Introduction
Rockbolts are a key element in the support of underground excavations. In recent years, fully encapsulated rockbolts have gained widespread use especially in the coal mining industry. The core argument that has favoured its use is the potential ability of fully encapsulated rockbolts to provide “…maximum support capacity at the point where bed separation could occur. (Moreover) this mechanism could operate simultaneously at many points along the rockbolt” (Gray, Hunt and Fabjanczyk, 1998).
In the context of ground control, a rockbolt provides support through its ability as a reinforcing member to enhance the material characteristics of rock strata, in particular by increasing the stiffness or the ability of the rock strata to resist movement when subjected to loading.
For this to occur, the rockbolt has to have superior material property characteristics in terms of strength and stiffness to share the load acting on the rock strata. And, equally important, effectiveness is dependent on the bonding between rock strata and rockbolt to allow the load transfer to occur. In the case of fully encapsulated rockbolts, a physical bond is created between the rock and rockbolt using a resin or grout along the entire length of the rockbolt. The stronger the coupling between the rock and rockbolt, the more effective the support provided.
Factors, which influence this bond, include the surface properties of the rockbolt (geometry and rib pattern and, surface condition), material properties of the resin and the rock surface characteristics (surface roughness, straightness, voids etc) (Hagan, 2003).
This paper deals with differences in the load transfer phenomena between rock strata and the rockbolt.
Significance of Load Transfer
After installation, the rockbolt initially provides passive support. Under the action of some load, either external or deadweight, strata separation can occur. As the rockbolt is embedded in the strata, it tends to resist the movement or displacement between strata due to its higher stiffness or modulus of elasticity and a load is induced in the rockbolt through shear action in the grout along the contact surfaces at the point of strata separation as illustrated in Figure 1. The load induced in the rockbolt is found to dissipate with distance along the rockbolt as the load is transferred into the surrounding rock mass.
The load transfer capacity is a term equated to the effectiveness of the rockbolt in providing support to the rock strata. Serous and Signor (1987) defined it as the change in load with respect to distance along the rockbolt. Gray, Hunt and Fabjanczyk (1998) defined it in terms of the maximum stress generated per unit area of the rockbolt. More effective support systems are characterised by high load transfer capacity with high loads generated at small displacements.
Figure 1. Load transfer between rock and rockbolt via the grout (after SCT).
The support system will fail when either the shear or yield strength of the individual components of the system is exceeded by the shear, compressive or tensile stresses generated with bed separation or, when the shear strength is exceeded between the rock/grout or grout/rockbolt interface.
Non-linear load transfer
The nature of the load distribution has been examined by Farmer (1975) who developed a model that indicated an exponential decay with distance along the rockbolt. Subsequent tests with varying applied loads and over two anchorage lengths verified the model. There was reasonable agreement with the model up to a given load at which point it was postulated that de-bonding occurred along the interfaces.
Nietzsche and Haas (1976) developed a model based on finite element methods and assumed total symmetry in the stress distribution. Their model again indicated a non-linear reduction in load with distance; however, it was also found that most of the load was dissipated within 400 mm from the free surface with little confining pressure generated at the far end of the rockbolt.
Linear load transfer
Serbousek and Signor (1987) developed a finite element model that indicated a non-linear reduction. The model also indicated that the rate of load reduction increased with both grout modulus and rock modulus; the stiffer the grout, the greater the rate of load transfer. Subsequent laboratory tests revealed results that in some cases appeared to conflict with the model. First, the rate of load reduction appeared to be linear at least for the majority of the load transfer. Second, the rate of load reduction increased with applied load. Third, they concluded that length of the rockbolt had little effect on the load transfer characteristics as nearly all the load was dissipated within the first 550 mm as shown in Figure 2. Finally, grout type (resin and gypsum) and hole diameter were found to have little effect on the rate of load transfer.
Radcliffe and Stateham (1980) conducted a field study using 50 strain-gauged bolts in bedded roof strata. They found a linear function for load transfer over the full length of a rockbolt. Similar results based on field studies have been reported by Strata Control Technologies, SCT (Fabjanczyk and Tarrant, 1992).
Figure 2. Load distribution for rockbolts of differing length
(after Serbousek and Signer, 1987).
Bed separation vs. face loading
Two models concerning the nature of load transfer with a fully encapsulated rockbolt have evolved over the past thirty years. One accounts for non-linear load transfer observed in pull tests undertaken in the laboratory and field. An alternate model accounts for linear load transfer also observed in field studies but where load transfer was initiated through bed separation.
Whitaker (1998) accounted for the two models as being due to differences in the method of loading the rockbolt. In a conventional pull test, an axial tensile load is applied at the free end of the grouted rockbolt usually using a hydraulic cylinder. At the same time, the resultant reactive force of the hydraulic cylinder induces a compressive load as it is made to press against the surface of surrounding rock. This arrangement is illustrated in Figure 3.
