- Email: [email protected]
Shear resistance contribution of support systems in double shear test
Tunnelling and Underground Space Technology 56 (2016) 168–175
Contents lists available at ScienceDirect
Tunnelling and Underground Space Technology journal homepage: www.elsevier.com/locate/tust
Shear resistance contribution of support systems in double shear test L. Li ⇑, P.C. Hagan, S. Saydam, B. Hebblewhite School of Mining Engineering, UNSW Australia, Sydney, Australia
a r t i c l e
i n f o
Article history: Received 17 August 2015 Received in revised form 1 December 2015 Accepted 23 March 2016
Keywords: Double shear test (DST) Rockbolt Shotcrete TSLs Shear resistance
a b s t r a c t Rockbolt and surface support systems such as shotcrete and thin spray-on liners (TSLs) are widely used as underground support elements to resist the convergence and maintain the stability of excavations. In order to evaluate the bearing capacity of combined reinforced rockbolt and surface support systems in preventing sliding along discontinuities, double shear tests (DST) was carried out using fully grouted rockbolts installed in three separate blocks. These blocks were covered with a 5 mm layer of TSL followed by a 50 mm layer of shotcrete. Two rockbolts were installed at an inclined angle of 45°, and 20 kN lateral constraining force was applied to clamp together the three blocks. Three different support combinations were tested: 50 mm shotcrete only, 5 mm TSL only, and combined shotcrete and TSL, with and without rockbolts. It was confirmed that the shotcrete plays a mechanical role in resisting the shear load, and TSLs increase the bond strength between shotcrete and substrate replicating the side wall of an excavation. The contribution of rockbolt and surface support system in resisting joint movement was also compared. The failure mechanism of rock substrate, rockbolt and surface support system was also analysed. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Design, installation and monitoring support systems are a major consideration in underground mining operations. The use of rockbolts dates back to the late 1900 s in mining and includes mechanically anchored (slot and wedge or expansion shell) bolts, fully grouted bolts and resin bolts. Generally, the purpose of a rockbolt is to develop inherent strength and stiffness in a rock mass by securing the loose surface blocks to deep stable rock masses (Hoek et al., 2000). Rockbolts develop their tensile and shear strength during the convergence of rock mass, in which the rockbolt can be considered as part of a rock mass. Rockbolts when combined with shotcrete and/or TSLs can cope with almost any ground condition encountered during mining and tunnelling. Thin spray-on liners (TSLs) are multicomponent polymer materials that can be applied to the rock mass surface as a sealant or as a surface support (Darling, 2011). When the shotcrete is sprayed onto the surface of rock, the adhesion strength develops over time. Moreover, any fracture on the surface is filled and sealed by the application of high speed injected shotcrete, helping maintain the integrity of the rock mass. Failure is normally due to adhesion failure between the shotcrete and rock interface that can be initiated after a small displacement.
⇑ Corresponding author. E-mail address: [email protected] (L. Li). http://dx.doi.org/10.1016/j.tust.2016.03.011 0886-7798/Ó 2016 Elsevier Ltd. All rights reserved.
Over the past four decades, research has confirmed that shear performance can be just as significant as tensile performance. This is evident by a high proportion of rockbolts that have been found to fail in shear in high stress rock masses (Li, 2010) and under rockburst conditions (Haile, 1999). A case study conducted by McHugh and Signer (1999) indicated that shear loading contributed significantly to the failure of anchoring systems. Malmgren and Nordlund (2008) pointed out that the behaviour of shotcrete in interaction with rock is very complex and the performance of shotcrete is influenced by a number of parameters, especially the present of discontinuities. However, there is less known about the effect of discontinuities on the failure of shotcrete. Many experimental tests have been performed in order to study the mechanical behaviour of rockbolts in resisting the shear force based on single shear tests (Haas, 1976; Haas, 1981; Yoshinaka et al., 1987; Egger and Zabuski, 1991; Holmberg, 1991; Pellet, 1993; Ferrero, 1995; Pellet and Egger, 1996; McHugh and Signer, 1999; Mahony and Hagan, 2006) and double shear tests in the laboratory (Aziz et al., 2003; Grasselli, 2005; Jalalifar, 2006; Craig and Aziz, 2010; Jalalifar and Aziz, 2010; Hyett and Spearing, 2013). The factors that can influence shear resistance include (Hartman and Hebblewhite, 2003): Rock mass condition: joint aperture/roughness/dilatancy, Strength and size of rock, The reinforcement elements system: type/diameter/mechanical properties of anchoring systems,
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
Grout type/thickness, Inclination of bolt and joint plane, and And loading conditions: pretension, normal pressure, confinement condition. Hass (1976) studied the ‘natural fracture’ of shear surface by splitting limestone blocks with 25 mm surface roughness. The results showed that surface roughness increased the initial shear resistance. However, the shear resistance decreased rapidly as the asperities were sheared off. Hass (1976) further compared the anchorage capacity of fully grouted rockbolts in a rifled hole and a smooth hole. He concluded that roughness plays a major role in increasing the tensile capacity and shear force. Grasselli (2005) pointed out that the variation in bolt inclination affects the maximum load mobilised in the rockbolt as well as the rigidity of a jointed system. Spang and Egger (1990) carried out shear tests with bolt diameters of 8 mm, 10 mm, and 40 mm. The results indicated that the maximum shear force and displacement is proportional to bolt diameter. Pull-out capacity of a rockbolt in different resin encapsulation annulus was examined by Hagan and Weckert (2004). They found that pull-out capacity of fully encapsulated rockbolt is not affected by changes in resin annulus of less than 4 mm. However, there was a reduction in pull-out capacity at greater resin thicknesses. There is little research about the resin annulus effect in shear tests. Hard rocks usually have higher shear resistance compared with soft rock and rockbolts embedded in harder rock usually require smaller displacements to attain a given resistance than those in softer rock (Spang and Egger, 1990). Ferrero (1995) mentioned that pretensioning does not influence the maximum resistance of the reinforced shear joint; however, it will heavily influence the magnitude of the shear displacement. An analysis of double shear testing results shows that the behaviour of a rockbolt under load follows three distinct stages, linear behaviour, non-linear behaviour and unconstrained plastic deformation. The contribution to shear strength is defined as the difference between the applied shear force and the frictional strength normalised to the ultimate tensile load of the bolt. The first and second stage refers to loading up to 75% and 90% of maximum load capacity respectively for a full scale rockbolt (Grasselli, 2005). Polypropylene-fibre is used to increase the ductility of shotcrete, leading to stress redistribution (Malmgren, 2007). The possible mechanism for shotcrete in interaction with the rock is by block interlock theory. Also the effect of sealing the joints by shotcrete helps to maintain the integrity of the rock mass (Malmgren et al., 2005; Stacey, 2001). The shear-bond strength of TSL was assessed by Yilmaz (2007) and the double-sided shear tests for various TSLs were conducted by Saydam et al. (2004) and Richardson et al. (2009). These researchers examined the capacity of shotcrete or TSLs under shear conditions; however, little is known about the performance of shotcrete and TSLs in combination with rockbolts. These studies examined shear resistance capacity associated with the movement of rock mass; however, they are insufficient to provide an integrated approach for combined underground support systems. It is common practice that surface support systems are often used in combination with rockbolts. Accordingly, this paper aims to address the mechanical behaviour of combined support systems quantitively, including rockbolt, TSL and shotcrete, through double shear tests.
