P.C.Pinzzi,A.J.S.(Sm)Spering,K.V.Jessu,P.Singh,R.Hwker

a Department of Mining and Metallurgical Engineering,Western Australian School of Mines,Curtin University,Kalgoorlie,WA 6430,Australia

b Minova Global,Nowra,NSW 2541,Australia

Keywords:Rock bolt Combined load Gap test Failure mechanism Axial load Shear load

ABSTRACT Rock bolts are subjected to different loading conditions along their lengths such as axial,bending,and/or shear forces,which can cause failure at lower loads than those considered for design purposes.The common existing methodologies do not consider the actual loading of the rock bolts and assume it is only pure axial or pure shear.This study was conducted to investigate the un-grouted rock bolt performance under combined load conditions.Two loading regimes were evaluated:the effect of initial shear displacement on axial load capacity and displacement,and the effect of axial displacement on the shear load capacity.The first regime was also conducted for shear with a gap,when there is a spacing between the shear interfaces.The results of this study showed that the rock bolt can resist higher axial loads than shear under pure or combined load conditions.Under combined load conditions,the rock bolt capacity decreased significantly for both regimes.However,when applying the shear load with a gap,the rock bolt load capacity was not affected significantly.Also,the total bar deformation was improved for shear and axial.The findings of this study show the need to improve the rock bolt design considering the complex loading conditions in situ with/without a gap.

1.Introduction

Rock bolts are used worldwide to reinforce potentially unstable rocks around excavations.The excavation reinforcement and support system plays an important role in mine safety since it is directly correlated to the fall of ground.Potvin and Nedin conducted a study that pointed out the risks and causes of rock fall focused during 1998-2003 period;according to their study,mining areas away from the working face are considered as ‘‘low risk”of injury[1].However,when analyzing the fatalities,the data showed that most of them happened away from the faces under reinforced and supported grounds.In this case,the causes were pointed out to be due to,in order of relevance,corrosion,bolts too short,broken bolts,incorrect installation and bolt spacing too wide.In 2015,the fall of ground was the major cause of fatalities in mines around the world;it counted for 29% of the International Council on Mining and Metals report’s statistical samples of 2016 [2].These facts show the need for improvements of the conventional rock bolt design methodologies.Even though there are many methodologies used for the reinforcement design,none of them are considered to be accurate[3].It is still a challenge to understand the complexity of the load in situ,mainly to replicate these loads on numerical modeling or laboratory analysis.Although there were studies conducted to understand the rock bolt behaviour under axial and shear load,only a few attempts were made to investigate the effect of complex loading conditions(i.e.shear,axial,and bending)on the rock bolt performance.

The loading conditions in situ can vary from mine to mine,and they require different reinforcement mechanisms.Tunnel support requirements depend on:the stress regime around the tunnel,stresses change during its lifetime,the shape and size of excavation and geological features[4].For each existing reinforcement mechanism(i.e.the key block,beam building,and suspension),the load conditions on the support will change and the occurrence of combined load conditions can be observed in these mentioned basic support conditions [3].Fig.1a represents the key block loading mechanism,which the rock bolts hold the dead-weight generated by the block.Depending on the rock bolt’s location (i.e.roof or wall)and the distribution and orientation of discontinuities,the loading acting on the rock bolt will change.The beam building mechanism is used when the roof is bedded,so the rock bolts will tie the beams together to form one ‘‘laminated”beam (Fig.1b).In the rock bolt located at the center of the excavation,the load is typically only axial,but the other rock bolts are subjected to a combination of axial,shear and bending.For the suspension mechanism,the loading mechanism is similar to the beam building mechanism;however,in this case,weak rock beds are suspended from a thicker and stronger rock layer(Fig.1c).Therefore,the loads that the rock bolt undergoes vary from each in-situ condition.

Fig.1.Support mechanisms (after [3]).