Figure 3. Arrangement for loading a
rockbolt in a pull test
(after Galvin et al 2001).
It is postulated that the more likely mechanism of loading a rockbolt in the field is due to bed separation with the rockbolt being drawn in opposite directions. This is illustrated in Figure 4 and shows the absence of any compressive load on the rock surface. In such a case, the displacement of the parting may be sufficiently small so that it is within the elastic limit of the grout and rock thereby minimising any resistance to subsequent closure of the parting. Alternatively, displacement exceeds the elastic limit resulting in slippage or breakage of the bonds between the rockbolt/grout or grout/rock strata or both. Such a situation would then tend to react against any closure of the parting.
Numerical modelling by Whitaker found some difference in load distribution along the rockbolt and stress distribution in the surrounding rock strata between the two loading arrangements.
Test Program
A test program was devised to investigate under controlled laboratory conditions, the modelling by Whitaker with the objective of determining what effect the method of loading a rockbolt had on the nature of load transfer for a fully encapsulated rockbolt (Hagan and Weckert, 2002).
A strain-gauged rockbolt was used to measure the variation in load along the length of the rockbolt. The rockbolt was a standard though shortened version of instrumented rockbolts supplied by SCT for use in field studies. Four pairs of strain gauges were arranged along the length of the rockbolt that lay within the grouted section (approximately 175 mm) of the rockbolt. A further three pairs of strain gauges lay outside the hole but within the loaded section of the rockbolt. The instrumented rockbolt had been calibrated in an Avery Universal test machine.
The test rock used in the program was made from a mixture of cementitous grout, sand and water that was poured into 145 mm diameter cylindrical poly pipe moulds and left to cure for 28 days. A borehole diameter of 26 mm was drilled in the test rock. A specially formulated mix-and-pour resin for the test program was used to bond the rockbolt in the borehole.
The test rock was contained within a biaxial load cell and a 10 MPa confining load applied during each test.
Figure 4. Application of load in a rockbolt due to bed separation (after Galvin et al 2001).
Loading arrangements
Two methods were used to apply load to the instrumented rockbolt.
The first method was intended to replicate a conventional pull test whereby a hydraulic cylinder is used to effectively extrude the rockbolt from the rock. As fluid is forced under pressure into the cylinder, a tensile load is induced in the rockbolt. The cylinder in turn reacts against the rock face. Varying levels of forces were applied by the cylinder. At each force increment, the change in strain at each of the seven pairs of strain gauges was recorded. The arrangement of equipment used in this test is shown in Figure 5.
Figure 5. Arrangement of rockbolt, hydraulic cylinder and biaxial load cell containing the rock sample as used in the test program.
The second arrangement was intended to mirror the loading condition of a rockbolt on bed separation. In this arrangement, the reactive force from the cylinder was transferred to the opposite end of the rock sample through the wall of a biaxial cell via two thick steel plates, one placed at either end of the cell. The rock core sample was fixed to the thick plate with three dynabolts. This arrangement is illustrated schematically in Figure 6. In this arrangement, no load was caused to act on the surface surrounding the rockbolt. Instead a tensile stress regime was created at the far end of the rock sample.
Figure 6. Schematic of the non-confinement
loading arrangement.
Results
Three replications of the pull test arrangement were undertaken with slight changes made between each test replication. The maximum force applied by the hydraulic cylinder to the strain-gauged rockbolt in this arrangement was 40 kN. The variation in load as indicated by the strain gauges along the rockbolt is shown in Figure 7.
Figure 7. Variation in load along a rockbolt in the conventional pull test arrangement.
The variation in load with the bed separation arrangement is shown in Figure 8. The load distribution was observed at five levels up to 60 kN. Unfortunately as the force was being increased to 80 kN, the rock sample failed along a plane parallel to the end point of the three dynabolt anchors.
Figure 8. Variation in load along a rockbolt in the bed separation arrangement.
An examination of the results presented in the two graphs showed a number of similarities.
· Load was found to reduce with distance along the rockbolt from the free surface.
· The rate of load reduction varied with the level of applied force on the rockbolt.
· Given the relatively short encapsulation length of 175 mm, extrapolation of the results indicated that at each level of applied force on the rockbolt, the load tended to zero at the end of the rockbolt.
As well there were some differences exhibited in the results.
· In the case of the pull test arrangement, there appeared to be an initial slight increase in load on the rockbolt near to the free surface that was above the force applied to the rock bolt. This is likely to be due to the compressive stress induced in the rock by the hydraulic cylinder. This tended to be more pronounced at the lower load levels and was found to extend up to 50mm into the rock. As the level of applied force was increased, this became less significant.