2. Double shear test Double shear test is a laboratory test widely used for measuring the shear resistant of support systems in jointed rockmass, including rockbolt, shotcrete, TSLs, and their combinations. In this paper,
169
double shear tests refer to an arrangement of three blocks reinforced with a fully grouted rockbolt at different angles. The two end blocks are fixed and a load is applied to the middle block causing shear loading. Shotcrete and TSLs were applied to the surface of the test blocks, a schematic view of this test is shown in Fig. 1. The strength performance of reinforced element for pull and shear tests was carried out in embedment tubes with a minimum thickness of 10 mm according to British Standard (BS7861-Parts 2:2009), is considered a single guillotine test. The disadvantage of this kind of test as discussed by Li et al. (2014a), is that the steel tube is much stiffer than the rock which may introduce an error into the measurement. A quantitative result reported by Aziz et al. (2015), is the shear load per side for each double shear cable bolt was significantly higher than the results of testing cables in the single guillotine shear test. 2.1. Sample preparation Blocks used in double shear tests were prepared using a cement mortar to ensure consistency of the test condition. The cementitious mixture was poured into plywood moulds, each measuring 300 mm 300 mm 200 mm. A plastic conduit of 24 mm in diameter was placed in the moulds to create a hole for rockbolt installation that was set at either 90° or 45° to the direction of loading. This study sets the inclined angle at 45°. The orientation of the bolt installation is shown in Fig. 1. The cast samples were left to cure initially in the mould for the first 24 h. The blocks were then removed from the moulds and placed in a water bath for the remaining period of 28 days to cure. Cement was poured into 40 mm diameter cylinder for uniaxial compression strength tests. Three block samples were clamped together two rockbolts, that was fully encapsulated using 77 MPa high strength grout. To ensure that the rockbolt was fully covered, the grout was injected using a grout gun. By doing this, smooth and plane joint surfaces were achieved. The top surface of the blocks was roughed to ensure good adhesive strength of the shotcrete and TSL. An acrylic-based commercially available TSL was uniformly applied on the surface, that provides a tensile strength of 3.0 MPa within 7 days. Sprayed concrete was applied on the membrane surface after it set, with a typical bond strength (concrete to membrane) greater than 1.0 MPa. A commercially available shotcrete product is a ready-to-use, cement based shotcrete mix with reduced rebound and active corrosion inhibition. Proper curing is extremely important in this process. Wet curing is recommended for 3–5 days. A high performance structural polypropylene fibre was used. It provides resistance to rupture force thereby enhancing concrete toughness and crack control. The usage of fibres was 0.4 kg per bag of shotcrete. The mechanical properties of the test materials are shown in Table 1. 2.2. Test methodology A hydraulic universal testing machine with a capacity of 3600 kN was used for double shear testing under static loading, as shown in Fig. 2.The testing machine is comprised of an indicator/ control console containing the hydraulic pump unit and valve gear, a dial indicator and electrical controls. The main piston is located within the walls of a cylinder that has hydraulic fluid fed via the top and bottom of the piston. The load is applied from the bottom piston. Four continuously threaded rods, each 24 mm diameter and two steel plates were used as end constraints as seen in Fig. 1. This differs from a fully enclosed test sample arrangement in which it is assumed the test blocks are under a perfect confining pressure to simulate an infinite rock mass. However, in this situation, the test
170
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
Fig. 1. Schematic arrangement of double shear test with cut-away showing (a) Rockbolt at 90° to joint planes, (b) rockbolt at 45° and (c) arrangement with shotcrete on top surface.
Table 1 The mechanical properties of materials. Material type
The parameters
Test block
UCS: 52.6 MPa (28 days cured) Young’s modulus: 29.7 GPa
Tested
Joint surface
Cohesion (cj ): steel/grout 250 kPa Friction angle (uj ): 32°
Tested
Dilatancy angle (hj ): 0° Grouts
UCS: 77.0 MPa (7 days cured) Young’s modulus: 13.2 GPa
Tested
Rockbolt
Diameter: 16 mm Ultimate strength: 660 MPa Yield strength: 572 MPa Young modulus: 35 GPa Poison’s ratio: 0.3
Tested
TSL
Tensile strength: >3.0 MPa (7 days) Application thickness: 4–5 mm Elongation at break: >60% (7 days) Bond strength: 1.7 MPa
Provided by manufacture
Shotcrete
Application thickness: 40–50 mm Compressive strength: 78 MPa (7 days) Flexural strength: 5.0 MPa (7 days)
Provided by manufacture
Fibre
Length 45mm Tensile strength 750 MPa to 850 MPa
Provided by manufacture
block is assumed to be non-dilatant when its size is not big enough, thereby, making it difficult to understand the interaction between the test block and rockbolt. To ensure testing was conducted under shear loading by rock, three steel plates each of 20 mm thickness were placed underneath the blocks to ensure even load distribution and to avoid sample failure before yielding of the rockbolt. A series of tests were conducted to determine the contribution of the rockbolt and surface support system in resisting shear load in double shear test. Three tests were conducted with the rockbolt installed at 90° to the applied load with varying confining pressures of 0, 20 kN and 50 kN respectively, which determine the influence on shear strength of bolted joints (Li et al., 2014b). The test condition involving different surface support systems (shotcrete only, TSL only, combined shotcrete and TSL) applied to the top of the block with and without inclined rockbolts, are shown in Table 2. 2.3. Instrumentation installation Two 2.5 tonne capacity load cells were installed at both ends of the blocks. They were used to monitor the initial axial confining load of 20 kN and 50 kN and the normal load mobilised along the rockbolts due to shearing. Two linear variable differential transformers (LVDT) were used to measure the shear displacement during the test, as shown in Fig. 3. It should be noted that the measured displacement is not the deflection of rockbolt. The shear load means the vertical force applied from the bottom of the sample. Shear load, normal load and vertical displacement were recorded and stored in data-loggers for later analysis. In this paper, the test block and support elements are considered as one system. 3. Analysis of results 3.1. Surface support system results Fig. 4 shows the corresponding results of three types of surface support in resisting double shear under 20 kN normal load. The
Table 2 Test combinations. Test plans 3 rock blocks 90° Rockbolt
Fig. 2. Experimental loading facility.
Variables Confining Load 0 kN 20 kN 50 kN
Surface support without Rockbolt
TSL Shotcrete TSL & Shotcrete
45° Rockbolt Surface support
TSL Shotcrete TSL & Shotcrete
171
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
LVDT-1
Load Cell-1 LVDT-1
Fig. 5. Shear load versus displacement of combined support system in double shear tests.
LVDT-2 LVDT-2
Fig. 3. The set-up of test block with load cell and LVDT connection.
1.2 kJ, 1.6 kJ and 2.1 kJ respectively, shown in Table 3. The energy absorbed at 20 mm is critical to the system due to the DST system is gradually stable, with a constant residual shear strength. A support system with high stiffness is essential for maintaining stability and resisting deformation. It can be seen from Fig. 4 that a combination of shotcrete and TSL stiffens the DST system, which can absorb 11.7 kJ per square meter, allowing the rockbolts to absorb more energy and sustain large displacement without failure of the excavation. 3.2. The combination of rockbolt and surface support systems results
Fig. 4. Shear load versus displacement for different surface support combinations in double shear tests.
maximum shear capacity of TSL, shotcrete, and shotcrete and TSL were 41.3 kN, 95.4 kN, and 106.6 kN respectively. The increase in shear resistance for three support systems related to the unsupported sample, were 42%, 75% and 78% respectively, as shown in Table 3. The results confirm that surface support can play a significant role in resisting movement of joints. After an initial rapid increase in shear resistance, it remains uniform thereafter. During initial loading, the applied shear force mainly overcomes the frictional force and the bond strength of TSL/shotcrete. When vertical shear displacement occurs, the shotcrete and TSL initiated developing their tension and flexural strengths until the system fails. Shotcrete and TSL’s failure behaviour are described later in Section 5 in this paper. Energy absorption capacity increased nearly fourfold due to the application of the shotcrete and TSL. The variation of energy absorption characterises the role of surface support. The energy absorbed at 20 mm movement for above four tests is 0.5 kJ,
Fig. 5 shows the static double shear test results with different combinations of rockbolt and surface support systems, comparing the double shear test with a rockbolt only. The peak shear loads for the rockbolt, rockbolt with TSL, rockbolt with shotcrete, and rockbolt when combined with shotcrete and TSL were 167 kN, 185 kN, 219 kN and 232 kN respectively. A normal load of 20 kN was applied in each test to clamp the joint plane tightly. Additional surface support provides a higher shear resistance compared with the rockbolt only, the percentage increases were 10%, 24% and 28% respectively. When the rockbolt is combined with surface support systems, the increased amount is less than only surface support systems. The application of surface support maximizes the function of rockbolt in resisting the shear in DST system. Shear resistance of combined support elements increase compared with the individual element. A two dimensional stress state was changed to a three dimensional stress state in-situ due to the application of surface support. The role of surface support is more like a confinement in double shear tests. Moreover, the residual strength of combined support elements is higher than single rockbolt. The energy absorbed with a rockbolt only is 2.3 kJ after 20 mm vertical displacement. More energy was absorbed by applying the TSL and shotcrete, as seen in Table 4. However, the reinforced system becomes ‘less stiff’ due to the application of surface support. At the initial loading stage, the energy was dissipated by both the surface support and rockbolt. Large coverage of shotcrete and TSL is a good buffer for the dissipation of the energy. The slope of the curve
Table 3 Test results of surface support systems under shear loading. Parameters
No support
TSL
Shotcrete
TSL & Shotcrete
Peak shear load (kN) Percentage increase in shear load (%) Displacement at peak shear load (mm) Energy at 20 mm/sample (J) Energy absorbed at 20 mm (J/m2) Steady shear load (kN) Max displacement (mm) Peak normal load (kN)
24.0 – 5.1 515 2860 18.2 17 –
41.3 42% 6.8 1220 6800 39.6 45 36.2
95.4 75% 11.0 1600 8910 105 38 38.7
107.0 78% 14.3 2110 11,700 118 50 41.2
172
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
Table 4 The test results of surface support systems with inclined rockbolts. Support elements
Rockbolt only
TSL & Rockbolt
Shotcrete & Rockbolt
Shotcrete & TSL & Rockbolt
Peak shear load at 10 mm (kN) Percentage of increased shear load (%) Vertical displacement (mm) Steady shear load (kN) Peak normal load (kN) Energy absorbed at 20 mm (J) Energy absorbed at 20 mm (kJ/m2)
167 – 19.9 137 47.4 2340 13
185 10% 11.4 132 52.0 2860 16
219 24% 12.6 165 62.6 3790 21
232 28% 19.2 197 80.3 2540 14
refers to the stiffness of the support system, and the peak shear load of the system increased as shown in Fig. 5. Because of the large energy dissipation with surface support systems, full advantage of reinforcement element can be taken that allows the support systems to develop its maximum capacity.
