Introduction
In most Australian underground coal mines, rockbolts form the basis of the primary roof support system. Previous studies have found however that despite an estimated industry-wide annual expenditure of over $A35 million on rockbolts, it is estimated that 30%-35% of the rockbolts do not perform to specification and may represent some risk to the maintenance of a safe workplace environment (Galvin et al 2001).
There are several factors contributing to the under-performance of rockbolts including installation procedures and the storage and handling of rockbolts, the effects of which can be mitigated through effective risk management practices.
In an effort to improve the performance of rockbolt systems, a
research facility has been established at the University of NSW (UNSW) with
funding provided by the Australian Coal Association Research Program (ACARP).
The objective of the research is to gain a better understanding of the load
transfer mechanisms of fully encapsulated rockbolts and identify the risk
factors that impact on the performance of rockbolt systems, thereby
contributing to an improvement in safety at mine sites.
Research undertaken using the facility has examined factors that may impact the anchorage performance or load bearing capacity of fully encapsulated rockbolt systems; the effect of bolt profile on load transfer; the nature of the load transfer between rockbolt, resin and rock; and, a study on the quality conformance of rockbolts at mine sites.
The objective of this paper is to review current understanding and knowledge on the anchorage performance of rockbolts combining this with the latest research findings to provide a foundation on which hazards may be identified and improvements made to workplace practices.
Risk Factors
The risk factors that can influence the performance of a rockbolt support system can be broadly classified into three groups – design, operation and quality control factors. Table 1 lists a number of risk factors related to each of these groups.
Many of the risk factors have a similar outcome in terms of a reduction in anchorage performance of rockbolts. The impacts of these risk factors and their underlying causes are discussed in the following section.
In addition to these factors, there are other variables that may also impact on rockbolt performance. Many of these lie outside the direct control and monitoring capabilities of personnel at the mine sites. Such factors include the metallurgical properties and deformation pattern of the rockbolt, the chemicals used in the manufacturing process of resin cartridges and consistency of these properties. Other measures are required to manage these risk factors by the suppliers such as quality control systems and independent auditing.
Resin Annulus
In general terms, the anchorage capacity of a rockbolt increases with rockbolt diameter – this holds true so long as the resin annulus or thickness of the resin between rockbolt and rock remains constant. With an increase in borehole diameter not only does the maximum load bearing capacity increase with rockbolt diameter but also the resistance to shear failure of the resin/rock interface with a larger rock surface area. Karabin and Debevec (1978) confirmed this general principal in pullout tests conducted with three different borehole sizes while maintaining a constant resin annulus, the results of which are shown in Figure 1.
Figure 1. Effect of borehole diameter on resistance to displacement of a rockbolt (after Karabin & Debevec, 1978)
For the case of relatively soft rock such as that associated with coal measures where the resin/rock interface is the weakest link and the length of encapsulation is limited, borehole diameter can be increased to achieve the required load capacity per unit length of the anchorage system. Snyder, Gerdeen and Viegelahn (1979) argued that increasing the borehole diameter must be accompanied by a commensurate increase in the diameter of the rockbolt as it would otherwise lead to an increase in resin thickness. Hence this would result in poor confinement of the resin leading to a reduction in the load coupling between the rockbolt, resin and rock.
Various researchers have noted the importance of minimising resin thickness. For example Franklin and Woodfield (1971) found when using a 19 mm rebar, a resin annulus of 6.4 mm resulted in the most rigid and strongest anchorage system. Durham (1973) suggested an optimum range of resin annulus of between 4 and 6 mm.
Figure 2. Effect of variation in resin annulus on rockbolt anchorage characteristic (after Hagan, 2003)
Work reported by Fabjanczyk and Tarrant (1992) on rockbolt push tests showed a marked reduction in load transfer performance of over 30% with an increase in borehole diameter from 27 mm to 29 mm when using a standard 22 mm rockbolt, that is with an increase in resin annulus of only 1 mm. They suggested the optimum borehole size being the smallest practical diameter taking account of bolt installation factors and resin viscosity.
Recent work by Hagan (2003) found in a laboratory pull test study that there was little significant variation in rockbolt behaviour with resin annulus sizes of 4 mm or less when using a standard 22 mm rockbolt as illustrated in Figure 2.
At an annulus size of 5 mm there was a near 25% reduction in the load bearing capacity of the rockbolt compared to that achieved with a 4 mm annulus. In addition there was a near 60% reduction in the resin/rockbolt stiffness. This is likely to have an adverse effect on the load transfer process between the rock mass and rockbolt with the rockbolt offering less resistance to the relative displacement of rock strata.
