Cyclic behaviour of bearing-type bolted connections with slot bolt holes

Abstract

This article experimentally studies the behaviour of high-strength bolted connections with slot bolt holes under cyclic load to evaluate its seismic performance. A total of six specimens varying in the bolt diameters and pretension are designed and tested. The results show the connections with slot holes have good energy dissipation capacity. It is found that connections with M30 bolts, of which the hysteretic loops are fuller, have better energy dissipation capacity than that of M20 bolts connections. With the increase of number of loading cycles, the sliding force of the connections decreases. On the other hand, the ultimate bearing capacity of the connections does not decrease much both for M30 and M20 bolts. As the experiments proceed, the friction coefficients and the areas of the hysteresis curves decrease in a similar trend because of the smoothing of surfaces. The experimental results indicate the potential application of the connections with slot bolt holes as dual-function parts in structures to simultaneously provide stiffness and energy dissipation capacity.

Keywords

cyclic behaviour energy dissipation capacity high-strength bolt sliding force slot hole

Introduction

Steel structures have many advantages such as lightweight, high strength, appealing architecture and recyclable use of materials. These advantages make them particularly suitable for application in high-rise buildings. Historically, most major structural damage in earthquakes is due to connection failure (Shi et al., 2013). Many brittle failures of connections have been observed in Northridge earthquake in 1994 due to their limited plastic deformation capacity (Plumier, 1994). During Kobe earthquake in 1995, a total of 988 steel constructions were damaged, most of which were due to the brittle failure of beam–column joints with limited ductility. In addition, connections account for more than half the cost of structural steelwork, and their design and detailing are therefore of primary importance for the safety and economy of structures.

Due to the huge amount of brittle failures of rigid connections, efforts have been made to enhance the ductility of moment-resisting steel frames. Chen and Lui (1988) recommended that rigid connections should have certain rotation capacity, in addition to the strength, to facilitate the moment redistribution. Engelhardt and Sabol (1997) pointed out that it was not possible to generate sufficient ductility of welded connections even if the welding quality was ensured by removing beam–column welding lining board. The enhancement for rigid connections by improving conventional welding was limited.

As an alternative, bolting is a preferred method of connecting members on site (Fiorino et al., 2014; Jensen and Quenneville, 2011). Based on the force transfer mechanism by bolts, the bolted connections can be classified as bearing type (relying on the bolts to transfer force) or friction type (relying on the friction between the plates clamped by bolts to transfer force). Bearing-type connections allow the slip of connected plates and thus the bearing capacity is determined by either shear strength of bolts or bearing capacity of the hole walls. High-strength friction grip (HSFG) bolts are commonly used for bolted connections. Chung et al. (2005) and Da Silva et al. (2004) did research on the different loading patterns of bolted connections. Egan et al. (2014) built three-dimensional (3D) explicit finite element models to predict the quasi-static bearing response of typical countersunk composite fuselage skin joints, where the initial joint sticking behaviour and the elastic loading response of single-bolt and three-bolt joints were accounted for. Choi et al. (1996) established a 3D finite element methodology to simulate bolted connections. The effect of bolt pretension, shape of bolt shank head and nut were taken into consideration in the model, which can properly simulate the actual behaviour of the bolt connections, especially the end-plate connection. Yang et al. (2011) studied the failure of bolted connections at elevated temperatures. Astaneh-Asl (1999) proposed to allow high-strength bolting joints to slip under strong earthquake. The friction between the interfaces and the extrusion of hole walls consumed earthquake energy, avoiding the brittle failure or collapse of structures. Swanson and Leon (2000) also pointed out that the energy dissipated by the bolts slipping took a large part of that dissipated by the whole connection. The design methods of high-strength bolt splice joints were presented in the references (Kulak et al., 2000; Kulak and Green, 1990; Sheikh-Ibrahim and Frank, 1998). Popov (1988) carried out cyclic loading experiments on the bolted beam–column joints. Ju and Oh (2016) and Yang et al. (2016) also did fatigue test research on bolt-connected joints. However, they were mainly concerned with the hysteretic behaviour of the structural joints, and the mechanical properties of the bolted connections were not studied. Calado et al. (1998) found that standard circular bolt holes would become elliptical holes under the cyclic loading, while their energy consumption capacity decreased gradually. The influence of different arrangement of bolts on their cyclic behaviour was studied by Gerami et al. (2011). They found bolt arrangement variation had insignificant effect on the moment-bearing capacity of connections. Moreover, the seismic performance of beam-to-column connections were studied. However, bolts with standard holes were used in their research (Brunesi et al., 2014, 2015; Daidie et al., 2007; Ksentini et al., 2015).