Localised crushing adjacent to the joint plane
3.3. Effect of rockbolt position Under the same confinement condition, Fig. 6 compares the shear performance of a rockbolt in different orientations with no pretension. The test results are in agreement with the previous studies conducted by Egger and Zabuski (1991) and Grasselli (2005). It is noticed that the maximum load mobilised by the rockbolt occurs for an initial inclination to the joint within the range of 30–60°. From Fig. 6, it can be seen that the variation in bolt inclination affects both the maximum load and the rigidity of jointed system. 4. Test block and support element failure modes 4.1. Test block failure modes During testing, it was observed that the test block was crushed on both sides of the joint plane. A gap or void was created around the rockbolt, as shown in Fig. 7. The crushed areas may occur in the middle block and side blocks. Cracks were extended to the outer surface of the test block. Once the fracture propagates into the block, the shear resistance of the support system decreases, because the energy dissipates through these cracks. Consequently the system cannot sustain any further shear load. Through the sample failure mode in this test, the vulnerable area can be seen similar to in-situ situation. 4.2. Rockbolt failure modes It is worth noting that the measured vertical displacement is not the rebar shear deformation as the crushed test block was
Fig. 6. Comparison of shear load and shear displacement relationship with different bolt orientations.
Fig. 7. Crushing of test sample surrounding the rockbolt adjacent to the joint plane due to bending during shear.
unable to provide confinement to the rockbolt. Double shear test is more reliable than the single shear test, as it avoids asymmetric loading. It can also improve the stability of the sample. According to the test design, two side test blocks were fixed and cannot move during the test. In order to analyse the behaviour, the average deformation result was used to produce the deformation curves, as seen in Fig. 8. The test results show that deformation of the rockbolt is not closely related with the peak load because the degree of damage of the sample is different. Fig. 8 demonstrates the rockbolt was sheared in various deformations due to different combination of surface support and rockbolt. The results indicate that rockbolt deformation is related to stiffness of the support system. The application of surface support enables a less deformation of rockbolt in shear. The deflection of rockbolt was measured by angle a and the distance A-B, as shown in Fig. 9. Fig. 9 demonstrates a 16 mm diameter bolt, placed in a drilled hole inclined 45° to the joint, after a shear test. The distance between the two hinge points A-B and the angle a of the bolt were measured in each test (Table 5). The formation of hinges is characterized by compression on one side of the bolt, and tension on the other. The elongation of the bolt is negligible because the rockbolt protruded from the two sides of blocks. Although the rockbolt was fully grouted, the inclined hole was subjected to lateral extrusion and shear force leading to the grout being crushed and failing. When the middle block moved up, the rockbolt was subjected to shear force as well as axial force. Rockbolt extruding is a common phenomenon in-situ, because the resultant force direction is not in line of rockbolt. Generally, the bond strength between grout and rockbolt is high in pull-out tests, which has been verified in previous tests (Yazici and Kaiser, 1992). It is rare to find failure caused by broken grout, once the grout is subjected to an inclined force, it can easily be crushed and lose its bond strength. However, the rockbolt retains the capability to hold together the broken rockmass. It should be noted that only 16 mm diameter rockbolts were tested in this study. A comparison study between 16 mm and 20 mm diameter rockbolts conducted by Grasselli (2005) shows increasing the bolt diameter increases the rigidity of the reinforced system due to the larger area of the single reinforcement. Masoumi et al. (2013) indicated that the size effect of brittle material (test block) leads to different point load, UCS and triaxial compressive strength, which may influence the double shear testing results as well. However, the rockbolt is not expected to be affected.
173
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
Shear deformation (mm)
50
30
Rockbolt TSL + Rockbolt Shotcrete + Rockbolt
20
Shotcrete + TSL + Rockbolt
Joint plane
40
10 0 0
5
10
15
20
25
30
35
40
45
50
55
Distance along the rockbolt from one end to the middle (mm) Fig. 8. Shear deformation of along the length of rockbolt.
A
α
B
Fig. 9. A example of 16 mm rockbolt deformed in shear.
Table 5 Rockbolt deformation parameters. Test Schemes
Bolt deformed angle a (°)
A-B (mm)
Bolt deformed angle a (°)
A-B (mm)
Rockbolts only TSL + Rockbolts Shotcrete + Rockbolts Shotcrete + TSL + Rockbolts
32.1 23.2 21.1
81 103 127
23.2 23.6 25.5
82 125 118
20.2
103
10.1
102
4.3. Surface support failure Shotcrete and TSLs are chemically bonded to the rock surface, allowing for earlier mobilisation of support. The role of surface support is complementary to reinforcement as it helps the rockmass to support itself (Potvin, 2004). In this study, surface support is activated as a reaction to block movement between the blocks. Although there were problems in achieving a consistent TSL thickness, the test was successful. Different failure mechanisms were identified for the shear test.
(a) Shotcrete crack initiation
(d) Shotcrete fracture
Fig. 10 shows the surface support failure modes. Tensile and adhesive strengths are critical to surface support. During the shear test, shotcrete gradually was cut off as the middle block moved up. The fibres contribute to its tensile strength and toughness. Fig. 10(a) shows the shotcrete crack initiation and fibres linked the cracked shotcrete; (b) with increasing displacement, the shotcrete was totally broken and separated as the fibre was pulled out; (c) when shotcrete was combined with TSL, there was better bonding of the shotcrete, bending failure occurred; (d) shotcrete cracked at the collar of rockbolt due to the mobilisation of the rockbolt; (e) rockbolt is seen protruding out with fully encapsulated rockbolt; (f) TSL penetrated the joints, increasing the bond strength between the two joints. 5. Analysis of shear performance of support system Reinforcement and surface support systems interact with each other and are usually combined to form an integrated ground support system. Each element in this system has a different role to play because of differences in the manner in which it interacts with a rockmass. Reinforcement elements such as rockbolts and cable
(b) Shotcrete broken
(e) Rockbolt protruding Fig. 10. Support failure modes.
(c) Shotcrete +TSL bending
(f) TSL penetrating into shear plane
174
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
bolts are inserted within a rockmass to provide resistance to any internal rock movement and thereby preserve the integrity of the rockmass as a structural material. Surface support systems such as TSL and shotcrete are applied to the surface of a rockmass linking the reinforcement elements together and thereby resist surface deformation, preventing rock fragments from falling or being ejected. Fig. 11 compares the test results illustrating the differences between a rockmass having no support, reinforcement only, surface support only, and a combined or integrated reinforcement and surface support system. These differences in behaviour can be modelled as shown in Fig. 12. In the case of the integrated support system, the shear load and displacement relationship can be divided into three stages, signified as A, B and C in the graph, that corresponds to changes in the mechanical reaction of the support system. It should be noted that while the shear resistance of the integrated support system is greater than each of the individual elements it is less than the summation of those elements. This is because the surface support and reinforcement elements each become active at different times during the process of shearing. In Fig. 12, the first stage of the integrated support system corresponds to elastic behaviour, whereby it sustains a large increase in load with small displacement. During this stage, no cracks were evident in the test block and surface support. The mechanical performance of the surface support was mobilised after some displacement causing some relative movement between the joints. In practice, this is an important stage as it indicates the rockbolt will provide resistance to any differential movement that might otherwise take place between the test blocks. During this stage, the rockbolt can mobilise up to 85% of its peak resistance. The second stage of the integrated support system is characterized by a non-linear behaviour corresponding to yielding of the rockbolt and crushing of the rock immediately surrounding the
Fig. 11. Comparison of shear load versus displacement with different support systems.