Studies have been conducted to understand the rock bolt response of each possible load condition.Freeman was the first to investigate axial loadings along the rock bolt length using strain gauges [5].Numerous studies were also conducted to understand the effect of axial load and also shear load on the rock bolt performance [6-14].Most of these studies were conducted to evaluate the mechanical properties of rock bolts,grout and installation procedure.Recent studies focused on the contribution of bolts to the joints reinforcement.In this case,the combined load is a result of shear and bending/axial load.In this paper,however,the combined load had two sources of loads:in the shear tests with a gap(Fig.2),when shear load is applied and axial load is also generated as a result of bending moment and as a result of more than one load being applied to the rock bolt.

1.1.Shear loads at discontinuities

In situ observations in hard rock have pointed out the rock mass condition and discontinuities characteristics in shallow and deep mines.Li observed that at deep conditions,the rock mass presented a higher quality because the induced stresses are higher[15].Thus,the discontinuities were closed.At shallow/intermediate conditions,the discontinuities were open (i.e.gap)because the stresses conditions are less.Therefore,the shear load with open discontinuities/gap is a common condition in shallow,hard rock mines.

The gap conditions can also occur in rock with no spacing between the discontinuities,in both shallow and deep mines,when the rock strength is low,as in coal mines.In this condition,as the rock bolt deforms,the rock at the discontinuity interface will break generating a spacing/gap [11].Although this condition has been identified in situ,this load condition is not well understood,and there are only a few studies in the literatures that consider the effect of the gap on the rock bolt performance [14,16].

The effect of a gap between discontinuities was first considered by Stimpson to evaluate the rock bolt performance;his study was conducted to develop an analytical method to better understand the rock bolt performance considering the bolt inclination,hole and bolt diameter,grout,Young’s modulus of rock and spacing of the discontinuities[16].However,it did not focus on the gap effect on the rock bolt capacity.Snell et al.conducted shear tests to evaluate the performance of fully grouted bolts with different gap sizes between the joints;it was found that as the gap size increases,the shear ultimate load capacity of the rock bolt decreases,and its resistance to displacement increases [14].Even though in shear tests with a gap the only load applied is shear,the rock bolt fails by a combination of shear and axial loads.The gap allows the bar to bend before failure and the bending moment results in an axial force.

Fig.2.Shear tests with a gap (between the shear platens).

Pellet and Egger applied the beam theory to understand the shear behaviour of the rock bolt across discontinuities and determined that the two endpoints of the gap (points A and O in Fig.2)influence the failure mechanism of the rock bolt [8].In this study,shear load was applied at point O,where the rock bolt failed under shear and axial.In this case,point A undergoes a maximum bending moment.According to the above-mentioned studies,when the rock bolt reaches the plastic stage,the bending stiffness drops and the bolt behaves as a truss,and only axial forces grow.This justifies why the rock bolt failed by a combination of shear and axial loads.

1.2.Combined load

In situ observations have shown the effect of combined load on the rock bolt performance as identified in the basic support mechanisms (i.e.key block,beam building,and suspension).Li conducted in situ observations of ground support in high-stress conditions at advance faces of cut-and-fill mining;his findings evidenced a combination of load mechanisms in situ as presented in Fig.3a and b,where it is possible to identify that both rock bolts in Fig.3a and b were subjected to axial,bending and shear before failure [17].

Few studies have been conducted to understand the effect of combined load on the rock bolt mechanical properties in which shear and axial loads are applied at the same time to the rock bolt with/without a gap [18].The purpose of this study was to understand the complexity of the failure mechanism and the rock bolt response to combined load conditions,at an early stage,and progress to be able to suggest improvements in the rock bolt design considering the existing complex loading conditions in situ.

2.Materials and methods

Tests were conducted to evaluate the rock bolt steel performance under combined load.The first set of tests were conducted using single loading conditions:pure tensile and pure shear.The pure conditions were used as a reference for applying complex load conditions as shear tests with a gap and combined load conditions.In total,131 tests were conducted using three different steel types(Table 1):Australian (AUS)R27,South African (SA)R27 and South African (SA)M24.

Fig.4 shows the Instron machine used for all the tests at the Minova facility in Nowra,New South Wales,Australia.The rock bolt was held by the two sets of jaws and tested to destruction.Throughout all the tests,the movement of the hydraulic platen was set at 5 mm/min unless otherwise stated.

Fig.3.In situ combined load conditions (after [17]).

Table 1 Rock bolt types and specifications.

Fig.4.Tensile test machine.