· In the bed separation arrangement, the opposite behaviour was observed with a slight reduction in load just below the surface of the rock. This was noticeable at all forces levels.
· The most significant difference, however, lay in the nature of load transfer.
§ In the pull out test arrangement, it was apparent that the rate of load transfer was non-linear though the limited number of measurement points along the rockbolt made it difficult to define the relation. Similar behaviour occurred at each level of applied force.
§ In the bed separation arrangement, load transfer appeared to follow a linear reduction with distance.
§ In both cases there appeared to be an edge effect with a change in the load transfer relation near to the surface.
Conclusion
Two methods of applying an external tensile force to a fully encapsulated rockbolt were examined in terms of the impact on the load transfer behaviour between a rockbolt and rock. One method replicated loading of a rockbolt in the pull out test. The second method replicated loading induced in the rockbolt as a result of bed separation.
It was found that load transfer occurred over the full length of encapsulation independent of the level of force applied and method of application; at least for an encapsulation length up to 175 mm. Previous research indicated that load transfer could be limited to less than the full length of the rockbolt. Further testing will be required to examine whether this observation holds for longer lengths of encapsulation, preferably with encapsulation lengths in excess of 400 mm.
There was a marked difference between the two methods in terms of the nature of the load transfer. In a pull out test arrangement, load transfer was observed to decrease in a non-linear fashion with distance. Whereas in the bed separation arrangement, load transfer decreased in a linear fashion. Further testing is required with closer spacing of strain gauges on an instrumented rockbolt to better define the nature of the relation in both cases.
As the rate of load reduction was much greater in the pull out test arrangement than in the bed separation arrangement, the former would be considered to be more effective in terms of load transfer and enhancing ground support. It is also suggested that caution is exercised when interpreting results based on the pull out test as it would tend to overestimate the level of support that would actually be provided in supporting rock through load transfer and confinement.
Acknowledgements
The author wishes to thank the support provided by the Australian Coal Association Research Council (ACARP), which provided funding for the project and Celtite Pty Ltd, which provided test materials. The author also acknowledges the contributions made by Alison Whitaker, Glenn Dawson and Steven Weckert to the rockbolting research program at the UNSW Mining Research Centre.
References
Fabjanczyk, M and Tarrant, G C, 1992. Load transfer mechanisms in reinforcing tendons, in Proceedings 11th International Conference on Ground Control in Mining, pp 1-8 (The Australasian Institute of Mining and Metallurgy: Melbourne).
Farmer, I W, 1975. Stress distribution around resin-grouted rock anchor. International Journal of Rock Mechanics, Mining Science and Geomechanics Abstracts, 12:347-351.
Galvin, J M, Offner, J C, Whitaker, A, Fabjanczyk, M and Watson, J O, 2001. Establishing anchorage and failure mechanisms of fully encapsulated roof support systems. ACARP Project C7018 (UNSW Mining Research Centre).
Gray, P, Hunt, N and Fabjanczyk, M W, 1998. New developments in ground support with particular reference to high capacity, high load transfer rock bolts, in Proceedings International Conference on Geomechanics/Ground Control in Mining and Underground Construction, pp 513-524 (University of Wollongong),
Hagan, P C, 2003. A study of the variability in some rockbolt parameters and their potential impact on anchorage performance in coal mines, in Proceedings of Conference on Effective Risk Management for Mining Project Optimisation, (The Australasian Institute of Mining and Metallurgy: Melbourne).
Hagan, P C and Weckert, S, 2002. Anchorage and failure mechanisms of fully encapsulated rockbolts – Interim Report, ACARP Project C10022, April (UNSW Mining Research Centre).
Nitzche, R N and Haas, C J, 1976. Installation induced stresses for roof bolts. International Journal of Rock Mechanics, Mining Science and Geomechanics Abstracts, 13:17-24.
Patrick, W C and Haas, C J, 1980. Strata separations and loads on grouted bolts in coal mine roofs, in Proceedings 21st Symposium on Rock Mechanics, pp757-768.
Radcliffe, D E and Stateham, R E, 1980. Stress Distribution Around Resin Grouted Bolts, USBM RI8440, (USA Dept of Interior).
SCT, Advanced Course in Roof Bolting – Load Transfer. Strata Control Technology.
Serbousek, M O and Signer, S P, 1987. Linear Load-Transfer Mechanics of Fully Grouted Roof Bolts, USBM RI9135 (USA Dept of Interior).
Whitaker, A, 1998. Critical Assessment of Past Research into Rock Bolt Anchorage Mechanisms. Unpublished BE (Mining Engineering) thesis, University of New South Wales, Sydney.
Original version of manuscript published in Proceedings of 1st Australasian Ground Control in Mining Conference, November, 2003 (UNSW: Sydney).