Similar findings have been made available by equipment suppliers. For example, Yeaby (1991) stated that “in essence encapsulation is reduced by 20% per millimetre of bit diameter” in terms of the reduction in rockbolt performance.
These results highlight the sensitivity of relatively small changes in resin annulus on the performance of a rockbolt. Hence it is important to control and monitor the dimensions of both the rockbolt and the borehole to ensure consistent performance is achieved. The optimum borehole diameter should be approximately 4 to 8 mm larger than the core diameter of the rockbolt. A larger borehole diameter is likely to result in a marked reduction in the load bearing capacity of the rockbolt. Alternatively, any smaller borehole diameter is likely to lead to poor distribution of resin along the length of the rockbolt due to a combination of small annulus clearance and resin viscosity.
Some of the actions required to address the various risks associated with the installation of rockbolts include:
· selecting the correct drill bit size for a given rockbolt diameter;
· measuring actual drilled diameter of the borehole;
· regular monitoring to ensure straight drill rods are used in drilling boreholes for rockbolts as bent drill rods will increase hole diameter;
· maintaining thrust force within the recommended design limits for the different rock types; and
· regular monitoring of the rockbolts to ensure they are within specification with respect to core diameter, rib height, bolt length and straightness.
Findings of Field Study
Variation in rockbolt dimensions
A study on the conformance of the physical dimensions of a rockbolt to manufacturer’s specification was undertaken by Hocking (2000) at seven coal mines in New South Wales. The purpose of the study was to gauge the extent of the variability of rockbolts supplied and used at the mine sites. The study examined several factors including core diameter and the rib height of a rockbolt as shown in Figure 3 as well as length and straightness of rockbolts and degree of surface corrosion.
Figure 3. Core diameter and height of deformation ribs
Sixteen batches of rockbolts were examined across seven coal mines with between 21 and 50 rockbolts examined in each batch. Usually two and in some cases three batches were examined at each site.
The study found a wide distribution in terms of the deviation of the average measured core diameter for each batch from the specified or nominal dimension for the rockbolt as illustrated in Figure 4. The core diameter was measured at 90° to the rolled area of the rockbolt at three locations along the length of the bolt; mid-length and 100 mm from the either end. The batches included rockbolts from three of the major rockbolt suppliers.
At four of the mine sites (designated as A, C, D and E in Figure 4) and in nine of the batches, the average measured core diameter was within 0.3 mm of specification. The deviation for the other seven batches was such that the average diameter was between 0.7 mm and 1.2 mm less then the specification.
Figure 4. Deviation of the average of sample measurements from the core diameter specified dimension
Interestingly the batches tended to be consistent in terms of the level of deviation at each mine site except at sites designated as B & F. This consistency may reflect the characteristics of products from the different manufacturers.
This finding emphasises the possible need for a quality control system to monitor conformance to specification of rockbolts from the suppliers. Interestingly if the guidelines of the American Society for Testing of Materials were applied, seven out of the sixteen batches or nearly 44% of the batches would not comply. Its guidelines for rock bolts (ASTM, 1995) state the core diameter should be to within 0.38 mm.
The study also found in some instances considerable variation within each batch of the measured core diameter. At five of the seven sites and in nine of the sixteen batches, the standard deviation on the mean diameter was calculated as ±0.15 mm, indicating a 95% confidence interval about the mean reading up to 0.3 mm. However at one site the standard deviation was as high as ±0.385 in a sample size of 20 rockbolts indicating a 95% confidence interval of 0.77 mm.
In terms of the height of ribs or the deformations along the length of the rockbolt, the study found the average height within each batch at the majority of mine sites was within 0.25 mm of specified dimension, ranging from 0.4 mm below to 0.2 mm above the specified height. However as Figure 5 illustrates, there was a bias towards the average rib height being under specification.
Figure 5. Deviation in the average of sample measurements from the rib height specified dimension
At two of the mine sites (F & G) the batches of rockbolts had both the average measured core diameter and rib height less than the specified dimensions.
Borehole roughness
It has been argued that as no chemical cross-linkage takes place between resin and rock, the resistance to any movement relies solely on the mechanical bond between the resin and rock. The strength of this bond therefore has a direct bearing on the anchorage performance and load transfer between the rockbolt and rock.