To solve the problem for connections with standard holes such as high demand of installation accuracy, limited energy dissipation capacity and decreased load-bearing capacity when deforming to elliptical shape, slot bolt holes have been applied. Compared to connections with standard holes, slot-hole connections are more convenient and able to reduce rework and delays caused by machining precision deficiency (Allan and Fisher, 1968). The length of the short axis of slot holes equals to the diameter of the corresponding standard hole, while the long axis is 1.7 times as long as the short axis. Grigorian et al. (1992) designed the elliptical slot bolt holes. The slot was parallel to the force direction of members. When the tensile force in a connection was larger than the static friction force, the connection would slide consuming energy. This kind of connection could also be used for beam-column connections in steel frames (Allan and Fisher, 1968; Butterworth and Clifton, 2000; Terblanche, 2015) to improve the ductility and avoid brittle failure at the joints. Peng et al. (2007) explored the relationship between the hole area of the high-strength bolted connections and the pretension loss and the friction coefficient, respectively, and gave the suggested value of the hole shape coefficients to reflect the influence of both hole shape and hole area. ANSI/AISC360-10 and European standard (EC3) also specify the hole shape coefficient for different hole shapes. They provide corresponding provisions in the case of different spacing and margin of bolt holes.

High-strength bolted connection with slot holes can be used in beam–column joints and other forms of structural connections. It is supposed to keep stable under normal circumstances while playing the role of friction energy dissipation in strong earthquake. Previous research on slot-hole connections concentrated on the static performance. This article experimentally explored the mechanical properties of high-strength bolted connections with slot bolt holes under cyclic loads. A total of six specimens varying in the bolt diameters and pretensions were designed and tested. The sliding forces and energy dissipation capacity of the connections were studied.

Test layout

Design of specimens

Traditional bolted connections of steel beams and columns are shown in Figure 1. The bottom flange of the beam is bolted connected to the corbel which is welded to the flange of the column. Stiffening ribs are added to avoid the local buckling of the corbel and beam web. Based on this kind of connection, a total of six specimens with sandblasted surfaces were designed and tested, as listed in Table 1 and Figure 2. The specimens were divided into two groups (S1 and S2) by varying the diameter of bolts and pretension forces. The diameters of M20 and M30 bolts are 20 mm and 30 mm, respectively. The grade 10.9 high-strength bolts were utilized with tensile strength of 1000 MPa and yield ratio of 0.9. Slot bolt holes were designed, and their dimensions are shown in Figure 3. The dimension of specimens is shown in Figure 4. In order to prevent the local buckling of specimens under compression, stiffening ribs were added.

Figure 1.

Schematic of steel beam-to-column connections.

Figure. 2.

Dimension of the specimens: (a) Type 1 and (b) Type 2.

Figure 3.

Dimension of the slot bolt holes on the corbel plate: (a) Hole of M20 and (b) Hole of M30.

Figure 4.

A specimen in the test.

Table 1.

List of the tested specimens.

Specimen	Type of specimen	Bolt class	Type of holes	Pretension (kN)
S1-1	Type 1	M20	Slot	139
S1-2	Type 1	M20	Slot	147
S1-3	Type 1	M20	Slot	140
S2-1	Type 2	M30	Slot	341
S2-2	Type 2	M30	Slot	344
S2-3	Type 2	M30	Slot	358

Pretension

There are two means to impose tensile forces in the bolt: torque method and angle method. The former is to apply torsion on the nut, and the latter is to apply a relative angle between the nut and shank of the bolt. The torque method was used in the tests, as it was simple and convenient to use. The pretension process was divided into initial and final twist based on Chinese codes (JGJ82-2011 and GB50205-2001). A torque of half the ultimate torque was applied in the initial twist. The ultimate torque of high-strength bolts can be calculated by equation (1)

T_{c} = k \times P_{c} \times d

(1)

where T_c is the ultimate torque, Nm; k is the average value of the torque coefficient of the bolt connecting pair which is determined by experiments; P_c is the pretension in the high-strength bolt, kN, which is determined by JGJ82-2011; and d is the diameter of the bolt screw, mm.