Rockbolt with surface support Shear Load (kN)
A
B
C Surface support only No support
Displacement (mm) Fig. 12. Shear load and vertical displacement relationship with the different support systems.
rockbolt need to the joint planes as well as cracking the surface support. During this stage of non-elastic deformation, failure begins to occur between the rockbolt and grout as it detaches near to the joint plane. Once the peak load has been reached, yielding occurs and no further increase in shear load is necessary to deform the rockbolt. In addition, while the load bearing capacity of the surface support has been reduced due to cracking, it still retains some capacity to resist shear due to fibres contained in the shotcrete. In the third and final stage, the integrated system maintains a constant level of resistance to shear. This is due to a combination of the lateral confinement of the blocks and surface friction along the joint planes. Despite a constant residual load being applied, the rockbolt continues to yield as evident by the increase in displacement along the rock joints with crushing of the grout and rock immediately surrounding the rockbolt. Hence in a rock support system, each of the elements will behave differently whereby the rockbolt will initially undergo elastic deformation and with sufficient load will yield leading to plastic deformation. Whereas with the test block, loading will cause it to fail in a brittle manner with very little deformation before failure occurs. Consequently in the case where there is no external confinement and load is applied to the test block, the rockbolt will yield while the surrounding block (representing rock) will remain undeformed. Eventually with sufficient load, the rockbolt will split the block causing the whole test system to prematurely fail. 6. Conclusions A double shear testing facility was developed to assess differences in the behaviour of individual and combined rock support systems including rockbolt, shotcrete and thin spray-on liner (TSL). Each test sample comprised three blocks made from cementitious material that represented rock. Measurements were made of applied shear load and vertical displacement using a load cell and LVDT with subsequent determination of the energy absorption and maximum load bearing capacity. Differences were observed in the behaviour between the various support systems applied to the test sample. Overall it was found that the best performance was achieved when all three support elements were combined in terms of energy absorption and load bearing capacity. It has been found that use of shotcrete and TSL enhanced the absorption of energy. The use of shotcrete increased the initial stiffness in a combined support system. Improved performance was achieved with use of TSL when placed between the test sample and shotcrete as it improved the adhesive properties of the shotcrete leading to greater shear resistance. Rockbolt shear deformation is a function of the stiffness of rocksupport system. The application of surface support reduces the magnitude of deformation of the rockbolt in shear. A high stiffness rock-support system is essential in maintaining stability by resisting any initial deformation and loss of strength. In the tests, displacements of up to 60 mm between the blocks in the test sample were achieved however while the rockbolt yielded, it did not fail nor was there any sign of necking. This indicates that despite this large level of displacement, the rockbolt was still able to provide support. Acknowledgement The first author would like to thank the China Scholarship Council (CSC) and UNSW Australia’s School of Mining Engineering for supporting her study in Australia. Special thanks are due to Mr Kanchana Gamage and Mr Mark Whelan for their help in conducting the tests.
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
References Aziz, N., Hawker, R., Mirza, A., Nemcik, J., Li, X., Rasekh, H., 2015. Strength characteristics of secura hollow groutable cable bolts. In: Proceedings The 2015 Coal Operators’s Conference, Wollongong, Australia. Aziz, N., Pratt, D., Williams, R., 2003. Double Shear Testing of Bolts. Proceedings Coal Operators’ Conference 2003, Wollongong, NSW, Australia, pp. 154–161. Craig, P., Aziz, N., 2010. Shear Testing of 28 mm Hollow Strand ’TG’ Cable Bolt. Proceedings Coal Operators’ Conference 2010, pp. 171–179. Darling, P., 2011. In: SME Mining Engineering Handbook. SME. Egger, P., Zabuski, L., 1991. Behaviour of rough bolted joints in direct shear tests. In: Proceedings 7th ISRM Congress, Aachen, Germany, pp. 1285–1288. Ferrero, A., 1995. The shear strength of reinforced rock joints. Int. J. Rock Mech. Mining Sci. Geomech. Abstracts 32 (6), 595–605. Grasselli, G., 2005. 3D Behaviour of bolted rock joints: experimental and numerical study. Int. J. Rock Mech. Min. Sci. 42 (1), 13–24. Haas, C.J., 1976. Shear resistance of rock bolts. Soc. Min. Eng. 260, 32–41. Haas, J., 1981. Analysis of rock bolting to prevent shear movement in fractured ground. Min. Eng. 33 (6), 698–704. Hagan, P., Weckert, S., 2004. Anchorage and Failure Mechanisms of Fully Encapsulated Rockbolts – Stage 2. UNSW Mining Research Centre, Sydney, Australia. Haile, A.T., 1999. Observation of the dynamic support performance of South African tunnel support systems. In: Proceedings Rock Support and Reinforcement Practice in Mining, Kalgoorlie, Australia, pp. 335–341. Hartman, W., Hebblewhite, B., 2003. Understanding the Performance of Pock Reinforcement Elements Under Shear Loading Through Laboratory Testing—A 30-Year History. Proceedings 1 St AGCM Conference, Sydney, Australia. Hoek, E., Kaiser, P.K., Bawden, W.F., 2000. In: Support of Underground Excavations in Hard Rock. Holmberg, M., 1991. The Mechanical Behaviour of Untensioned Grouted Rock Bolts. Department of Soil and Rock Mechanics. Hyett, A., Spearing, S., 2013. Enlightening Bolts: Using Distributed Optical Sensing to Measure the Strain Profile Along Fully Grouted Rock Bolts. Proceedings 32nd International Conference on Ground Control in Mining. Jalalifar, H., 2006. A New Approach in Determining the Load Transfer Mechanism in Fully Grouted Bolts PhD thesis (unpublished). University Of Wollongong, NSW, Australia. Jalalifar, H., Aziz, N., 2010. Analytical behaviour of bolt–joint intersection under lateral loading conditions. Rock Mech. Rock Eng. 43 (1), 89–94. Li, C., 2010. Field observations of rock bolts in high stress rock masses. Rock Mech. Rock Eng. 43 (4), 491–496. Li, L., Hagan, P., Saydam, S., 2014a. A review of ground support systems performance subjected to dynamic loading. Proceedings Eurock 2014, Vigo, Spain. CRC Press, pp. 1063–1068.
175
Li, L., Hagan, P., Saydam, S., 2014b. Tensile stress mobilization along a rockbolt under shear load. Proceedings International Young Scholars’ Symposium On Rock Mechanics, Xi’an, China. Mahony, L., Hagan, P., 2006. A laboratory facility to study the behaviour of reinforced elements subjected to shear. Proceedings Coal Operators’ Conference 2006, Wollongong, NSW, Australia, pp. 245–254. Malmgren, L., 2007. Strength, ductility and stiffness of fibre-reinforced shotcrete. Mag. Concr. Res. [Online] 59. Available:.. Malmgren, L., Nordlund, E., 2008. Interaction of shotcrete with rock and rock bolts— A numerical study. Int. J. Rock Mech. Min. Sci. 45 (4), 538–553. Malmgren, L., Nordlund, E., Rolund, S., 2005. Adhesion strength and shrinkage of shotcrete. Tunn. Undergr. Space Technol. 20 (1), 33–48. Masoumi, H., Douglas, K., Russell, A., 2013. Investigation Into the Mechanical Behaviour of Intact Rock at Different Sizes, UNSW Australia. McHugh, E., Signer, S.D., 1999. Roof bolt response to shear stress: laboratory analysis. In: Proceedings 18th International Conference on Ground Control in Mining, Morgantown, WV, pp. 232–238. Pellet, F., 1993. Strength and Deformability of Jointed Rock Masses Reinforced by Rock Bolts (LMR). Pellet, F., Egger, P., 1996. Analytical model for the mechanical behaviour of bolted rock joints subjected to shearing. Rock Mech. Rock Eng. 29 (2), 73–97. Potvin, Y., 2004. Thin spary-on liners (TSLs) – a quick reference guide. In: Potvin, Y., Stacey, T., Hadjigeorgiou, J. (Eds.), Surface Support in Mining. Australian Centre for Geomechanics, WA Australia, pp. 3–43. Richardson, J., Mitra, R., Saydam, S., 2009. Investigation of Thin Spray-on Liners Using Numerical Modeling. Proceedings (American Rock Mechanics Association: Arma). Saydam, S., Yilmaz, H., Stacey, T., 2004. A new testing approach for thin spray-on liners: double-sided shear strength (DSS) test. In: Potvin, Y., Stacey, T., Hadjigeorgiou, J. (Eds.), Surface Support in Mining. Australian Centre for Geomechanics, WA, Australia, pp. 151–156. Spang, K., Egger, P., 1990. Action of fully-grouted bolts in jointed rock and factors of influence. Rock Mech. Rock Eng. 23 (3), 201–229. Stacey, T., 2001. Review of membrane support mechanisms, loading mechanisms, desired membrane performance, and appropriate test methods. Yazici, S., Kaiser, P.K., 1992. Bond strength of grouted cable bolts. Int. J. Rock Mech. Min. Sci. Geomech. Abstracts 29 (3), 279–292. Yilmaz, H., 2007. Shear-bond strength testing of thin spary-on liners. J. South Afric. Inst. Min. Metall. 107, 519–530. Yoshinaka, R., Sakaguchi, S., Shimizu, T., Arai, H., Kato, E., 1987. Experimental study on the rock bolt reinforcement in discontinuous rocks. In: Proceedings the 6th ISRM Congress.