2.1.Tensile tests

Tensile tests were performed to evaluate the properties and behaviours of a material under a uniaxial load condition.In these tests,tensile loads against displacement were recorded.

2.2.Shear tests

Pure shear/guillotine tests and shear tests with a gap were conducted using the apparatus shown in Figs.5 and 6.The shear rig test is composed by a fixed side and a removable side.In the internal part,it is composed of removable pipes(Fig.5a),where the rock bolt(Fig.5b)is placed.For the guillotine tests,the rock bolt was inserted into the steel pipes,and a load was applied in one side of the steel pipe until the rock bolt failure.In the tests conducted with a gap,the rock bolt was inserted into the steel rig,and the removable steel pipes inside the rig were adjusted to the spacing required for the tests (5,10,15 and 20 mm),as represented in the Fig.6a and b.For the gap tests,a nut and a plate were hand tighten on each side to avoid the bolt movement as the load was applied.

2.3.Combined load tests

Fig.5.Schematic steel rig setup used for shear tests.

Fig.6.Shear test with a gap schematic figure and the actual setup.

A new laboratory test was designed to evaluate the effect of combined load on the rock bolt performance.Two regimes were conducted to evaluate the effect of shear displacement on axial load capacity and the effect of axial displacement on shear load capacity.

The Instron machine was used to apply the shear load to the rock bolt placed inside the shear test rig as shown in Fig.7.The tensile load was applied to the bar using a 6×104kg capacity Enerpac hydraulic cylinder(RCH 606).Fig.8 represents each part of the test set up and also the position of the hydraulic cylinder.For pure shear tests,the rock bolts were not tightened.Under combined load,AUS R27 and SA M24 steel rock bolts were tested.

Two regimes of combined load were conducted as follows.

2.3.1.Axial displacement effect on shear load

Axial displacement was applied to the rock bolt,using the hydraulic cylinder.The axial displacement was kept,and the shear load was applied to failure.The axial displacement applied to the bar varied according to the maximum displacement the bar can undergo under pure loading conditions.The percentages of displacements applied compared to the bar maximum displacement capacity are given in parentheses.Two steel types were evaluated on this regime:(1)for AUS R27,axial displacements applied were 20 mm (16%),30 mm (25%),40 mm (33%),45 mm (37%),50 mm(41%),60 mm (49%),70 mm (58%),80 mm (66%)and 100 mm(82%);and (2)for SA M24,axial displacements applied were 20 mm (12%),30 mm (17%),40 mm (23%),50 mm (26%),and 60 mm (33%).

2.3.2.Shear displacement effect on axial load

Shear displacement was applied to the rock bolt by the Instron machine.The load was kept to the bar,and the axial load was applied to failure,using the hydraulic cylinder.The percentages of the shear displacements applied compared to the maximum shear displacement the bar can undergo are given in parentheses.For this condition,only AUS R27 was used since SA M24 was failing at the threads.In addition to the tests with pure shear displacement,some tests were conducted with a gap.For AUS A27,shear displacements applied were 2 mm (20%),4 mm (39%),6 mm(59%),7.5 mm (74%),and 9 mm (88%).This method was also conducted for shear with a 5 or 10 mm gap.The shear displacements applied were 7 mm(65%)and 9 mm(86%)for 5 mm gap and 9 mm(50%)and 12.5 mm (69%)shear displacements for 10 mm gap.

Fig.7.Combined load test set up.

Fig.8.Schematic diagram of combined load tests set up.

2.4.Calibration of hydraulic cylinder

Two different types of equipments were used to apply axial load to the rock bolt:Instron machine and Enerpac hydraulic cylinder using hand pump and foot pump.Thus,the hydraulic cylinder used for combined load tests was calibrated with the Instron machine for the accuracy.The calibration curve is shown in Fig.9.

3.Results and discussion

3.1.Tensile and shear (guillotine)tests

The first set of tests were conducted under only axial or only shear loading to benchmark the two extremes of the loading.These results were then used as a reference to compare rock bolt performance under combined load mechanisms(axial then shear to failure and shear then axial to failure).Table 2 shows the mechanical properties of each rock bolt type.