The design pretensions of M20 and M30 bolts were 155 and 355 kN according to JGJ82-2011, respectively. Considering the complexity in the load-applying process, the 5% deviation is acceptable. As a result, it is required that the applied pretension of bolts in practice should fall in the range of 0.95–1.05P_c. During the twist, the bolt pretension was measured by the strain gauges, as shown in Figure 5. The strain gauges were arranged in the two grooves on symmetrical positions of the shank below the nut. The data wires of the strain gauges passed through the holes predrilled in the nut. The actual pretension of each specimen is also listed in Table 1.

Figure 5.

Arrangement of strain gauges on the bolt.

Measurement

The deformation was measured by symmetrically arranged linear variable differential transformers (LVDTs) shown in Figure 6. Considering that the total load stroke does not exceed 30 mm, we choose YHD-30 LVDTs with a maximum stroke of 30 mm, instrumental error less than 0.017 mm. The variation of pretensions of the bolts was monitored by strain gauges. The strain gauge coefficient is equal to 2.06.

Figure 6.

Layout of displacement measurements: (a) plan view and (b) elevation view.

Loading

Loads were imposed based on Chinese codes GB50205-2001 and JGJ 101-2015, as shown in Figure 7. At the beginning of the tests, force control was adopted until the load increased up to 10% of the slip load, after which it was applied smoothly in a speed of 3–5 kN/s until the occurrence of sliding of the connection. The displacement control was then adopted in an interval of 1 mm until the contact of the bolt and hole walls. For each displacement level, three cyclic loadings were conducted. Once the bolt shank contacted the hole walls, a cyclic loading of 60 circles was conducted to study the low cycle fatigue performance and energy dissipation capacity of the bolted connections with slot bolt holes.

Figure 7.

Loading of the specimens.

Test phenomenon

In the initial stage of loading, the specimens were in elastic stage. The relative displacement of the connections increased with the increase of the load, and the connections did not make a sound. When the load reached the sliding load, along with a loud noise, the members experienced the first slide. Further loading on the connections led to gradual siding accompanied by successive noises. For the later cyclic loading stage, the connections issued a continuous ringing sound, which gradually declined until the loading was completed.

After loading, the iron filings slipped from the specimen due to friction and the specimens tended to be smooth, as shown in Figure 8.

Figure 8.

Specimens after loading: (a) slot holes after loading and (b) iron filings.

Test results

Load versus slip displacement curves

The relationship between the slip displacement and force of bolts obtained through experiments is shown in Figure 9. It can be seen that:

The force–displacement curve was linear before the occurrence of bolt sliding in each circle. After the sliding, the stiffness of the connections degenerated rapidly, accompanying by a large increment of displacement as the force kept stable. The hysteresis curves were stretched as the cycles of loading increased, and the area of the hysteresis loops increased greatly.

With the increase of number of loading cycles, the sliding force decreased for both M20 and M30 bolts. The friction coefficient which equals sliding force divided by the initial pretension to evaluate the surface roughness also decreased.

The ultimate load-bearing capacity of the connections did not decrease much with the increasing cycles of hysteresis loops. The sliding has insignificant influence on the shear strength of bolts and bearing capacity of the hole walls.

The hysteresis curves were symmetrical, and the middle part of the sliding section was rectangular and full. This indicated the significant energy dissipation capacity of bolted connections with slot bolt holes.

Compared with the hysteresis loops of M20 bolts, those of M30 bolts were more dissipative, indicating that the slot holes of M30 bolts had a better seismic performance.

Figure 9.

Hysteresis loops of specimens: (a) S1-1, (b) S1-2, (c) S1-3, (d) S2-1, (e) S2-2 and (f) S2-3.

To quantitatively explore the variation of hysteresis loops against number of circles, the hysteresis curves for the 1st, 30th and 60th circles were compared, as shown in Figure 10. It can be seen that the sliding load decreased more in the first 30 circles compared with the latter 30 circles. For M20 bolts (Figure 10(a) to (c)), the sliding force decreased from about 400 to about 200 kN for the first 30 cycles and further to about 120 kN after 60 cycles. For M30 bolts (Figure 10(d) to (f)), the sliding force decreased from about 1200 to 600 kN after 30 cycles and to about 400 kN after 60 cycles. The decrease of sliding load should be attributed to the fact that the friction surface tended to be smooth along with the coefficient of friction decrease. After the first 30 cycles of sliding, the smoothness of the friction surface became stable and thus there is no significant reduction in the sliding force for the latter 30 circles. This explanation was consistent with the phenomenon of continuous sound in first 30 circles and less sound in the latter 30 cycles. During the experiments, the screw constantly contacted the hole walls, causing the deformation of hole walls which was more obvious in the first 30 circles.