Contents lists available at ScienceDirect
Tunnelling and Underground Space Technology journal homepage: www.elsevier.com/locate/tust
Shear resistance contribution of support systems in double shear test L. Li ⇑, P.C. Hagan, S. Saydam, B. Hebblewhite School of Mining Engineering, UNSW Australia, Sydney, Australia
a r t i c l e
i n f o
Article history: Received 17 August 2015 Received in revised form 1 December 2015 Accepted 23 March 2016
Keywords: Double shear test (DST) Rockbolt Shotcrete TSLs Shear resistance
a b s t r a c t Rockbolt and surface support systems such as shotcrete and thin spray-on liners (TSLs) are widely used as underground support elements to resist the convergence and maintain the stability of excavations. In order to evaluate the bearing capacity of combined reinforced rockbolt and surface support systems in preventing sliding along discontinuities, double shear tests (DST) was carried out using fully grouted rockbolts installed in three separate blocks. These blocks were covered with a 5 mm layer of TSL followed by a 50 mm layer of shotcrete. Two rockbolts were installed at an inclined angle of 45°, and 20 kN lateral constraining force was applied to clamp together the three blocks. Three different support combinations were tested: 50 mm shotcrete only, 5 mm TSL only, and combined shotcrete and TSL, with and without rockbolts. It was confirmed that the shotcrete plays a mechanical role in resisting the shear load, and TSLs increase the bond strength between shotcrete and substrate replicating the side wall of an excavation. The contribution of rockbolt and surface support system in resisting joint movement was also compared. The failure mechanism of rock substrate, rockbolt and surface support system was also analysed. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Design, installation and monitoring support systems are a major consideration in underground mining operations. The use of rockbolts dates back to the late 1900 s in mining and includes mechanically anchored (slot and wedge or expansion shell) bolts, fully grouted bolts and resin bolts. Generally, the purpose of a rockbolt is to develop inherent strength and stiffness in a rock mass by securing the loose surface blocks to deep stable rock masses (Hoek et al., 2000). Rockbolts develop their tensile and shear strength during the convergence of rock mass, in which the rockbolt can be considered as part of a rock mass. Rockbolts when combined with shotcrete and/or TSLs can cope with almost any ground condition encountered during mining and tunnelling. Thin spray-on liners (TSLs) are multicomponent polymer materials that can be applied to the rock mass surface as a sealant or as a surface support (Darling, 2011). When the shotcrete is sprayed onto the surface of rock, the adhesion strength develops over time. Moreover, any fracture on the surface is filled and sealed by the application of high speed injected shotcrete, helping maintain the integrity of the rock mass. Failure is normally due to adhesion failure between the shotcrete and rock interface that can be initiated after a small displacement.
⇑ Corresponding author. E-mail address: [email protected] (L. Li). http://dx.doi.org/10.1016/j.tust.2016.03.011 0886-7798/Ó 2016 Elsevier Ltd. All rights reserved.
Over the past four decades, research has confirmed that shear performance can be just as significant as tensile performance. This is evident by a high proportion of rockbolts that have been found to fail in shear in high stress rock masses (Li, 2010) and under rockburst conditions (Haile, 1999). A case study conducted by McHugh and Signer (1999) indicated that shear loading contributed significantly to the failure of anchoring systems. Malmgren and Nordlund (2008) pointed out that the behaviour of shotcrete in interaction with rock is very complex and the performance of shotcrete is influenced by a number of parameters, especially the present of discontinuities. However, there is less known about the effect of discontinuities on the failure of shotcrete. Many experimental tests have been performed in order to study the mechanical behaviour of rockbolts in resisting the shear force based on single shear tests (Haas, 1976; Haas, 1981; Yoshinaka et al., 1987; Egger and Zabuski, 1991; Holmberg, 1991; Pellet, 1993; Ferrero, 1995; Pellet and Egger, 1996; McHugh and Signer, 1999; Mahony and Hagan, 2006) and double shear tests in the laboratory (Aziz et al., 2003; Grasselli, 2005; Jalalifar, 2006; Craig and Aziz, 2010; Jalalifar and Aziz, 2010; Hyett and Spearing, 2013). The factors that can influence shear resistance include (Hartman and Hebblewhite, 2003): Rock mass condition: joint aperture/roughness/dilatancy, Strength and size of rock, The reinforcement elements system: type/diameter/mechanical properties of anchoring systems,
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
Grout type/thickness, Inclination of bolt and joint plane, and And loading conditions: pretension, normal pressure, confinement condition. Hass (1976) studied the ‘natural fracture’ of shear surface by splitting limestone blocks with 25 mm surface roughness. The results showed that surface roughness increased the initial shear resistance. However, the shear resistance decreased rapidly as the asperities were sheared off. Hass (1976) further compared the anchorage capacity of fully grouted rockbolts in a rifled hole and a smooth hole. He concluded that roughness plays a major role in increasing the tensile capacity and shear force. Grasselli (2005) pointed out that the variation in bolt inclination affects the maximum load mobilised in the rockbolt as well as the rigidity of a jointed system. Spang and Egger (1990) carried out shear tests with bolt diameters of 8 mm, 10 mm, and 40 mm. The results indicated that the maximum shear force and displacement is proportional to bolt diameter. Pull-out capacity of a rockbolt in different resin encapsulation annulus was examined by Hagan and Weckert (2004). They found that pull-out capacity of fully encapsulated rockbolt is not affected by changes in resin annulus of less than 4 mm. However, there was a reduction in pull-out capacity at greater resin thicknesses. There is little research about the resin annulus effect in shear tests. Hard rocks usually have higher shear resistance compared with soft rock and rockbolts embedded in harder rock usually require smaller displacements to attain a given resistance than those in softer rock (Spang and Egger, 1990). Ferrero (1995) mentioned that pretensioning does not influence the maximum resistance of the reinforced shear joint; however, it will heavily influence the magnitude of the shear displacement. An analysis of double shear testing results shows that the behaviour of a rockbolt under load follows three distinct stages, linear behaviour, non-linear behaviour and unconstrained plastic deformation. The contribution to shear strength is defined as the difference between the applied shear force and the frictional strength normalised to the ultimate tensile load of the bolt. The first and second stage refers to loading up to 75% and 90% of maximum load capacity respectively for a full scale rockbolt (Grasselli, 2005). Polypropylene-fibre is used to increase the ductility of shotcrete, leading to stress redistribution (Malmgren, 2007). The possible mechanism for shotcrete in interaction with the rock is by block interlock theory. Also the effect of sealing the joints by shotcrete helps to maintain the integrity of the rock mass (Malmgren et al., 2005; Stacey, 2001). The shear-bond strength of TSL was assessed by Yilmaz (2007) and the double-sided shear tests for various TSLs were conducted by Saydam et al. (2004) and Richardson et al. (2009). These researchers examined the capacity of shotcrete or TSLs under shear conditions; however, little is known about the performance of shotcrete and TSLs in combination with rockbolts. These studies examined shear resistance capacity associated with the movement of rock mass; however, they are insufficient to provide an integrated approach for combined underground support systems. It is common practice that surface support systems are often used in combination with rockbolts. Accordingly, this paper aims to address the mechanical behaviour of combined support systems quantitively, including rockbolt, TSL and shotcrete, through double shear tests.