Based on the ultimate load,the ratio of shear load to tension was found to be:(1)0.73 for AUS R27 hard rock;(2)0.73 for SA R27 hard rock;(3)0.67 for SA M24 coal.

Figs.10 and 11 show that the axial load capacity of the rock bolt is superior to shear load capacity.However,the percentage in elongation is higher in shear.

Failure mechanisms in pure shear and pure tension are representd in Fig.12a and b,respectively.In Fig.12b,the necking before failure can be easily seen.

3.2.Rock bolt performance under combined load

3.2.1.Shear tests with a gap analysis

Gap shear tests were conducted to evaluate the rock bolt performance of two rock bolt types,AUS R27,and SA M24.26 tests were conducted in which 16 tests were on AUS R27,and 10 were on SA M24.The AUS R27 and SA M24 shear capacity was evaluated under different gap sizes,5,10,and 15 mm gap for both steels and 20 mm gap for AUS R27.The tests conducted with 20 mm gap were not repeated because as the rock bolt deformation increased for this gap size,the bar started to touch the steel rig at point A (Fig.2)and to shear at the contact point.

The load and displacement capacity were evaluated under shear load varying the gap size.The increase in gap size resulted in a decrease of shear capacity and an increase in total displacement(Fig.13a and b).The variance in load is not yet well understood.However,the increase in total displacement is caused by the failure mechanism,which is a combination of localized shear,axial,and bending,which is not presented in the pure shear tests.

Fig.9.Calibration curve between the Instron and the Enerpac.

Table 2 Properties of the three rock bolts steels tested.

Fig.10.Ultimate tensile strength for all the rock bolts tested.

Fig.11.Ultimate shear strength for all the rock bolts tested.

Fig.12.Pure tests.

In the shear tests with a gap,the displacement capacity was the most affected.The displacement needed to reach the maximum shear capacity of the rock bolt increased with gap size (Fig.13a and b).This displacement measured by the Instron unit is the total displacement (vertical displacement),which is the combination of the bending,axial,and shear as described by Pellet and Egger [8].The total displacement was used to calculate the combined displacement as shown in Fig.14.The combined displacement was defined by Pythagorean Theorem (Eq.(1)).From this analysis,the resultant force (combined load)angle (α)was calculated from Eq.(2).For pure shear,the bar breaks in 90° to the bar and for pure axial load,the bar break at 0° to the bar.Under combined load mechanism,the angle will vary between 0° and 90°.The total displacement,combined displacement,and α for each gap size are presented in Table 3.

where C is the combined displacement (axial,bending,and shear displacements);T the total displacement,given by the Instron;Gap the spacing between shear interfaces;and α the angle of combined load acting on the bolt.

Fig.13.Shear capacity under a gap.

Fig.14.Shear tests with gap loading mechanisms.

Table 3 Total,combined,shear and axial displacement for AUS R27 and SA M24 shear tests with a gap.

The combined load condition in the gap tests can be identified in the bar failure mechanism for AUS R27 from Fig.15b-e.When shear is applied with a gap,the rock bolt will bend before failure,and at the failure interface,the rock bolt will fail by shear and axial load.Moreover,as the gap increased the bending effect also increased.The same behaviour was seen on the SA M24 rock bolts failure.The M24 failures are represented in Fig.16a-d.For both steels,the effect of 5 mm gap was minimum.The failure mechanism at this gap size is similar to the shear failure in pure condition;however,it can be observed that there was less bending.For the other gap sizes,as the gap increases the effect of bending and shear also increases at the failed interface.

Due to the steel composition difference(i.e.content of C,P,Mn,Si,S,Ni,Cr,Mo,etc.)between SA M24 and AUS R27 (Figs.15 and 16),the failure characteristic was found to be slightly different.However,the total and combined displacement with the varied gap sizes were similar (Table 3).

For both steel types,AUS R27 and SA M24,the shear displacement at failure did not vary as the gap size increased.However,the total displacement increased because as the gap increased,the bar underwent bending and axial,compression and tension(on different sides of the bolt as it bends,Fig.14).The total displacement was found to be proportional to the gap size.From Table 3,it can be observed that the total displacement is almost twice the gap size.Although,the composition of both steels is different,and the nominal diameters are 23.5 and 22 mm for AUS R27 and SA M24,respectively,the combined displacement and resultant force angle showed to be similar for both steels.

Evidence of combined load acting on the bar has also been given by the angle α (Eq.(2)),which is the angle of the resultant forces acting on the rock bolt.The angle α varied from 59° to 65° for all gap size conditions.For same gap sizes,α showed to be similar even though the steels have different mechanical properties.This similarity shows that the effect of gap on the rock bolt performance depends on the gap size,and not as much on the rock the bolt properties.

These results found in the gap tests also illustrate the effect of the gap on the ratio of shear load to axial load.The AUS R27 shear load to axial load ratio dropped from 0.73 to 0.59(Fig.17a),and for SA M24,it dropped from 0.70 to 0.58 (Fig.17b)with the varying gap sizes.

From the anova statistical analysis,it was found that there is a significant difference in the ultimate loads between the guillotine shear test and gapped test.Thus,the gap had a significant influence on the rock bolt performance.Overall,the SA M24 performance was more affected by the gap in load comparing to the rock bolt AUS R27.

3.2.2.Combined load tests analysis

(1)Effect of shear displacement on the axial load and displacement performance for AUS R27 hard rock with/without a gap In total,29 tests were conducted to understand the influence of a shear displacement on the tensile strength.Fig.18 shows the tensile strength and displacement of the rock bolts when subjected to shear displacements of 0 mm,2 mm (20%),4 mm (39%),6 mm(59%),7.5 mm (74%),and 9 mm (88%).

Applying a shear displacement first from 2 to 6 mm had a minimal effect on the ultimate tensile capacity but it had a major effect on the displacement capacity of the bar.However,as the shear displacement increased,the tensile capacity and displacement were affected.Applying a shear displacement of 9 mm (88% of the ultimate shear displacement),the axial elongation decreased by 92%,which shows that the shearing of the rock bolt significantly decreases the ability to elongate axially.This behaviour is explained because when the bar is under shear,the load will concentrate at the weakest point,which is the shear interface for 9 mm shear displacement.Thus,when applying tensile load,the load will be distributed on half of the bar,from the source of axial load to the shear interface.As a consequence,the total bar displacement will be most affected as it is affected by the decrease in bar length under tension.

The failure mechanism for each scenario varied as shown in Fig.19a-e.At lower shear displacement,from 2 to 6 mm,the predominant failure mechanism was axial load,and the bar broke between the nut and the shear interface.When the shear displacement is increased to 7.5 mm,the major failure mechanism was tensile load,and the bar broke at the shear interface.At 6 mm shear displacement,the bar broke at the threads for the three tests performed with 1.5 m bar length.In this study,threads failure was considered to give erroneous results because at the threads the load capacity is decreased since the rock bolt diameter is decreased by the threading procedure.Increasing the shear displacement to 9 mm,the tensile performance of the bar decreased and the major load mechanism identified was shear.In this last scenario,the rock bolt underwent 88% of its total shear displacement before failure,which justifies the failure response.

13 tests were therefore conducted to try and understand the influence of shear with 5 and 10 mm gap on tensile strength for AUS R27.In the tests with a 5 mm gap,the shear displacements of 7.5 mm (69%)and 9 mm (86%)were applied and with 10 mm gap,9 mm (50%)and 12 mm (66%)shear displacements were applied,as the rock bolt can undergo more shear displacement with the increase in gap.Tensile load and displacement at failure are shown in Figs.20 and 21 for 5 and 10 mm gap tests,respectively.

The shear displacement over a gap had a minimal effect on the tensile capacity of the bolt,while the elongation needed to achieve the tensile capacity decreased significantly.For 5 mm gap,in one of the tests with 7.5 mm shear displacement,the bar broke at the threads and in the other tests it broke at the shear interface.For 10 mm gap,the bar broke at the shear interface for all the tests.Shear tests with the gap,in combined load condition,allows the bar to deform more before breaking.They resulted in failure at the threads at higher shear displacement (7.5 mm)compared to the condition without a gap (6.0 mm).

Fig.15.Failure mechanism of AUS R27 with and without a gap.

Fig.16.Failure mechanism of SA M24 with and without a gap.

Fig.17.Effect of gap on the correlation between tensile and shear load capacity.

Fig.18.Influence of shear load on tensile strength.

Fig.19.Failure mechanism on the combined load:applying shear displacement and then axial load.

Fig.20.Influence of shear load with 5 mm gap on tensile strength.

Fig.21.Influence of shear load with 10 mm gap on tensile strength.

To understand the effect of shear load with a gap on the axial capacity,the tensile capacity and the displacement were compared at 9 mm shear displacement with gap of 0,5 and 10 mm.The ultimate tensile capacity for pure tensile load condition was found to be 314 kN and total displacement 121.5 mm(1280 mm bar length)when using the hand pump.When using the foot pump,the tensile capacity was found to be 299 kN and displacement 133.3 mm(1297 mm bar length).These values were calibrated with those from the Instron machine.The tests conducted with 0 and 10 mm gap were conducted with the hand pump,and for 5 mm gap,the foot pump was used.Thus,the pure tests are the reference values for these findings.The findings for axial performance with 9 mm of shear displacement were:(a)with no gap:the tensile capacity dropped to 30%,and displacement decreased by 91%;(b)for 5 mm gap,the load capacity increased by 0.67% and decreased by 59% in displacement;(c)for 10 mm gap,it decreased in load capacity by 3% and displacement by 49%.

Fig.22.Combined load tests with 7.5 and 9 mm shear displacements with 5 mm gap.

Fig.23.Combined load tests with 9 and 12.5 mm shear displacements with 10 mm gap.

Applying a combined load with a gap means that two combined load mechanisms are acting on the rock bolt (Figs.22 and 23).When shear is applied with a gap as the first load,the rock bolt can undergo more axial deformation as well as bending compared to the guillotine condition.The initial gap length is deformed by bending,and as the shear deformation grows the bar will also deform by axial load.Therefore,the bar failure mechanism will be a combination of both shear and axial loads.For the same total displacement applied to the bar,the smallest the gap size is,the sooner the bar will start to deform axially compared to bigger gaps.It explains the difference of the axial displacement the bars underwent when applying axial load to failure and using the hydraulic cylinder after applying shear with a gap.Therefore,with increase in gap,the axial deformation of the rock bolt improved under combined load.

It might be an Italian or a Spaniard, remarked the clergyman.But to the fisherman s wife these nations seemed all the same, andshe consoled herself with the thought that the child was baptized as aChristian.

(2)Effect of axial displacement on the shear load and displacement performance for AUS R27 and SA M24 In total,47 tests were conducted under initial axial displacement to investigate the rock bolt shear performance for AUS R27 and SA M24.32 tests were conducted to understand the influence of axial on shear capacity of AUS R27.The axial displacements applied to the steel were 20 mm (16%),30 mm (25%),40 mm(33%),45 mm (37%),50 mm (41%),60 mm (49%),70 mm (58%),80 mm (66%),and 100 mm (82%).Fig.24 shows the effect of tension on shear capacity and displacement.There was a gradual decrease in shear capacity with the increase in the axial displacement while the shear displacement was found to be about 10%higher than the shear displacement at pure shear conditions.

Fig.24.Influence of axial load on shear strength for AUS R27.

The SA M24 was also evaluated for this combined load condition.15 tests were conducted using SA M24 to investigate its shear capacity under tension.The rock bolt was subjected to 20 mm(12%),30 mm (17%),40 mm (23%),50 mm (26%),and 60 mm(33%)axial displacement.The higher axial displacements were not tested as the rock bolt failed on threads at 80 mm axial displacement.The shear capacity decreased as the axial displacement increased from 20 to 60 mm axial displacement (Fig.25a).Both rock bolt types presented to have the shear capacity and displacement affected by axial displacement.

The rock bolts AUS R27 and SA M24 bars have presented similar behaviour under combined load (Figs.24 and 25).The load variation is a result of combined load acting on the bar,but more research is needed to understand it fully.The variation in displacement or its increase is a result of this combined load mechanism acting on both rock bolts type.Figs.26 and 27 present the failure mechanisms for both AUS R27 and SA M24.The failure mechanism suggests that the rock bolt underwent bending,and axial deformation as it was the load being applied before failing by shear load.

As a result of these findings,the load capacity and also displacement was affected by a combination of loads.Thus,the total displacement recorded by the Instron machine was a sum of the combined load(shear and axial/bending)being applied to the rock bolt.The increase in displacement by the Instron is explained by this combination of loads acting on the bar.Similarly to what happens when applying shear with a gap,before failing by shear the bar bends increasing the total displacement.Comparing AUS R27 and SA M24,the AUS bar deformed slightly more,when applying axial load then shear to failure,compared to the SA (Figs.24b and 25b).The total displacement increased:(a)for AUS R27,at maximum 1.1 mm compared to the displacement at pure shear,which is 10.2 mm;(b)for M24,at maximum 0.7 mm compared to its total capacity,which is 8.4 mm.

Fig.25.Influence of axial load on shear strength for SA M24.

Fig.26.Rockbolt surfaces after failures for AUS R27.

Fig.27.Rockbolt surfaces after failures for SA M24.

Fig.28.Effect of combined load on the correlation between shear and tensile load.

These results showed that both rock bolts underwent a similar failure mechanism,shear or axial/bending.However,on the SA M24,a structural failure was identified due to its difference in composition and mechanical properties as presented in the previous method.

Figs.26 and 27 show the rockbolt surfaces after failure.It can be noticed that at lower axial displacement,the failure mechanism is predominantly shear load,e.g.at 20 mm axial displacement.As the axial displacement applied increases,the bar underwent more axial load as can be seen at 100 and 60 mm axial displacement for AUS R27 and SA M24,respectively.

The effect of the axial displacement on the shear load has also affected the ratio of shear load to axial load(Fig.28).The main reason is that the tensile capacity is considered as the weakest link in the main rock bolt design methodologies.In fact,under this condition,for SA M24 the shear load capacity can almost reach half of the axial load capacity.

4.Rock bolt design

Under combined load conditions with a gap,the gap has also shown to have a significant influence on the rock bolt performance.In the combined regime,without a gap,load capacity and displacement decreased.Nevertheless,when applying shear with a gap and then axial load to failure,the rock bolt axial displacement increased as the gap increased.The findings of this study have suggested that in a gap and combined load conditions with a gap,which are common situations in situ,the rock bolt undergoes more deformation before failure.Then,when identifying these loads conditions,a rock bolt that can undergo more deformation should be considered in the support design.

These results show evidence of the need in creating a support design capable of considering the impact of shear as well as combined load conditions and conditions with a gap.It is necessary to implement the laboratory-based combined load studies into the reinforcement design to increase the safety and optimize the ground support design.

5.Conclusions

Rock bolts have shown to be subjected to complex loading conditions,e.g.combined load.The early stage of this study was developed to understand the complexity of failure mechanism of the rock bolt and its response to complex loading conditions.Laboratory tests were conducted using un-grouted rock bolts steels because they are useful to help understand the response of rock bolt under different loading conditions.Every combined load test performed showed an influence on the rock bolts performance differently.However,different steel types have shown to perform similarly under the same load conditions.

Based on the research and the lab tests to date,the following conclusions can be made:

(1)Current rock bolt support design methodologies consider only the axial capacity.

(2)For typical rock bolt steel in Australia,the shear capacity is about 70% of the axial (tensile)capacity.

(3)Most of the rock bolts are loaded along their length under varying combinations of axial,shear,and bending.

(4)Under both shear and combined load with a gap condition,a low-stiff bar is suggested in the design since it can undergo more displacement.

(5)Combined load conditions are still not considered in the rock bolt design;however,it could decrease the rock bolt shear and axial capacity by 30% and 18%,respectively.

To suggest improvements on the rock bolt design,further investigations will be undertaken to evaluate the effect of combined load on the system (i.e.rock bolt,grout,and cement block)with/without instrumentations.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors would like to thank Mining3,Minerals Research Institute of Western Australia,Curtin University and Peabody Energy for funding this research project.They also wish to thank Minova Global and its personnel who assisted in completing all the tests conducted at their facility in Nowra,NSW and for providing the rock bolts for testing.