Figure 10.

Force-displacement curves of specified circles: (a) S1-1, (b) S1-2, (c) S1-3, (d) S2-1, (e) S2-2 and (f) S2-3.

Table 2 lists the sliding force of the specimens for different cycles. The average results of three specimens for S1 and S2 were calculated, respectively. The table showed that, after the first 30 cycles, the sliding force was reduced by more than 50% for both M20 and M30 bolts, while it was reduced by more than 70% after 60 cycles. The reduction of sliding forces for the M20 bolts was greater than that for the M30 bolts. The ultimate loads were not reached, since the loading was stopped before the shear failure of the bolts or the bearing failure of the hole walls, for a safety consideration. The ultimate bearing capacity of the connections with M30 bolts is about 1.5 times the maximum load of actuators based on calculation.

Table 2.

Sliding forces of the specimens for different number of cycles.

Specimen	Bolt	Design value (kN)	Measured value (kN)
Specimen	Bolt	Design value (kN)	Fs30	(Fs1–Fs30)/Fs1	Fs60	(Fs1–Fs60)/Fs1	Fs30
S1	M20	372	535	198	63%	120	77%
S2	M30	852	1197	567	53%	377	69%

Energy dissipation capacity

To clearly reflect the energy dissipation capacity and friction coefficients of the tested connections, the relationship between them and the number of cyclic loading cycles is expressed in Figure 11. The ‘force ratio’ represents the ratio of the external force to the sliding force at the moment when the displacement is zero. Similarly, the ‘area ratio’ represents the ratio of the hysteresis loop area enclosed by each circle to that of the first circle. The force ratios for the pushing and pulling loading process were presented. As shown in Figure 11, the hysteretic curve area and surface friction coefficient decreased with the increase of number of loading cycles in a similar trend. The similar trend indicates that the energy consumption of high-strength bolted connections with slot holes mainly depended on the friction of the surfaces.

Figure 11.

Decrease of force along with loading cycles: (a) S1-1, (b) S1-2, (c) S1-3, (d) S2-1, (e) S2-2 and (f) S2-3.

The friction coefficient and the area of the hysteresis curves decreased by half after the first 30 cycles. After 60 circles, the reduction ratio of S1 (M20 bolts) and S2 (M30 bolts) dropped to 30% and 40% of the initial values, respectively. Therefore, the connection with M30 bolted connections had a better energy dissipation capacity and fatigue performance compared with M20 bolted connections.

Conclusion

This article experimentally investigated the cyclic behaviour of high-strength bolted connections with slot bolt holes. The following conclusions can be drawn:

High-strength bolted connections with slot holes had full hysteretic loops and thus good energy dissipation capacity.

It was found that connections with M30 bolts had better energy dissipation capacity and load-bearing capacity than those with M20 bolts.

The sliding forces of the connections decreased with the increase of the number of loading cycles. The sliding force reduced by more than 50% and 70% after the 30 and 60 cycles, respectively.

The ultimate bearing capacity of the connections did not decrease much with the increase of the number of loading cycles.

The friction coefficient and the area of the hysteresis curve decreased by half after the first 30 cycles. After 60 circles, the reduction ratio of connections with M20 bolts and M30 bolts dropped to 30% and 40% of their initial values, respectively.

The experimental results showed that the bolted connections with slot holes had good energy dissipation capacity and acted as a friction damper by dissipating energy through the friction of plate surfaces. This type of connections can also provide both bearing capacity and stiffness before sliding, which means the connection with slot holes can be utilized as dual-function components in structures to provide both stiffness and energy dissipation. However, the large decrease of friction coefficient along with sliding should be avoided by improving the property of friction surfaces. The use of paint on the contact surface may be an effective method to solve this problem which needs further researches.

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 work presented in this paper was supported by the National Natural Science Foundation of China with grants 51408418 and 51408620.

References

Allan

Fisher

(1968) Bolted joints with oversize or slotted holes. Journal of the Structural Division 94(9): 2061–2080.

Astaneh-Asl

(1999) Seismic performance and design of bolted steel moment-resisting frames. Engineering Journal 36(3): 105–120.

Brunesi

Nascimbene

Rassati

(2014) Response of partially-restrained bolted beam-to-column connections under cyclic loads. Journal of Constructional Steel Research 97: 24–38.

Brunesi

Nascimbene

Rassati

(2015) Seismic response of MRFs with partially-restrained bolted beam-to-column connections through FE analyses. Journal of Constructional Steel Research 107: 37–49.

Butterworth

Clifton

(2000) Performance of hierarchical friction dissipating joints in moment resisting steel frames. In: Proceedings of the 12th world conference on earthquake engineering, 30 January–4 February, Auckland, New Zealand.

Calado

Castiglioni

Bemuzzi

(1998) Behaviour of steel beam-to-column joints under cyclic reversal loading: an experimental study. Stability & Ductility of Steel Structures, 279–292.

Chen

Lui

(1988) Static flange moment connections. Journal of Constructional Steel Research 10: 39–88.

Choi

Chung

Song

(1996) Refined three-dimensional finite element model for end-plate connections. Advances in Steel Structures 122(11): 365–370.

Chung

Wang

(2005) Structural performance of cold-formed steel column bases with bolted moment connections. Steel & Composite Structures 5(4): 289–304.

10.

Da Silva

De Lima

LRO

Vellasco

PCGDS

et al . (2004) Behaviour of flush end-plate beam-to-column joints under bending and axial force. Steel & Composite Structures 4(2): 77–94.

11.

Daidie

Chakhari

Zghal

(2007) Numerical model for bolted t-stubs with two bolt rows. Structural Engineering & Mechanics 26(3): 343–361.

12.

Egan

Mccarthy

Frizzell

et al . (2014) Modelling bearing failure in countersunk composite joints under quasi-static loading using 3D explicit finite element analysis. Composite Structures 108(1): 963–977.

13.

Engelhardt

Sabol

(1997) Seismic-resistant steel moment connections: developments since the 1994 Northridge earthquake. Progress in Structural Engineering and Materials 1(1): 68–77.

14.

Fiorino

Macillo

Mazzolani

(2014) Mechanical behaviour of bolt-channel joining technology for aluminium structures. Construction and Building Materials 73: 76–88.

15.

Gerami

Saberi

et al . (2011) Cyclic behavior of bolted connections with different arrangement of bolts. Journal of Constructional Steel Research 67(4): 690–705.

16.

Grigorian

Yang

Popov

(1992) Slotted bolted connection energy dissipators. Earthquake Spectra 9: 491–504.

17.

Jensen

Quenneville

(2011) Experimental investigations on row shear and splitting in bolted connections. Construction and Building Materials 25(5): 2420–2425.

18.

(2016) Static and fatigue performance of the bolt-connected structural jointed of deep corrugated steel plate member. Advances in Structural Engineering 19(9): 1435–1445.

19.

Ksentini

Combes

Abbes

et al . (2015) Simplified model to study the dynamic behaviour of a bolted joint and its self loosening. Structural Engineering & Mechanics 55(3): 639–654.

20.

Kulak

Green

(1990) Design of connectors in web-flange beam or girder splices. Engineering Journal 27: 41–48.

21.

Kulak

Fisher

Struik

JHA

(2000) Guide to Design Criteria for Bolted and Riveted Joints. Hoboken, NJ: Wiley.

22.

Peng

Hou

Wen

et al . (2007) Experiment on effect of hole size and hole shape on bearing capacity of high strength bolt friction type connection. Steel Construction 8(22): 30–34 (in Chinese).

23.

Plumier

(1994) Behaviour of connections. Journal of Constructional Steel Research 29: 95–119.

24.

Popov

(1988) Seismic moment connections for MRFs. Journal of Constructional Steel Research 10: 163–198.

25.

Sheikh-Ibrahim

Frank

(1998) The ultimate strength of symmetric beam bolted splices. Engineering Journal 35(3): 106–118.

26.

Shi

Wang

et al . (2013) Cyclic behavior of 460 mpa high strength structural steel and welded connection under earthquake loading. Advances in Structural Engineering 16(3): 451–466.

27.

Swanson

Leon

(2000) Bolted steel connections: tests on t-stub components. Journal of Structural Engineering 126(1): 50–56.

28.

Terblanche

JLF

(2015) Modelling of slotted bolted friction connections as seismic energy dissipaters in braced steel frames. Doctoral Dissertation, Stellenbosch University, Stellenbosch, South Africa.

29.

Yang

Hsu

Chen

(2011) Shear strength of high-strength bolts at elevated temperature. Construction and Building Materials 25(8): 3656–3660.

30.

Yang

Lei

Chen

(2016) Constant amplitude fatigue test research on m20 high-strength bolts in grid structure with bolt–sphere joints. Advances in Structural Engineering 20(10): 1466–1475.