Work by Karabin and Debevec (1978) demonstrated that changes in drilling conditions influence the nature and size of interstitial sites formed along the surface of the borehole wherein resin can flow and set. A greater amount and extent of these interstitial sites would lead to an increase in the shear force required to break the bond between the resin and rock. This finding is illustrated in Figure 6 which compares the anchorage performance in terms of resistance to loading between two boreholes that were reamed smooth to 31.8 mm (1¼ in) and 34.9 mm (13/8 in) and a third borehole with a diameter of 25.5 mm (1 in) that had the surface roughened.
Figure 6. Effect on borehole surface roughness on the anchorage resistance
of an installed rockbolt
(after Karabin and Debevec, 1978).
Work by Gerdeen et al (1977) found that random grooving of a borehole increased anchorage performance three-fold when compared to smooth clean boreholes as shown in Table 2. “As drilled” boreholes were found to have twice the anchorage capacity as smooth, clean boreholes. Though their test work was incomplete, they found evidence of other factors such as the presence of water and the amount of cuttings left in the borehole also had a marked affect on anchorage capacity. Interestingly Snyder, Gerdeen and Viegelahn (1979) found that while there was a marked variation in test results under the same conditions, the variation reduced with increasing surface roughness that is the results became more consistent.
These results indicate that in terms of managing this risk factor on rockbolt performance and ensuring consistent levels of anchorage, it is important to maintain the optimum level of rotation speed and thrust during drilling for the rock conditions as these parameters effect the surface roughness of the borehole.
Rockbolt surface condition
While the design of the rib pattern, spacing and height of a rockbolt does impact the performance of a rockbolt, it is not a variable that is normally controlled and monitored at the mine site. There are other factors however that can impact on rockbolt performance which vary at the mine site and therefore require monitoring and control mechanisms.
The most significant factor concerning rockbolt surface is the
degree of surface corrosion. In a study by Cox and Fuller (1977) on the effects
of changes in surface finish of steel reinforcement in cement grout, they found
mill-finished indented wire offered greater resistance to displacement than plain
mill-finished wire. The raised surfaces surrounding the indents were seen as
enhancing the resistance to the relative movement between the wire and grout.
Cox and Fuller also observed that even greater resistance occurred compared to the indented wire when rusted wire was used as shown in Figure 7. But performance was slightly degraded with a combination of rust and indents in the wire.
Figure 7. Effect of surface rust and indents on load resistance of wire in grout (after Cox and Fuller 1977)
Fabjanczyk, Hurt and Hindmarsh (1998) found similar results when testing rock bolts with rusted bolts compared to clean, non-corroded surface (Figure 8). The rusted bolts at least initially offer greater stiffness at low displacements. In some ways this is a cosmetic effect since once any reasonable movement in rock strata occurs then both rusted and smooth rockbolts tend to behave in a similar manner.
Figure 8. Comparison of the load transfer characteristic between a clean
and rusted bolt
(after Fabjanczyk, Hurt and Hindmarsh, 1998)
The study by Hocking also examined the extent of surface corrosion on rockbolts. The study used a qualitative scale factor of one through eight as shown in Table 3. The full spectrum was observed in the average level of surface corrosion for each batch of rockbolts examined across the mine sites as shown in Figure 9.
Table 3
Qualitative scale used to assess rockbolt surface corrosion
|
1. |
No visible signs of corrosion along the surface of the rockbolt |
|
2. |
Mild level of corrosion occurring on less than 5% of the total surface area. The corrosion spots are usually a light bronze colour. Corrosion spots less than 3 mm in diameter |
|
3. |
Mild level of corrosion of between 5% and 20% of total surface area. Corrosion spots are red or brown in colour and not more than 10 mm in diameter |
|
4. |
Moderate corrosion over 20% to 50% of the total surface area. Corrosion spots less than 25 mm in length. Can also occur as a short single narrow line along the axis of the rockbolt |
|
5. |
Moderate corrosion over 50% to 80% of the total surface area. Corrosion spots less than 50 mm in length and also seen as a single narrow line plus other corrosion spots |
|
6. |
Moderate corrosion over the entire surface area though some parts of metal still visible |
|
7. |
High level of corrosion over the entire surface area. Corrosion is red or brown in colour. The deform ribs are discernable from a distance of 2 m |
|
8. |
Severe corrosion with platelets formed over the surface. The deform ribs are not discernable from a distance of 2 m |
Figure 9. Variation in the average level of surface corrosion observed in batches between mine sites
Corrosion appeared to be less affected by proximity to the sea than to the level of protective covering such as plastic wrapping and grease, position in the bundled batch of rockbolts, exposure to mine water and time spent in storage. Rockbolts which were covered in grease at one mine site were observed to show little surface corrosion. Bundles covered in plastic similarly showed lower levels of corrosion. Also surface corrosion appeared to be lower for those rockbolts within the tied bundles rather than on the outside edge of the bundle.