2. Double shear test Double shear test is a laboratory test widely used for measuring the shear resistant of support systems in jointed rockmass, including rockbolt, shotcrete, TSLs, and their combinations. In this paper,
169
double shear tests refer to an arrangement of three blocks reinforced with a fully grouted rockbolt at different angles. The two end blocks are fixed and a load is applied to the middle block causing shear loading. Shotcrete and TSLs were applied to the surface of the test blocks, a schematic view of this test is shown in Fig. 1. The strength performance of reinforced element for pull and shear tests was carried out in embedment tubes with a minimum thickness of 10 mm according to British Standard (BS7861-Parts 2:2009), is considered a single guillotine test. The disadvantage of this kind of test as discussed by Li et al. (2014a), is that the steel tube is much stiffer than the rock which may introduce an error into the measurement. A quantitative result reported by Aziz et al. (2015), is the shear load per side for each double shear cable bolt was significantly higher than the results of testing cables in the single guillotine shear test. 2.1. Sample preparation Blocks used in double shear tests were prepared using a cement mortar to ensure consistency of the test condition. The cementitious mixture was poured into plywood moulds, each measuring 300 mm 300 mm 200 mm. A plastic conduit of 24 mm in diameter was placed in the moulds to create a hole for rockbolt installation that was set at either 90° or 45° to the direction of loading. This study sets the inclined angle at 45°. The orientation of the bolt installation is shown in Fig. 1. The cast samples were left to cure initially in the mould for the first 24 h. The blocks were then removed from the moulds and placed in a water bath for the remaining period of 28 days to cure. Cement was poured into 40 mm diameter cylinder for uniaxial compression strength tests. Three block samples were clamped together two rockbolts, that was fully encapsulated using 77 MPa high strength grout. To ensure that the rockbolt was fully covered, the grout was injected using a grout gun. By doing this, smooth and plane joint surfaces were achieved. The top surface of the blocks was roughed to ensure good adhesive strength of the shotcrete and TSL. An acrylic-based commercially available TSL was uniformly applied on the surface, that provides a tensile strength of 3.0 MPa within 7 days. Sprayed concrete was applied on the membrane surface after it set, with a typical bond strength (concrete to membrane) greater than 1.0 MPa. A commercially available shotcrete product is a ready-to-use, cement based shotcrete mix with reduced rebound and active corrosion inhibition. Proper curing is extremely important in this process. Wet curing is recommended for 3–5 days. A high performance structural polypropylene fibre was used. It provides resistance to rupture force thereby enhancing concrete toughness and crack control. The usage of fibres was 0.4 kg per bag of shotcrete. The mechanical properties of the test materials are shown in Table 1. 2.2. Test methodology A hydraulic universal testing machine with a capacity of 3600 kN was used for double shear testing under static loading, as shown in Fig. 2.The testing machine is comprised of an indicator/ control console containing the hydraulic pump unit and valve gear, a dial indicator and electrical controls. The main piston is located within the walls of a cylinder that has hydraulic fluid fed via the top and bottom of the piston. The load is applied from the bottom piston. Four continuously threaded rods, each 24 mm diameter and two steel plates were used as end constraints as seen in Fig. 1. This differs from a fully enclosed test sample arrangement in which it is assumed the test blocks are under a perfect confining pressure to simulate an infinite rock mass. However, in this situation, the test
170
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
Fig. 1. Schematic arrangement of double shear test with cut-away showing (a) Rockbolt at 90° to joint planes, (b) rockbolt at 45° and (c) arrangement with shotcrete on top surface.
Table 1 The mechanical properties of materials. Material type
The parameters
Test block
UCS: 52.6 MPa (28 days cured) Young’s modulus: 29.7 GPa
Tested
Joint surface
Cohesion (cj ): steel/grout 250 kPa Friction angle (uj ): 32°
Tested
Dilatancy angle (hj ): 0° Grouts
UCS: 77.0 MPa (7 days cured) Young’s modulus: 13.2 GPa
Tested
Rockbolt
Diameter: 16 mm Ultimate strength: 660 MPa Yield strength: 572 MPa Young modulus: 35 GPa Poison’s ratio: 0.3
Tested
TSL
Tensile strength: >3.0 MPa (7 days) Application thickness: 4–5 mm Elongation at break: >60% (7 days) Bond strength: 1.7 MPa
Provided by manufacture
Shotcrete
Application thickness: 40–50 mm Compressive strength: 78 MPa (7 days) Flexural strength: 5.0 MPa (7 days)
Provided by manufacture
Fibre
Length 45mm Tensile strength 750 MPa to 850 MPa
Provided by manufacture
block is assumed to be non-dilatant when its size is not big enough, thereby, making it difficult to understand the interaction between the test block and rockbolt. To ensure testing was conducted under shear loading by rock, three steel plates each of 20 mm thickness were placed underneath the blocks to ensure even load distribution and to avoid sample failure before yielding of the rockbolt. A series of tests were conducted to determine the contribution of the rockbolt and surface support system in resisting shear load in double shear test. Three tests were conducted with the rockbolt installed at 90° to the applied load with varying confining pressures of 0, 20 kN and 50 kN respectively, which determine the influence on shear strength of bolted joints (Li et al., 2014b). The test condition involving different surface support systems (shotcrete only, TSL only, combined shotcrete and TSL) applied to the top of the block with and without inclined rockbolts, are shown in Table 2. 2.3. Instrumentation installation Two 2.5 tonne capacity load cells were installed at both ends of the blocks. They were used to monitor the initial axial confining load of 20 kN and 50 kN and the normal load mobilised along the rockbolts due to shearing. Two linear variable differential transformers (LVDT) were used to measure the shear displacement during the test, as shown in Fig. 3. It should be noted that the measured displacement is not the deflection of rockbolt. The shear load means the vertical force applied from the bottom of the sample. Shear load, normal load and vertical displacement were recorded and stored in data-loggers for later analysis. In this paper, the test block and support elements are considered as one system. 3. Analysis of results 3.1. Surface support system results Fig. 4 shows the corresponding results of three types of surface support in resisting double shear under 20 kN normal load. The
Table 2 Test combinations. Test plans 3 rock blocks 90° Rockbolt
Fig. 2. Experimental loading facility.
Variables Confining Load 0 kN 20 kN 50 kN
Surface support without Rockbolt
TSL Shotcrete TSL & Shotcrete
45° Rockbolt Surface support
TSL Shotcrete TSL & Shotcrete
171
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
LVDT-1
Load Cell-1 LVDT-1
Fig. 5. Shear load versus displacement of combined support system in double shear tests.
LVDT-2 LVDT-2
Fig. 3. The set-up of test block with load cell and LVDT connection.
1.2 kJ, 1.6 kJ and 2.1 kJ respectively, shown in Table 3. The energy absorbed at 20 mm is critical to the system due to the DST system is gradually stable, with a constant residual shear strength. A support system with high stiffness is essential for maintaining stability and resisting deformation. It can be seen from Fig. 4 that a combination of shotcrete and TSL stiffens the DST system, which can absorb 11.7 kJ per square meter, allowing the rockbolts to absorb more energy and sustain large displacement without failure of the excavation. 3.2. The combination of rockbolt and surface support systems results
Fig. 4. Shear load versus displacement for different surface support combinations in double shear tests.
