.75	−0.15	5
P2	+	41.59	35.03	1.19	167.77	0.32	28.40 (1.42)	98.60 (4.93)	3.47	1.46	−0.56	38
P3	−	90.45	99.38	0.91	364.87	0.84	25.80 (1.29)	65.20 (3.26)	2.52	3.50	−0.26	7
P3	+	45.03	36.23	1.24	167.77	0.32	24.60 (1.23)	65.20 (3.26)	2.65	1.83	—	—

^aRelated to negative moment; + Related to positive moment.

In which, $j d$ is moment arm length between the tension and compression force centerlines in beam cross-section, $L_{c}$ is length of the column between the two hinge supports. Also, the nominal joint shear capacity ( $V_{n}$ ) was obtained from the following equation

V_{n} = γ \sqrt{f_{c}^{'}} b_{j} h_{c}

(3)

Where,

γ

is 1 for an exterior joint without slabs, and

b_{j}

is effective joint width, which was obtained from the minimum value of the following relations

b_{j} = {\begin{matrix} (b_{b} + b_{c}) / 2 \\ b_{b} + m h_{c} / 2 \\ b_{c} \end{matrix}

(4)

In these relations, $b_{c}$ is column width and in the absence of an eccentricity between the beam and the column centerline, coefficient $m$ is equal to 0.5. The results show that in all joints, the maximum joint shear ratio to the nominal joint shear capacity is less than 1.

Parameters relating to the deformation capacity of the joints include the yield displacement ( $δ_{y}$ ) and the maximum displacement ( $δ_{u}$ ) which were obtained from the envelope curve of the test specimens are shown in Figure 5(d). To obtain the deformation parameters, the idealized curve was used. In the post-peak region, according to ACI374.1-05 (2005), a displacement corresponding to the 75% maximum resistance was considered as the maximum displacement and a displacement corresponding to the yield resistance was obtained from the equal energy principle in the pre-peak region by idealizing the actual curve with the bilinear curve. Also, for the idealization of the actual curve in the post-peak region, the linear curve was used with the same enclosed area. The ratio of the maximum displacement to the yield displacement is known as the connection ductility. The slope of the first line in the pre-peak region in the idealized curve was considered the yield stiffness ( $k_{y}$ ) and the slope of the line in the post-peak region was considered as the post-yield stiffness ( $k_{p}$ ). The $k_{p} / k_{y}$ ratio is shown in Table 3. In specimen P3, $k_{p}$ value has not been calculated because there was no strength degradation until the maximum displacement under the positive moment.

The energy dissipation and the cumulative energy dissipation for the first cycle of per drift ratio are shown in Figure 6. As shown in Figure 6(a), the energy dissipation in specimens P2 and P3 is more than specimen M1 until the 4% drift ratio. As exhibited in Figure 6(b), the total energy dissipation in specimens P2 and P3 in the last drift ratio (5% drift ratio) is 1.36 and 1.12 times specimen M1 at the 5% drift ratio. This ratio was 1.16 and 0.96 in the last drift of specimen P2 and P3 to the last drift in specimen M1 (6% drift ratio), respectively. As a result, specimen P2 has higher energy dissipation than specimen M1. The extensive amount of cracking at the location of plastic hinge formation in specimen P2 compared to specimen M1 is in accordance with the results of this section and could be due to the more distance between the plastic hinge zone and the beam-column joint as well as its proximity to cyclic load.

Figure 6.

Secant stiffness degradation and Energy dissipation (a) Energy dissipation per drift ratio (b) Cumulative energy dissipation per drift ratio (c) Secant stiffness per drift ratio (d) Secant stiffness definition.