Finally as would be expected there was a correlation between the level of surface corrosion and duration in storage as shown in Figure 10. Given the wide variation in the level of surface corrosion that was observed across the seven mine sites, it is recommended that the effect on anchorage performance should be quantified.
Figure 10. Variation in average level of surface corrosion in batches with the length of storage
Conclusion
With reference to the risks associated with maintaining the stability of underground excavations, certain hazards have been identified regarding the use of fully encapsulated rock bolts. An increase in resin annulus from 4 mm thickness to 5 mm has been shown to result in a near 25 per cent reduction in the load bearing capacity of a rockbolt. There are several types of hazards that need to be controlled in order to ensure optimum performance of the rockbolt at the operational level such as consistently in achieving the designed diameter and length of borehole diameter and, the quality of rockbolts with respect to meeting specifications on core diameter and height of deform ribs.
A field study on the variability in the quality of rockbolts across seven mine sites found that 43% of the bundles tested had an average core diameter less than the specification by 0.7 mm or greater. The deviation from specification for the other 57% of bundles was less than 0.3 mm. The deviation in rib thickness was on the whole less significant although for 81% of the bundles, the rib thickness was less than specification. The study also found a large degree of variation in the level of observed surface corrosion on the rockbolts reflecting differences in storage conditions and practices between the mine sites.
The risks can be reduced by implementing quality systems that incorporate standard procedures and regular auditing of the quality of materials supplied, storage and handling and, the process of rockbolt installation so that the intended performance of fully encapsulated rockbolts in ground support is consistently achieved.
Acknowledgements
The author wishes to acknowledge the various contributions made by Rodney Hocking, Daniel Peel, Steven Weckert and Alison Whitaker to the research on rockbolting at the UNSW Mining Research Centre.
References
ASTM 1995. Standard Specifications for Roof and Rock Bolts and Accessories, F432 (American Society for Testing Materials).
Cox, R H T and Fuller, P G, 1977. Load transfer behaviour between steel reinforcement and cement based grout, CSIRO Division of Applied Geomechanics, Technical Rpt No 40.
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).
Fabjanczyk, M, Hurt, K and Hindmarsh, D, 1998. Optimisation of roof bolt performance, in Proceedings International Conference on Geomechanics/Ground Control in Mining and Underground Construction, pp 413-424 (The Australasian Institute of Mining and Metallurgy: Melbourne).
Franklin, J and Woodfield, P, 1971. Comparison of a polyester resin and a mechanical rockbolt anchor, Transactions of the Institute of Mining and Metallurgy, (A)80:91-A100.
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 – end of grant summary report. UNSW Mining Research Centre, ACARP Project C7018.
Gerdeen, J C, Synder, V W, Viegelahn, G L and Parker, J, 1977. Design criteria for roof bolting plans using fully grouted nontensioned bolts to reinforce bedded mine roof. USBM Contract No. JO366004.
Hagan, P C, 2003. The effect of resin annulus on anchorage performance of fully encapsulated rockbolts, in Technology Roadmap for Rock Mechanics - Proceedings 10th International Congress on Rock Mechanics ISRM 2003, (South African Institute of Mining and Metallurgy: Johannesburg).
Hagan, P C and Weckert, S, 2002. Anchorage and failure mechanisms of fully encapsulated rockbolts – Stage 2, UNSW Mining Research Centre, ACARP Project C10022.
Hocking, R J, 2000. A study of the variation of important rockbolt parameters and their impact on performance, BE thesis (unpublished), University of New South Wales, Sydney.
Karabin, G J and Debevec, W J, 1978. Comparative evaluation of conventional and resin bolting systems, MESA Report IR1033 (USA Dept of Interior).
Snyder, V W, Gerdeen, J C and Viegelahn, G L, 1979. Factors governing the effectiveness of roof bolting systems using fully resin-grouted nontensioned bolts, in Proceedings 20th Symposium on Rock Mechanics, pp 607-613.
Yeaby, M, 1991. Practical guide to rock bolting. (ANI Arnall).