maximum shear capacity of TSL, shotcrete, and shotcrete and TSL were 41.3 kN, 95.4 kN, and 106.6 kN respectively. The increase in shear resistance for three support systems related to the unsupported sample, were 42%, 75% and 78% respectively, as shown in Table 3. The results confirm that surface support can play a significant role in resisting movement of joints. After an initial rapid increase in shear resistance, it remains uniform thereafter. During initial loading, the applied shear force mainly overcomes the frictional force and the bond strength of TSL/shotcrete. When vertical shear displacement occurs, the shotcrete and TSL initiated developing their tension and flexural strengths until the system fails. Shotcrete and TSL’s failure behaviour are described later in Section 5 in this paper. Energy absorption capacity increased nearly fourfold due to the application of the shotcrete and TSL. The variation of energy absorption characterises the role of surface support. The energy absorbed at 20 mm movement for above four tests is 0.5 kJ,
Fig. 5 shows the static double shear test results with different combinations of rockbolt and surface support systems, comparing the double shear test with a rockbolt only. The peak shear loads for the rockbolt, rockbolt with TSL, rockbolt with shotcrete, and rockbolt when combined with shotcrete and TSL were 167 kN, 185 kN, 219 kN and 232 kN respectively. A normal load of 20 kN was applied in each test to clamp the joint plane tightly. Additional surface support provides a higher shear resistance compared with the rockbolt only, the percentage increases were 10%, 24% and 28% respectively. When the rockbolt is combined with surface support systems, the increased amount is less than only surface support systems. The application of surface support maximizes the function of rockbolt in resisting the shear in DST system. Shear resistance of combined support elements increase compared with the individual element. A two dimensional stress state was changed to a three dimensional stress state in-situ due to the application of surface support. The role of surface support is more like a confinement in double shear tests. Moreover, the residual strength of combined support elements is higher than single rockbolt. The energy absorbed with a rockbolt only is 2.3 kJ after 20 mm vertical displacement. More energy was absorbed by applying the TSL and shotcrete, as seen in Table 4. However, the reinforced system becomes ‘less stiff’ due to the application of surface support. At the initial loading stage, the energy was dissipated by both the surface support and rockbolt. Large coverage of shotcrete and TSL is a good buffer for the dissipation of the energy. The slope of the curve
Table 3 Test results of surface support systems under shear loading. Parameters
No support
TSL
Shotcrete
TSL & Shotcrete
Peak shear load (kN) Percentage increase in shear load (%) Displacement at peak shear load (mm) Energy at 20 mm/sample (J) Energy absorbed at 20 mm (J/m2) Steady shear load (kN) Max displacement (mm) Peak normal load (kN)
24.0 – 5.1 515 2860 18.2 17 –
41.3 42% 6.8 1220 6800 39.6 45 36.2
95.4 75% 11.0 1600 8910 105 38 38.7
107.0 78% 14.3 2110 11,700 118 50 41.2
172
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
Table 4 The test results of surface support systems with inclined rockbolts. Support elements
Rockbolt only
TSL & Rockbolt
Shotcrete & Rockbolt
Shotcrete & TSL & Rockbolt
Peak shear load at 10 mm (kN) Percentage of increased shear load (%) Vertical displacement (mm) Steady shear load (kN) Peak normal load (kN) Energy absorbed at 20 mm (J) Energy absorbed at 20 mm (kJ/m2)
167 – 19.9 137 47.4 2340 13
185 10% 11.4 132 52.0 2860 16
219 24% 12.6 165 62.6 3790 21
232 28% 19.2 197 80.3 2540 14
refers to the stiffness of the support system, and the peak shear load of the system increased as shown in Fig. 5. Because of the large energy dissipation with surface support systems, full advantage of reinforcement element can be taken that allows the support systems to develop its maximum capacity.
Localised crushing adjacent to the joint plane
3.3. Effect of rockbolt position Under the same confinement condition, Fig. 6 compares the shear performance of a rockbolt in different orientations with no pretension. The test results are in agreement with the previous studies conducted by Egger and Zabuski (1991) and Grasselli (2005). It is noticed that the maximum load mobilised by the rockbolt occurs for an initial inclination to the joint within the range of 30–60°. From Fig. 6, it can be seen that the variation in bolt inclination affects both the maximum load and the rigidity of jointed system. 4. Test block and support element failure modes 4.1. Test block failure modes During testing, it was observed that the test block was crushed on both sides of the joint plane. A gap or void was created around the rockbolt, as shown in Fig. 7. The crushed areas may occur in the middle block and side blocks. Cracks were extended to the outer surface of the test block. Once the fracture propagates into the block, the shear resistance of the support system decreases, because the energy dissipates through these cracks. Consequently the system cannot sustain any further shear load. Through the sample failure mode in this test, the vulnerable area can be seen similar to in-situ situation. 4.2. Rockbolt failure modes It is worth noting that the measured vertical displacement is not the rebar shear deformation as the crushed test block was
Fig. 6. Comparison of shear load and shear displacement relationship with different bolt orientations.
Fig. 7. Crushing of test sample surrounding the rockbolt adjacent to the joint plane due to bending during shear.
unable to provide confinement to the rockbolt. Double shear test is more reliable than the single shear test, as it avoids asymmetric loading. It can also improve the stability of the sample. According to the test design, two side test blocks were fixed and cannot move during the test. In order to analyse the behaviour, the average deformation result was used to produce the deformation curves, as seen in Fig. 8. The test results show that deformation of the rockbolt is not closely related with the peak load because the degree of damage of the sample is different. Fig. 8 demonstrates the rockbolt was sheared in various deformations due to different combination of surface support and rockbolt. The results indicate that rockbolt deformation is related to stiffness of the support system. The application of surface support enables a less deformation of rockbolt in shear. The deflection of rockbolt was measured by angle a and the distance A-B, as shown in Fig. 9. Fig. 9 demonstrates a 16 mm diameter bolt, placed in a drilled hole inclined 45° to the joint, after a shear test. The distance between the two hinge points A-B and the angle a of the bolt were measured in each test (Table 5). The formation of hinges is characterized by compression on one side of the bolt, and tension on the other. The elongation of the bolt is negligible because the rockbolt protruded from the two sides of blocks. Although the rockbolt was fully grouted, the inclined hole was subjected to lateral extrusion and shear force leading to the grout being crushed and failing. When the middle block moved up, the rockbolt was subjected to shear force as well as axial force. Rockbolt extruding is a common phenomenon in-situ, because the resultant force direction is not in line of rockbolt. Generally, the bond strength between grout and rockbolt is high in pull-out tests, which has been verified in previous tests (Yazici and Kaiser, 1992). It is rare to find failure caused by broken grout, once the grout is subjected to an inclined force, it can easily be crushed and lose its bond strength. However, the rockbolt retains the capability to hold together the broken rockmass. It should be noted that only 16 mm diameter rockbolts were tested in this study. A comparison study between 16 mm and 20 mm diameter rockbolts conducted by Grasselli (2005) shows increasing the bolt diameter increases the rigidity of the reinforced system due to the larger area of the single reinforcement. Masoumi et al. (2013) indicated that the size effect of brittle material (test block) leads to different point load, UCS and triaxial compressive strength, which may influence the double shear testing results as well. However, the rockbolt is not expected to be affected.
173
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
Shear deformation (mm)
50
30
Rockbolt TSL + Rockbolt Shotcrete + Rockbolt
20
Shotcrete + TSL + Rockbolt
Joint plane
40
10 0 0
5
10
15
20
25
30
35
40
45
50
55
Distance along the rockbolt from one end to the middle (mm) Fig. 8. Shear deformation of along the length of rockbolt.
A
α
B
Fig. 9. A example of 16 mm rockbolt deformed in shear.
Table 5 Rockbolt deformation parameters. Test Schemes
Bolt deformed angle a (°)
A-B (mm)
Bolt deformed angle a (°)
A-B (mm)
Rockbolts only TSL + Rockbolts Shotcrete + Rockbolts Shotcrete + TSL + Rockbolts
32.1 23.2 21.1
81 103 127
23.2 23.6 25.5
82 125 118
20.2
103
10.1
102
4.3. Surface support failure Shotcrete and TSLs are chemically bonded to the rock surface, allowing for earlier mobilisation of support. The role of surface support is complementary to reinforcement as it helps the rockmass to support itself (Potvin, 2004). In this study, surface support is activated as a reaction to block movement between the blocks. Although there were problems in achieving a consistent TSL thickness, the test was successful. Different failure mechanisms were identified for the shear test.
(a) Shotcrete crack initiation
(d) Shotcrete fracture
Fig. 10 shows the surface support failure modes. Tensile and adhesive strengths are critical to surface support. During the shear test, shotcrete gradually was cut off as the middle block moved up. The fibres contribute to its tensile strength and toughness. Fig. 10(a) shows the shotcrete crack initiation and fibres linked the cracked shotcrete; (b) with increasing displacement, the shotcrete was totally broken and separated as the fibre was pulled out; (c) when shotcrete was combined with TSL, there was better bonding of the shotcrete, bending failure occurred; (d) shotcrete cracked at the collar of rockbolt due to the mobilisation of the rockbolt; (e) rockbolt is seen protruding out with fully encapsulated rockbolt; (f) TSL penetrated the joints, increasing the bond strength between the two joints. 5. Analysis of shear performance of support system Reinforcement and surface support systems interact with each other and are usually combined to form an integrated ground support system. Each element in this system has a different role to play because of differences in the manner in which it interacts with a rockmass. Reinforcement elements such as rockbolts and cable
(b) Shotcrete broken
(e) Rockbolt protruding Fig. 10. Support failure modes.