The value of the secant stiffness ( $k_{\sec}$ ) in term of the drift ratio is depicted in Figure 6(c). In this figure, the secant stiffness was obtained from the slope of the line passing through the peak points of per drift ratio at the first cycle under the positive and the negative moments (Figure 6(d)). As is obvious, the most stiffness degradation has occurred at the initial drift ratios of the tests. The secant stiffness at the 6% and 5% drift ratios to the initial stiffness in specimen M1 is 6% and 11%, respectively, and at the 5% drift ratio in specimens P2 and P3 is 10% and 4%, respectively. The results indicate more secant stiffness in specimens P2 and P3 with respect to specimen M1 until the 5% and 3% drift ratios. Making use of the peripheral tube at the beam-column joint and end girder increased the stiffness in the precast specimens.

To check the strength degradation in each cycle, the strength ratios of the specimens are shown in Figure 7. The strength ratios in this figure were obtained by dividing the strength value in the second and the third cycles of each drift ratio to that of the first cycle. As shown in Figure 7, in specimens M1, P2, and P3, the minimum strength ratio under the negative moment is 0.71, 0.81, and 0.18 at the 5%, 5%, and 4% drift ratios, and under positive moments is 0.27, 0.58, and 0.91 at the 6%, 5%, and 5% drift ratios, respectively, corresponding to the third cycles. So, the results indicate less strength degradation for specimen P2 under the negative moment and for specimen P3 under the positive moment at their last drift ratios. Deformation concentration at the region of plastic hinge in the precast specimens resulted in less damage in the precast beam and column, as well as the joint during each drift in comparison to the in-situ specimen and rendered less strength degradation in the second and third cycles.

Figure 7.

Strength degradation (a) Specimen M1 (b) Specimen P2 (c) Specimen P3.

Connections qualification

In order to evaluate the seismic performance of the connections, the different parameters of the hysteresis curves of the specimens were compared with the ACI374.1-05 (2005) acceptance criteria. These parameters include strength degradation, energy dissipation, and stiffness degradation at a drift ratio not less than 3.5%. In this study, the ACI374.1-05 (2005) performance criteria were evaluated in the 4% drift ratio, conservatively. According to ACI374.1-05 (2005), the strength degradation ratio of the third cycle to the first cycle should not be less than 25% (in the load direction). Also, the energy dissipation ratio should not be less than 0.125. As shown in Figure 8, this ratio was obtained from the division of the enclosed area of the hysteresis curve in the third cycle to the enclosed area of the dotted line. Also, the secant stiffness in the third cycle should not be less than 0.05 of the initial stiffness in the first cycle of the elastic drift. As shown in the figure, the secant stiffness was obtained from the slope of the passing line from the −0.10 to +0.10 maximum drift ratio of the third cycle.

Figure 8.

ACI 374.1-05 seismic performance criteria.

The comparison of the seismic performance of the specimens with the ACI374.1-05 (2005) acceptance criteria is presented in Table 4. As can be seen, specimen P2, like specimen M1, is able to meet all of the criteria in this code about load carrying capacity (

P_{4} / P_{4max} \geq 0 .75

), energy dissipation (

E_{I} / E_{II} \geq 0 .125

), and stiffness (

k_{4} / k_{i} \geq 0 .05

). But specimen P3 does not meet the criteria in terms of the load-carrying capacity and stiffness. The presence of the cover plate limited the plastic hinge deformations and increased the over-strength within this area and caused a premature failure in specimen P3.

Table 4.

Comparison of the test results with ACI 374.1-05 (2005) acceptance criteria.

Connection	Load-carrying capacity^b				Energy dissipation^c			Stiffness^d
Connection	Load direction^a	Maximum resistance of 4% drift ratioP_4max (kN)	Resistance at third cycle of 4% drift ratio P₄ (kN)	Resistance ratioP₄/P_4max	Elastic-plastic energy dissipation E_II (kN.m)	Actual energy dissipation E_I (kN.m)	E _I /E _II	Initial stiffness k_i (kN/mm)	Secant stiffness at third cycle of 4% drift ratio k₄ (kN/mm)	k ₄ /k _i
M1	−	72.10	67.79	0.94	11.36	3.87	0.34	4.28	0.20	0.05
M1	+	28.25	27.57	0.98	11.36	3.87	0.34	4.49	0.23	0.05
P2	−	98.39	79.85	0.81	15.10	4.40	0.29	5.83	0.29	0.05
P2	+	41.59	36.49	0.88	15.10	4.40	0.29	5.43	0.26	0.05
P3	−	10.30	4.51	0.44	5.06	0.99	0.20	5.90	0.05	0.01
P3	+	45.03	42.67	0.95	5.06	0.99	0.20	5.57	0.06	0.01

^aRelated to negative moment; + Related to positive moment.

^bMaximum allowable degradation in load-carrying capacity at 4% drift ratio is 25% according to ACI 374.1-05 (2005).

^cMinimum allowable energy dissipation ratio at 4% drift ratio is 12.5% according to ACI 374.1-05 (2005).

^dMinimum allowable secant stiffness at 4% drift ratio to initial stiffness is 5% according to ACI 374.1-05 (2005).

Cracking pattern and failure mode

Crack distribution and crushing in the reinforced concrete of specimens as well as the failure of the SBCs different steel parts to last drift ratio were indicated in Figure 9.

Figure 9.

Cracking pattern and failure mechanism of specimen M1 (a–d), P2 (e–h), P3 (i–l) (a) 4% drift ratio (b) 5% drift ratio (cycle1) (c) 5% drift ratio (cycle3) (d) 6% drift ratio (e) 2% drift ratio (f) 4% drift ratio (cycle1) (g) 4% drift ratio (cycle2) (h) 5% drift ratio (cycle2) (i) 2% drift ratio (j) 3% drift ratio (cycle1) (k) 4% drift ratio (cycle1) (l) 5% drift ratio.

The crack initiation of all specimens started from the bottom cover concrete of the beam. It was observed in the vicinity of the girder end at the 0.25% drift ratio for the precast specimens and in the beam end in the vicinity of the column face at the 0.5% drift ratio for in-situ specimen, respectively. Due to the proximity of the cracking zone and applied load location for precast specimens with respect to in-situ specimen, more deformation demand was observed and cracking occurred at the lower drift ratio. Despite the in-situ specimen in which the cracks penetrated into the beam-column joint at the 1.5% drift ratio, it was not observed any cracking at the column face in the precast specimens until the last drift.

The first spalling in specimen M1 occurred at the 4% drift ratio and located on the bottom concrete cover of the beam in the vicinity of the column face. It was observed in the vicinity of the girder end at the 2% and 3% drift ratios for specimens P2 and P3, respectively.

The concrete crushing of the specimens M1 and P2 was observed at the spalling points at the 5% and 3% drift ratios, respectively, whereas specimen P3 experienced cracking in the weld metal of the top plate to the column steel tube under the negative moment after the spalling of its concrete and fractured at the 3.7% drift ratio. The cover plate of the beam in the plastic hinge zone prevented the spread of the concrete cracks and limited the deformations in this area. Therefore, high over-strength was observed in specimen P3 under the negative moment and the failure mechanism was transferred to the beam-column joint after the formation of the plastic hinge.

As was expected, there was not any concrete crushing within the beam-column joints as considered in primary design. The first fracture of the longitudinal bars took place at the specimens M1 and P2 similarly at the 5% drift ratio in the bottom part of the beam where the concrete collapsed. However, the girder and its steel connection plates and the column steel tube of the specimen P2 experienced no damage.

Strain measurement

To evaluate the stress level in the different parts of the in-situ and the precast specimens, the strain was measured at their specified points. These points are shown in Figure 10. According to the figure, the strains were measured on the longitudinal reinforcing bars for the in-situ specimen and in addition to them were taken from the steel plates in the precast specimens.

Figure 10.

Strain gage layout (a) Specimen M1 (b) Specimen P2 (c) Specimen P3.

The selected locations for the strain gauges included predicted places for the formation of the plastic hinge based on the design procedure of the connections, the beam-column joints, the beams end adjacent to column faces, and the connection steel parts.

The measured strains in specimens M1, P2, and P3 are shown in Figures 11–13, respectively. The negative and positive values on the horizontal axis are related to the tension and compression strains and on the vertical axis denote the pull and push directions, respectively.

Figure 11.

Strain measured on specimen M1.

Figure 12.

Strain measured on specimen P2.

Figure 13.

Strain measured on specimen P3.

Results indicated that the first yielding of all specimens occurred in the bottom longitudinal bars at the 0.75% drift ratio and was placed at the end of the beam in the vicinity of the column face for the in-situ specimen and in the vicinity of the girder flange end for precast specimens. As the drift ratio got higher, the top longitudinal bars of the same places started yielding. Hence, regarding the strain measurement, the initial formation of the plastic hinge had conformity with the initial cracking observed in part 4.3.

At the drift ratio of 1.5%, the strain of column longitudinal bars in the vicinity of the top face of the beam exceeded the linear zone in the in-situ specimen. This result indicates plastic hinge has penetrated in beam-column joint at this specimen. On the other side, the plastic hinge expanded in the beam with a length of d (beam effective depth) from the column face. But, in precast specimens, the longitudinal bars of the columns remained elastic until the last drift. As included in the precast specimen’s design, the steel connection members, including the top and bottom flanges of the girder and the top and seat plates as well as the column steel tube remained elastic until the last drift. The strains values of the precast specimens show that the plastic hinge has extended with a length of d from the girder end into the concrete beam. Despite the yielding of the cover plate at the plastic hinge region in specimen P3, the failure occurred at the 3.7% drift ratio in the steel welded plate at the column face. This premature failure occurred due to the unpredicted over-strength observed at the plastic hinge location in the test.

Drift contribution

To measure the deformation of various parts of the specimens, including bending deformation of the beam, bending deformation of the column, shear deformation of the beam-column joint, and plastic hinge rotation, 10 LVDT were used according to Figure 14. Two vertical LVDTs on the hinge supports determined the rigid body rotation of the specimens. Two other vertical Linear variable differential transformer (LVDTs) on the column and adjacent to the beam faces measured the assembled rotation due to the column curvature and the beam-column joint deformation and two diagonal LVDTs measured the shear deformation of the beam-column joint. Also, four LVDTs were located on the beam in the vicinity of the column face and the distance of 300 mm and 600 mm from it in order to calculate the deformation of the beam by the differential curvature of it between these points depending on the location of the plastic hinge. The bending deformation of the beam was calculated based on the differential value of the assembled resultant deformation of each part with the applied deformation of the hydraulic jack.

Figure 14.

Drift contribution measurement. (a) LVDTs layout in tests setup (b) Drift contribution of different parts.

The contribution to the total drift ratio in specimen M1 under the negative and the positive moments from the 1%–6% drift ratio is shown in Figure 15(a). As indicated in the figure, the plastic hinge rotation contribution has increased from zero at the 1% drift ratio to 71% and 62% at the 6% drift ratio under the positive and negative moments, respectively. With more inelastic deformation demands in the plastic hinge region at the higher drift ratios, the bending deformation of the beam and column have been reduced. As can be observed, the shear deformation of the beam-column joint had approximately 10% contribution in all drift ratios. LVDT results of the hinge supports showed the rigid body rotation was negligible and was not considered in the calculations.

Figure 15.

Drift contribution in specimens (a) specimen M1 (b) specimen P2 (c) specimen P3.

The contribution to the total drift ratio in specimens P2 and P3 under the negative and positive moments from the 1%–5% drift ratio is shown in Figure 15(b) and (c). As is obvious, the plastic hinge rotation contribution in specimen P2 under the positive and negative moments at the 1% drift ratio with the value about 5% and 6% up to the 5% drift ratio is increased to 74% and 70%, respectively. This value for specimen P3 is 5% and 4%, respectively, and has reached 73% in the last drift ratio under the positive moment and 53% under the negative moment in the 3% drift ratio, respectively. Top plate to steel tube weld metal began to crack in specimen P3 from the 3% drift ratio under the negative moment and up to the last drift ratio with the fracture of the weld metal, this part has a 94% contribution in total drift ratio. As can be seen, in both precast specimens, the increase in the plastic hinge rotation contribution to the total drift ratio coincides with the decrease in the contribution of the bending deformation of the beam and column. However, shear deformation of the beam-column joint in both specimens had a contribution up to 2% of the total drift ratio. Due to the concrete confinement of the beam-column joint with the peripheral steel tube in the precast specimens, the contribution of the shear deformation of this part and its damage is much less than the in-situ specimen.

Conclusion

This study presented experimental research on two types of precast connections and an in-situ one under cyclic loading. The parameters obtained from the hysteresis curves of the connections were compared entirely and assessed based on the seismic qualification criteria of ACI374.1-05 (2005). Furthermore, the strains of the various parts of the connections were measured to determine the plastic hinge location. Finally, the drift contribution of the specimen elements was compared together. The results of this study can be summarized as follows:

1. SBC P2 can be used as a moment resistant precast beam-to-column connection according to the ACI374.1-05 (2005) acceptance criteria.

2. SBC P2 revealed 1.37 and 1.41 times ICBC M1 resistance under the negative and positive moments, respectively. This higher resistance was developed due to the proximity of the plastic hinge location to cyclic loads.

Additionally, the amount of energy dissipation and ductility of these connections were approximately identical.

3. SBC P3 failed to meet ACI374.1-05 (2005) acceptance criteria in terms of strength and stiffness degradation. The premature failure of this specimen occurred due to the unpredicted over-strength observed at the plastic hinge location in the test.

4. However, this connection with a maximum resistance of 1.25 and 1.54 times ICBC M1, under negative and positive moments, could be used in zones with low seismic hazards.

5. Plastic hinge formation in specimen P2, in contrast to specimen M1, occurred within the proposed precast beam instead of the beam-column joint, which coincides with the strong-column and weak-beam principle in precast frames and postpones the total fracture mechanism of structure.

6. Due to using of the peripheral steel tube in the SBCs P2 and P3, the concrete damage and drift contribution of the beam-column joints for these connections was less than the ICBC M1.

7. Precast inverse T shape beams, provide suitable conditions for seating the precast slabs and reduce the total thickness of the precast floor.

8. The precast proposed beams have benefits such as easy production and assembling process compared to semi-monolithic ones and prevent reinforcing bars congestion in beam-column joint.

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) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Neyram Ahooghalandary

Appendix

References

ACI 318 (2015) Building Code Requirements for Structural Concrete, ACI 318-14. Farmington Hills, MI: ACI.

ACI 374 (2005) Acceptance Criteria for Moment Frames Based on Structural Testing and Commentary, ACI 374.1-05. Farmington Hills, MI: ACI.

AISC 341 (2010) Seismic Provisions for Structural Steel Buildings, AISC 341-10. Chicago, IL: AISC.

Aninthaneni

Dhakal

Marshall

, et al. (2018) Nonlinear cyclic behaviour of precast concrete frame sub-assemblies with dry end plate connection. Structures 14: 124–136.

ASTM A615 (2012) Standard Specification for Deformed and Plain Carbon-Steel Bars for Concrete Reinforcement, ASTM A615/A615M. West Conshohocken, PA: ASTM.

Bringas

(2016) Handbook of Comparative World Steel Standards. West Conshohocken, PA: ASTM International.

Dadmand

Pourbaba

Sadaghian

, et al. (2020b) Experimental & numerical investigation of mechanical properties in steel fiber-reinforced UHPC. Computers and Concrete 26(5): 451–465.

Dadmand

Pourbaba

Sadaghian

, et al. (2020a) Effectiveness of steel fibers in ultra-high-performance fiber-reinforced concrete construction. Advances in Concrete Construction 10(3): 195–209.

Deng

Song

, et al. (2020) Seismic performance of interior precast concrete beam-column connections with highly ductile fiber-reinforced concrete in the critical cast-in-place regions. Engineering Structures 210: 110360.

10.

Dong

Pan

, et al. (2018) Mechanical behavior of a novel precast beam-to-column connection with U-shaped bars and engineered cementitious composites. Advances in Structural Engineering 21(13): 1963–1976.

11.

Eghbali

Mirghaderi

(2017) Experimental investigation of steel beam to RC column connection via a through-plate. Journal of Constructional Steel Research 133: 125–140.

12.

Elsanadedy

Almusallam

Al-Salloum

, et al. (2017) Investigation of precast RC beam-column assemblies under column-loss scenario. Construction Building Material 142: 552–571.

13.

Esmaeili

Ahooghalandary

(2019) Introducing a precast moment resistant beam-to-column concrete connection comparable with in-situ one. Computers and Concrete 23(3): 203–215.

14.

Gou

Ding

Fan

, et al. (2018) Seismic performance of a novel precast concrete beam-column connection using low-shrinkage engineered cementitious composites. Construction Building Material 192: 643–656.

15.

Guan

Yang

, et al. (2019) Cyclic loading test on a locally post-tensioned precast concrete beam–column connection. Advances in Structural Engineering 22(12): 2699–2711.

16.

Imanpour

Torabian

Mirghaderi

(2019) Seismic design of the double-cell accordion-web reduced beam section connection. Engineering Structures 191: 23–38.

17.

Liu

Yan

(2017) Seismic performance comparison between precast beam joints and cast-in-place beam joints. Advances in Structural Engineering 20(9): 1299–1314.

18.

Morshedi

Dolatshahi

Maleki

(2017) Double reduced beam section connection. Journal of Constructional Steel Research 138: 283–297.

19.

PCI (2010) PCI Design Handbook: Precast and Prestressed Poncrete. Chicago, IL: Prestressed Concrete Institute.

20.

Pourbaba

Joghataie

Mirmiran

(2018) Shear behavior of ultra-high performance concrete. Construction and Building Materials 183: 554–564.

21.

Pourbaba

Sadaghian

Mirmiran

(2019) A comparative study of flexural and shear behavior of ultra-high-performance fiber-reinforced concrete beams. Advances in Structural Engineering 22(7): 1727–1738.

22.

Rodriguez

Torres-Matos

(2014) Seismic behavior of a type of welded precast concrete beam-column connection. PCI Journal 58(3): 81–94.

23.

Shufeng

Qingning

Hao

, et al. (2018) Experimental study of a fabricated confined concrete beam-to-column connection with end-plates. Construction Building Material 158: 208–216.

24.

Shin

Kim

Halterman

, et al. (2017) Seismic toughness and failure mechanisms of reduced web-section beams: phase 2 tests. Engineering Structures 141: 607–623.

25.

Song

, et al. (2021) Seismic design of hybrid unbonded post-tensioned precast concrete joints based on bearing capacity and energy dissipation capacity. Advances in Structural Engineering 24(8): 1724–1738.

26.

Vella

Vollum

Kotecha

(2018) Headed bar connections between precast concrete elements: design recommendations and practical applications. Structures 15: 162–173.

27.

Yan

Chen

Xie

(2018) Seismic experimental study on a precast concrete beam-column connection with grout sleeves. Engineering Structures 155: 330–344.

28.

Zhang

Zheng

Zhao

(2019) Seismic performance of steel beam-to-column moment connections with different structural forms. Journal of Construction Steel Research 158: 130–142.