(c) Shotcrete +TSL bending
(f) TSL penetrating into shear plane
174
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
bolts are inserted within a rockmass to provide resistance to any internal rock movement and thereby preserve the integrity of the rockmass as a structural material. Surface support systems such as TSL and shotcrete are applied to the surface of a rockmass linking the reinforcement elements together and thereby resist surface deformation, preventing rock fragments from falling or being ejected. Fig. 11 compares the test results illustrating the differences between a rockmass having no support, reinforcement only, surface support only, and a combined or integrated reinforcement and surface support system. These differences in behaviour can be modelled as shown in Fig. 12. In the case of the integrated support system, the shear load and displacement relationship can be divided into three stages, signified as A, B and C in the graph, that corresponds to changes in the mechanical reaction of the support system. It should be noted that while the shear resistance of the integrated support system is greater than each of the individual elements it is less than the summation of those elements. This is because the surface support and reinforcement elements each become active at different times during the process of shearing. In Fig. 12, the first stage of the integrated support system corresponds to elastic behaviour, whereby it sustains a large increase in load with small displacement. During this stage, no cracks were evident in the test block and surface support. The mechanical performance of the surface support was mobilised after some displacement causing some relative movement between the joints. In practice, this is an important stage as it indicates the rockbolt will provide resistance to any differential movement that might otherwise take place between the test blocks. During this stage, the rockbolt can mobilise up to 85% of its peak resistance. The second stage of the integrated support system is characterized by a non-linear behaviour corresponding to yielding of the rockbolt and crushing of the rock immediately surrounding the
Fig. 11. Comparison of shear load versus displacement with different support systems.
Rockbolt with surface support Shear Load (kN)
A
B
C Surface support only No support
Displacement (mm) Fig. 12. Shear load and vertical displacement relationship with the different support systems.
rockbolt need to the joint planes as well as cracking the surface support. During this stage of non-elastic deformation, failure begins to occur between the rockbolt and grout as it detaches near to the joint plane. Once the peak load has been reached, yielding occurs and no further increase in shear load is necessary to deform the rockbolt. In addition, while the load bearing capacity of the surface support has been reduced due to cracking, it still retains some capacity to resist shear due to fibres contained in the shotcrete. In the third and final stage, the integrated system maintains a constant level of resistance to shear. This is due to a combination of the lateral confinement of the blocks and surface friction along the joint planes. Despite a constant residual load being applied, the rockbolt continues to yield as evident by the increase in displacement along the rock joints with crushing of the grout and rock immediately surrounding the rockbolt. Hence in a rock support system, each of the elements will behave differently whereby the rockbolt will initially undergo elastic deformation and with sufficient load will yield leading to plastic deformation. Whereas with the test block, loading will cause it to fail in a brittle manner with very little deformation before failure occurs. Consequently in the case where there is no external confinement and load is applied to the test block, the rockbolt will yield while the surrounding block (representing rock) will remain undeformed. Eventually with sufficient load, the rockbolt will split the block causing the whole test system to prematurely fail. 6. Conclusions A double shear testing facility was developed to assess differences in the behaviour of individual and combined rock support systems including rockbolt, shotcrete and thin spray-on liner (TSL). Each test sample comprised three blocks made from cementitious material that represented rock. Measurements were made of applied shear load and vertical displacement using a load cell and LVDT with subsequent determination of the energy absorption and maximum load bearing capacity. Differences were observed in the behaviour between the various support systems applied to the test sample. Overall it was found that the best performance was achieved when all three support elements were combined in terms of energy absorption and load bearing capacity. It has been found that use of shotcrete and TSL enhanced the absorption of energy. The use of shotcrete increased the initial stiffness in a combined support system. Improved performance was achieved with use of TSL when placed between the test sample and shotcrete as it improved the adhesive properties of the shotcrete leading to greater shear resistance. Rockbolt shear deformation is a function of the stiffness of rocksupport system. The application of surface support reduces the magnitude of deformation of the rockbolt in shear. A high stiffness rock-support system is essential in maintaining stability by resisting any initial deformation and loss of strength. In the tests, displacements of up to 60 mm between the blocks in the test sample were achieved however while the rockbolt yielded, it did not fail nor was there any sign of necking. This indicates that despite this large level of displacement, the rockbolt was still able to provide support. Acknowledgement The first author would like to thank the China Scholarship Council (CSC) and UNSW Australia’s School of Mining Engineering for supporting her study in Australia. Special thanks are due to Mr Kanchana Gamage and Mr Mark Whelan for their help in conducting the tests.
L. Li et al. / Tunnelling and Underground Space Technology 56 (2016) 168–175
References Aziz, N., Hawker, R., Mirza, A., Nemcik, J., Li, X., Rasekh, H., 2015. Strength characteristics of secura hollow groutable cable bolts. In: Proceedings The 2015 Coal Operators’s Conference, Wollongong, Australia. Aziz, N., Pratt, D., Williams, R., 2003. Double Shear Testing of Bolts. Proceedings Coal Operators’ Conference 2003, Wollongong, NSW, Australia, pp. 154–161. Craig, P., Aziz, N., 2010. Shear Testing of 28 mm Hollow Strand ’TG’ Cable Bolt. Proceedings Coal Operators’ Conference 2010, pp. 171–179. Darling, P., 2011. In: SME Mining Engineering Handbook. SME. Egger, P., Zabuski, L., 1991. Behaviour of rough bolted joints in direct shear tests. In: Proceedings 7th ISRM Congress, Aachen, Germany, pp. 1285–1288. Ferrero, A., 1995. The shear strength of reinforced rock joints. Int. J. Rock Mech. Mining Sci. Geomech. Abstracts 32 (6), 595–605. Grasselli, G., 2005. 3D Behaviour of bolted rock joints: experimental and numerical study. Int. J. Rock Mech. Min. Sci. 42 (1), 13–24. Haas, C.J., 1976. Shear resistance of rock bolts. Soc. Min. Eng. 260, 32–41. Haas, J., 1981. Analysis of rock bolting to prevent shear movement in fractured ground. Min. Eng. 33 (6), 698–704. Hagan, P., Weckert, S., 2004. Anchorage and Failure Mechanisms of Fully Encapsulated Rockbolts – Stage 2. UNSW Mining Research Centre, Sydney, Australia. Haile, A.T., 1999. Observation of the dynamic support performance of South African tunnel support systems. In: Proceedings Rock Support and Reinforcement Practice in Mining, Kalgoorlie, Australia, pp. 335–341. Hartman, W., Hebblewhite, B., 2003. Understanding the Performance of Pock Reinforcement Elements Under Shear Loading Through Laboratory Testing—A 30-Year History. Proceedings 1 St AGCM Conference, Sydney, Australia. Hoek, E., Kaiser, P.K., Bawden, W.F., 2000. In: Support of Underground Excavations in Hard Rock. Holmberg, M., 1991. The Mechanical Behaviour of Untensioned Grouted Rock Bolts. Department of Soil and Rock Mechanics. Hyett, A., Spearing, S., 2013. Enlightening Bolts: Using Distributed Optical Sensing to Measure the Strain Profile Along Fully Grouted Rock Bolts. Proceedings 32nd International Conference on Ground Control in Mining. Jalalifar, H., 2006. A New Approach in Determining the Load Transfer Mechanism in Fully Grouted Bolts PhD thesis (unpublished). University Of Wollongong, NSW, Australia. Jalalifar, H., Aziz, N., 2010. Analytical behaviour of bolt–joint intersection under lateral loading conditions. Rock Mech. Rock Eng. 43 (1), 89–94. Li, C., 2010. Field observations of rock bolts in high stress rock masses. Rock Mech. Rock Eng. 43 (4), 491–496. Li, L., Hagan, P., Saydam, S., 2014a. A review of ground support systems performance subjected to dynamic loading. Proceedings Eurock 2014, Vigo, Spain. CRC Press, pp. 1063–1068.
175
Li, L., Hagan, P., Saydam, S., 2014b. Tensile stress mobilization along a rockbolt under shear load. Proceedings International Young Scholars’ Symposium On Rock Mechanics, Xi’an, China. Mahony, L., Hagan, P., 2006. A laboratory facility to study the behaviour of reinforced elements subjected to shear. Proceedings Coal Operators’ Conference 2006, Wollongong, NSW, Australia, pp. 245–254. Malmgren, L., 2007. Strength, ductility and stiffness of fibre-reinforced shotcrete. Mag. Concr. Res. [Online] 59. Available: