Field of the invention

The invention relates to a rock bolt assembly for use in mining and tunnelling operations, construction and civil engineering applications.

Background to the invention

In many hard rock mines, such as gold, platinum, chrome, manganese and diamond mines, bodies of ore are accessed by means of development tunnels which are excavated by blasting with explosives. The creation of any mine excavation disturbs the Earth's crust. This results in stress concentrations that interfere with the stability of the rock mass surrounding the excavation. It is essential to modify the internal behaviour of the rock mass surrounding excavations in order to reinforce the ground and stabilise excavations.

As a portion of tunnel is excavated, the exposed surface of the rock in the tunnel may be supported by means of rock bolts. Resin-grouted rock bolts are extensively used for this purpose. When effectively installed, a rock bolt exerts a clamping force and a bonding strength that keeps the rock mass together. This increases the cohesive strength of discontinuities in the rock mass and enhances the stability of the excavation.

The two basic functions of a rock bolt are thus to reinforce the surrounding rock mass and to secure it in place. The characteristics of the underground environment and requirements of the mining operation may determine not only the type of rock bolt to be used, but also the dimensions of the hole to be drilled to accommodate the rock bolt. Accordingly, there may be certain hole size limitations to deal with when installing rock bolts. In particular, it will be appreciated by those skilled in the art that, due to the underground conditions (e.g. depth and rock characteristics) it may be necessary to drill holes of a certain diameter or within a certain range of diameters. The relationship between the diameter of the hole and the diameter of the rock bolt has a significant effect on the efficiency of the reinforcement. Therefore, this often necessitates the use of rock bolts with complementary diameters.

As an example, the Applicant has found that when installing rock bolts in the Mponeng gold mine in South Africa (currently the world’s deepest mine from ground level), low yield requirements may dictate that a conventional carbon steel rock bolt has a relatively small diameter (e.g. less than 20 mm), while underground conditions/characteristics may dictate the drilling of 38 mm to 42 mm diameter holes, which would necessitate the use of rock bolts with diameters of, say, about 27 mm to 30 mm, to obtain optimal resin coverage between the rock bolt and the hole edge.

The Applicant has found that hole size restrictions or constraints such as those identified above may have negative consequences. In particular, a rock bolt installer may be forced to select a rock bolt of a certain diameter due to these hole size limitations, but the selected rock bolt may lack the appropriate load bearing, deformation, energy-absorbing and/or rock supporting characteristics to perform in the desired manner. This is undesirable as the use of incorrect rock bolts may cause failures and rockfalls underground, which in turn create safety hazards.

In light of the above, a need has been identified for a rock bolt which can better accommodate the requirements of certain installations, particularly relatively large- hole installations. For instance, there is a need for a rock bolt which, when encapsulated in resin in, say, a 38 mm to 42 mm diameter hole, provides suitable resin mixing and point load anchoring performance along with excellent energy-absorbing and yielding performance.

The Applicant previously developed a rock bolt made from austenitic steel with high strength and high ductility. Certain aspects of the rock bolt are described in a Patent Cooperation Treaty (PCT) patent application filed by the Applicant, published as PCT Publication No. WO/2019/053653 (hereinafter referred to in this background section as the “Applicant’s rock bolt”).

The Applicant’s rock bolt preferably employs alloyed steel with a manganese content of 10% to 24%. It has been found that the incorporation of this relatively high manganese content (among other things) in the steel provides improved energy absorbing and yielding characteristics. The Applicant’s rock bolt exhibits stiff behaviour at the onset of loading, as well as high strength and excellent deformation characteristics which allows it to overcome certain problems associated with previously known rock bolts, such as conventional carbon steel rock bolts.

The Applicant’s rock bolt exhibits, post a yield point thereof, under static load conditions, an increase in load capacity and elongation with a substantially uniform reduction in diameter without necking or breaking along an entire displacement zone thereof until the break or fail point of the rock bolt assembly is reached.

Under dynamic load conditions, post the yield point thereof, the load capacity and displacement of the rock bolt increases until a threshold is reached at which a first end of the rock bolt is dislocated from an anchoring composition or dislocated from an anchor point at which the first end is anchored in the rock. The tensile strength of the Applicant’s rock bolt increases with deformation and when an axial load is relaxed the increased tensile strength is maintained.

As a result of these characteristics, the Applicant’s rock bolt is useful in combatting problems such as high stress-induced instability problems, including rock-bursts and rock squeezing commonly found in deep mines.

While the Applicant’s rock bolt has proven to be useful and effective, it may be relatively expensive to manufacture. For instance, it currently costs significantly more to manufacture a rock bolt from austenitic steel with 10% to 24% manganese content than it costs to manufacture a conventional carbon steel rock bolt.

The Applicant has thus identified a need for a rock bolt or assembly which is capable of both exhibiting the advantageous characteristics referred to above and better accommodating the requirements flowing from hole size limitations, but which is more cost-effective to manufacture.

Embodiments of the present invention aim to address the needs identified above, at least to some extent.

Summary of the invention

According to a first aspect of the invention, there is provided a rock bolt assembly comprising: a first section made substantially from carbon steel; a second section defining an effective yielding length, wherein the second section is made from austenitic steel with a manganese content in the range of 10% to 24%, and wherein the second section has an ultimate tensile strength of more than 575 MPa and exhibits elongation to failure of more than 10% of the effective yielding length; and a connecting arrangement for connecting a first end region of the first section to a first end region of the second section such that the first section and the second section function as a unitary rock bolt, in use.

The rock bolt assembly may essentially provide a combination bolt or hybrid bolt, a part of which is made from carbon steel and a part of which is made from high strength, austenitic steel with high ductility, preferably with a composition and characteristics substantially as described in PCT Publication No. WO/2019/053653.

The two sections may be separate rods/bolts secured or securable to each other by the connecting arrangement, with the second section being the utilised as a primary (or sole) yielding component. The first section may be referred to as an anchoring section (or a resin mixing and anchoring section) and the second section may be referred to as a yielding section. Preferably the first and second section are secured to each other in a substantially co-axial manner.

The first section may be configured to be inserted into a hole before the second section. In other words, the first section may be configured to be located in a top or end region of the hole, in use, while the second section extends from the first section up to a hole opening or to a point near the hole opening.

Embodiments of the invention provide for the first and second sections to have different diameters.

The first section may be a carbon steel rod. In one embodiment, the first section may have a diameter of between about 25 mm and about 32 mm and a length of between about 450 mm and about 750 mm, and the rock bolt assembly may be configured for insertion or installation into a 38 mm to 42 mm diameter hole. However, various different diameters and lengths may be employed in embodiments of the invention.

The first section may define one or more radial protrusions or paddles to facilitate mixing of an anchoring composition and to provide a larger surface area for bonding with the composition. The first section may thus include a profiled end.

The first and second sections may be connected such that they extend end-to-end and substantially co-axially in the hole, thereby to function as a unitary rock bolt in use.

The second section may be a substantially smooth rod or at least the majority of the second section may be smooth. In one embodiment, the second section may have a diameter of about 16 mm to 20 mm and a length of about 1 .5 m to 2.5 m, and may be configured for insertion along with the first section into a 38 mm to 42 mm diameter hole. However, various different diameters and lengths may be employed in embodiments of the invention.

In embodiments of the invention, the second section may have a substantially smaller diameter than the first section, e.g. around 18 mm compared to around 30 mm. This has been found to be useful in meeting conflicting requirements in certain underground environments, which will become clearer from the descriptions below.

The second section may include a threaded first end region, a threaded second end region, and a non-work hardened zone there between. The non-work hardened zone is preferably a smooth zone with a circular cross-section. The smooth zone may have a length of about 1 .4 m to 1 .8 m. This smooth zone may define the effective yielding length, and may also be referred to as a displacement zone. The threaded zones may be work hardened by cold threading of these zones during manufacture. The smooth zone (effective yielding length / displacement zone) has not been work hardened by cold threading which permits displacement of the smooth zone when a load is applied to the rock bolt assembly.

In one embodiment, the rock bolt assembly has an overall length of approximately 2.4 m.

The rock bolt assembly may include securing means and may include, at the second end region of the second section, an external thread for attachment of securing means to the second section. The securing means may be in the form of a nut for tightening a bearing plate relative to a rock face.

The steel of the second section has the ability to withstand dynamic impact loads that are typical with seismic movements and also strengthens under increased load.

The second section may exhibit, post a yield point thereof, under static load conditions, an increase in load capacity and an increasing elongation, with a substantially uniform reduction in diameter of the second section, until a break or fail point is reached. Under static load conditions, the increase in load capacity may be substantially linear. Under static load conditions, the ultimate tensile strength and break point of the second section may be substantially the same.

The second section may further exhibit, post the yield point thereof, under dynamic load conditions, an increase in load capacity and elongation/deformation until a break or fail point or threshold is reached. The second section may have a dynamic load capacity greater than the static load capacity thereof.

It has been found that the increase in load capacity and deformation/elongation exhibited by the second section under static and/or dynamic load conditions significantly exceeds that of traditional carbon steel bars or rods, e.g. the first section. Preferably the ultimate tensile strength of the second section defines the ultimate tensile strength of the rock bolt assembly.

The steel of the second section preferably has a manganese content of between 10% and 24% and is known as a transformation induced plastic (TRIP/TWIP) high manganese steel which is a high strength, austenitic steel with high ductility.

Preferably, when a load is applied in use, the second section yields at a yield point (e.g. at 95 to 110 kN for a diameter of about 18 mm). Thereafter it continues to elongate with increasing load in a linear or near liner fashion, but with lower stiffness. Failure may occur at elongation of about 14% to 20% of a yielding section length (effective yielding length) of the second section and at a load of greater than 180 kN, breaking the bond between the steel and the resin. No conventional necking occurs and no de-bonding layer is required. Energy is absorbed by elongation of the alloyed steel of the yielding section. There is preferably substantially no deformation or elongation in the first section and no relative movement between the first section and the anchoring composition (resin).

This specification refers to the second (manganese) section being of “high strength” and “high ductility”. Normal rebar typically has an ultimate tensile strength (UTS) of about 250 MPa, while typical “mining bar” has a minimum UTS of 575 MPA. Accordingly, for purposes of interpreting this specification, the term “high strength” refers to a UTS of more than 575 MPa. However, preferred embodiments of the invention provide a UTS of more than 600 MPa, and preferably more than 700 MPa.

For purposes of interpreting this specification, the term “high ductility” refers to the second (manganese) section of the rock bolt assembly exhibiting an elongation to failure of more than 10% of an effective yielding length (displacement zone) of the second section, preferably more than 15% of the effective yielding length of the second section. It is important to note that, in the context of this specification, elongation is determined as a percentage of the entire effective yielding length (displacement zone) of the second section, which as at least 1 m. As an example and as more fully described below, prototypes of the rock bolt assembly according to the invention were tested and the lowest elongation/yield to failure measured was 17.7% over the length of the yielding part of the bolt. It is noted that certain standards such as ASTM International define ductility by considering elongation or yield over a much shorter section of the bolt, e.g. 4 or 5 times diameter or a predefined length such as 200 mm.

The rock bolt assembly may, in use, be configured as a fully grouted rock bolt with an anchoring composition such as resin or cement. The assembly may thus be referred to as a “resin-grouted rock bolt assembly”. Different bolt types exist, namely mechanically anchored, friction anchored and grouted - and embodiments of the invention are specifically applied to the grouted type.

The connecting arrangement may include complementally shaped formations on the first section and the second section and/or at least one separate coupling component. For instance, the connecting arrangement may include:

• a parallel male thread on the first end region of the second section and a complemental threaded female orifice on the first end region of the first section (or vice versa);

• a roof bolt coupling as illustrated in South African Registered Design No. F2013/00339, currently in the name of the Applicant;

• a coupling system as described in South African Patent No. 2014/08984, currently in the name of the Applicant; or

• any suitable conventional external parallel coupling.

The connecting arrangement may thus be provided by complementally shaped threaded zones on the first section and the second section.

According to a second aspect of the invention, there is provided a rock bolt assembly comprising: a first rod or bar having a first diameter; a second rod or bar having a second diameter which is less than the first diameter; and a connecting arrangement for connecting a first end region of the first rod to a first end region of the second rod such that the first section and the second section function as a unitary rock bolt, in use. The first rod may be made from carbon steel and the second rod may be made from high strength, austenitic steel with high ductility. The second rod preferably includes the manganese content of 10% to 24% referred to above.

According to a third aspect of the invention, there is provided a second section or rod substantially as described above, the second section or rod being configured to be secured to a first section or rod substantially as described above to form a rock bolt assembly.

According to a fourth aspect of the invention, there is provided a rock bolt kit which includes the first section or rod and the second section or rod substantially as described above. The kit may also include the connecting arrangement substantially as described above.

Brief description of the drawings

The invention will now be further described, by way of example, with reference to the accompanying drawings.

In the drawings:

Figure 1 is a plan view of an embodiment of a rock bolt assembly according to the invention, installed in a rock mass;

Figure 2 is an enlarged plan view of a first section and part of a second section of the assembly of Figure 1 , with the first section being shown in longitudinal sectional view to illustrate a connecting arrangement of the assembly;

Figure 3 is a three-dimensional view of an example of a connecting arrangement which may be utilised in embodiments of the invention; and

Figure 4 is a plan view of the example of Figure 3; Figure 5 is a plan view of a further example of a connecting arrangement which may be utilised in embodiments of the invention; Figure 6 is a longitudinal sectional view of a coupler of the connecting arrangement of Figure 5; Figure 7 is a conceptual illustration of a further example of a connecting arrangement which may be utilised in embodiments of the invention;

Figure 8 is three dimensional view of female part of the connecting arrangement of Figure 7; Figure 9 is a side view of a male part of the connecting arrangement of Figure 7; Figure 10 is a plan view of a further example of a connecting arrangement which may be utilised in embodiments of the invention; Figures 11 -15 are photographs showing results of a resin mixing and encapsulation test; Figures 16-18 are load-deformation curves for three tested prototype bolts made from austenitic steel with a high manganese content as employed in an embodiment of the invention;

Figures 19-21 are load-deformation curves for tested prototype assemblies in a third set of tests; Figures 22-24 are load-deformation curves for tested prototype assemblies in a fourth set of tests; Figures 25 & 26 are diagrams of workstations used for dynamic drop testing of a rock bolt assembly according to the invention; and Figure 27 is a schematic diagram of a rock bolt assembly which was subjected to the dynamic drop testing; and Figures 28 to 37 are graphs illustrating the results of dynamic testing, with Figures

28 to 37 illustrating the results of test numbers 1 to 10, respectively.

Detailed description with reference to the drawings

The following description of the invention is provided as an enabling teaching of the invention, is illustrative of the principles of the invention and is not intended to limit the scope of the invention. It should be appreciated to those skilled in the art that, without derogating from the scope of the invention as described, it is possible that there are various alternative embodiments or configurations or adaptions of the invention and its features. As a result, it is possible that the described rock bolt assembly may be modified such that it can be used or applied in other industries, to assist with and improve reinforcement, without derogating from the scope of the invention. The term “rock bolt” or “rock bolt assembly” as it applies to the current invention, may therefore be used to describe a similar bolt which is used or adapted to be used in civil engineering applications such as geotechnical applications and/or seismic designs for buildings, amongst others. Such a bolt or assembly may therefore be anchored, embedded, installed or otherwise in other environments, or bodies/volumes of other material/s.

Figure 1 illustrates an embodiment of a rock bolt assembly 10 according to the invention. Figure 1 shows the rock bolt assembly 10 installed in a drill hole 20 in a rock mass 22, surrounded by an anchoring composition in the form of resin grout 28.

The assembly 10 comprises a first section in the form of a carbon steel bar/rod 12 and a second section in the form of a high strength, austenitic steel bar/rod 14 with high ductility. The assembly 10 is installed such that the rod 12 goes into the hole 20 first and is situated at the top of the hole 20, and part of the rod 14 extends through an opening 30 of the hole 20 at the opposite end.

The assembly 10 has two distinct parts: the first rod 12 provides a resin mixing and anchoring section and the second rod 14 provides a yielding section.

In this example embodiment, the rod 12 has a main diameter of 27 mm (and a paddle diameter of 30 mm, as clarified below) and a length of 650 mm, while the rod 14 has a diameter of 18 mm and a length of 1.8 m. The rod 14 is thus significantly longer than the rod 12, but has a smaller diameter. The hole 20 has a diameter of 38 mm, providing space for the resin 28 (or any other suitable anchoring composition) around the assembly 10.

It will be appreciated that a wide range of lengths and diameters may be employed in other embodiments of the invention. For example, the first section may have a diameter of between about 15 mm and 45 mm and a length of between about 300 mm and 2 m. The second section may, for example, have a length of between about 1 m and 4 m and a diameter of between about 10 mm and 35 mm.

A first end region 16 of the rod 12 is connected to a first end region 18 of the rod 14 such that the rods 12, 14 extend in an end-to-end and co-axial manner along a length of the hole 20. The rods 12, 14 are connected by a connecting arrangement consisting of a parallel male external thread 24 at the first end 18 of the rod 14 and a complementally shaped, threaded female hole 26 in the first end 16 of the rod 12 (see Figure 2). Figures 3 and 4 illustrate a further example of such a connecting arrangement between a carbon steel bar 42 (with a greater diameter) and a high strength austenitic steel bar 44 (with a smaller diameter). Flowever, it will be appreciated that the rods 12, 14 may be secured to each other in many different ways without departing from the scope of the invention. To illustrate this, further examples are shown in Figures 5 to 10 and described below.

Turning again to Figures 1 and 2, an outer surface of the rod 12 is provided with protrusions or “paddles” 38 to facilitate mixing of the resin 28 and to provide a larger surface area for bonding with the anchoring composition. The rod 12 may thus be provided with scalloped edges. It will be appreciated that the paddles 38 enlarge the diameter of the rod 12 in certain areas, hence the reference to paddle diameter above.

A second end region 36 of the rod 14 is configured to receive a nut 32 and a bearing plate 34.

The rod 14 is manufactured from and comprises austenitic steel with a manganese content in the range 10 to 24%, more preferably in the range of 10 to 18%, or even more preferably about 17%. The steel used in the rod 12 is known as a transformation induced plastic (TRIP/TWIP) high manganese steel which is a high strength, austenitic steel with high ductility.

The threaded regions 24, 36 of the rod 14 are work-hardened (e.g. by cold threading) while the region between them is not work-hardened. The latter region can be referred to as a smooth bar region 40 (shown in Figure 1 ) and this is, in use, the effective yielding section of the rod 14. In this embodiment, the yielding section is approximately 1.6 m long and this length can be referred to as the “effective yielding length”.

In use, the assembly 10 may be supplied in a fully assembled and pre-torqued condition so as to lock/secure the connection between the first section 12 and the second section 14. The sections 12, 14 function as a unitary rock bolt.

A hole with suitable diameter (e.g. 34 to 42 mm) may be drilled, for instance about 2.35 m deep in the rock 22 to be supported. Resin capsules of the appropriate diameter, length and setting times may be inserted. The assembly 10 is inserted into the hole.

Upon installation, the paddles 38 help to mix the resin 28, thereby anchoring the rock bolt assembly 10 to the rock 22. After the waiting time recommended for the resin 28 has expired, the nut 32 is then tightened against the bearing plate 34 and subsequently tightened against the rock 22. This introduces a tensile load on the rock bolt assembly 10 which supports the rock 22.

The assembly 10 may be used in conjunction with rock surface support such as mesh, lacing cables, sprayed cement, or other elements.

The carbon section 12 has low dynamic deformation characteristics, while the austenitic steel section 14 acts as the fundamental yielding or deforming component of the assembly 10.

It has been found that the rod 12 provides the correct point load anchoring performance when encapsulated in the resin 28 (optimal resin thickness coverage between the rod 12 and the 38mm hole annulus), whilst at the same time the smaller diameter (18 mm) of the coupled rod 14 with a lower yield strength than the carbon rod 12, offers a yield point at a load of between 80 - 110 kN. Besides yielding at a lower level, the manganese rich section 14 also ensures that the overall rock bolt assembly 10 has unique dynamic impact characteristics, which allows the assembly 10 to deform and strengthen when dynamic loads are applied. As mentioned above, the rod 14 exhibits, post the yield point thereof, under static load conditions, an increase in load capacity and an increasing displacement (i.e. elongation) until a break or fail point is reached. Under static load conditions, the increase in load capacity is be substantially linear. Under static load conditions, the ultimate tensile strength and break point of the rod 14 are substantially the same. The rod 14 further exhibits, post the yield point thereof, under dynamic load conditions, an increase in load capacity and elongation until a break or fail point or threshold is reached at which the rod 14 or the rock bolt assembly 10 is dislocated from the anchoring composition or dislocated from an anchor point. As it is dislocated, it starts anchor ploughing or dragging against its surroundings which in turn absorbs additional energy. This has been found to be an additional advantage of the use of the material described herein.

The rod 14 has a dynamic load capacity greater than the static load capacity thereof. The tensile strength increases with deformation and when the axial load is relaxed the increased tensile strength is maintained.

The TRIP/TWIP high manganese steel used to make the rod 14 is a transformation induced plasticity steel, in which the metastable austenite transforms to martensite during deformation of the steel. The mechanical properties of the steel are the result of the transformation induced plasticity in the steel which leads to enhanced work hardening rate, postponed onset of necking and improved formability.

In the TRIP/TWIP high manganese steel, the metastable austenite will not only deform plastically, but it transforms to the more stable a’- martensite upon application of a tensile load. The exceptional mechanical properties of the steel are directly related to this strain-induced phase transformation. Exceptional work hardening as well as phase transformation occurs during mechanical deformation. The deformation of the steel occurs by a combination of slip or dislocation glide and a secondary transformation to martensite. The martensite platelets that form as a result of the transformation act as planar obstacles and reduce the mean free path of the dislocation glide. Dislocations pile up at interfaces between these planar defects and the matrix and causes significant back stresses that impede the progress of similar dislocations. The significant work hardening caused by these planar defects delays local necking and results in increasing linear displacement.

The combination of carbon steel and manganese steel sections makes the rock bolt assembly 10 more economical to manufacture. The carbon steel section 12 costs significantly less than the rod 14 with high manganese content. In this regard, the Applicant has found that whilst the carbon section 12 offers nominal dynamic characteristics, by combining it with a significant manganese section 14, the overall coupled bolt assembly 10 (“combination anchor”) still offers excellent characteristics, both statically and dynamically.

The Applicant is aware of other prior art rock bolts with manganese content developed by third parties, but the prior art rock bolts of which the Applicant is aware are fundamentally different to the Applicant’s rock bolt. For example, the well-known “SWELLEX” rock bolt is a frictional bolt made from manganese steel with less than 5% manganese content. Manganese steel has been used in rock bolts mainly for its toughness properties. Furthermore, it has been introduced into bolts in low proportions (less than 5%) which cannot provide the uniform reduction in diameter as provided in the present invention.

Not only is the methodology of using frictional resistance in a rock bolt fundamentally different to the methodology of using the rock bolt of the present invention, the manganese content in rock bolts such as the SWELLEX bolt is insufficient to allow the rock bolt to exhibit the properties described above. Furthermore, these known bolts do not provide a hybrid/combination assembly with two different components and/or diameters as described herein.

The rock bolt assembly as described herein is useful in combatting instability problems such as high stress-induced instability problems, including rock-bursts and rock squeezing.

In the event of either static or dynamic movement of the rock 22 occurring in the direction of the bearing plate 34, which is the downward movement of the rock 22, the tensile load on the rock bolt assembly 10 will increase. This results in the displacement of the rod 14. The displacement of the rod 14 causes the diameter of the bolt assembly 10 to be reduced in the smooth bar region 40 which breaks the bond between the rock bolt assembly 10 and the resin 28 along the length of the smooth bar region 40 of the rod 14.

The configuration of the rod 14 with two work-hardened end regions and the smooth bar region 40 there between, is specifically configured to be used with a rock bolt which is manufactured using the manganese content described above. As far as the Applicant is aware, a rock bolt manufactured from any other material or combination of materials, which has the same configuration as described above, will not achieve the same level of success as the rock bolt with the rod 14 as described.

As mentioned above, the rods may be secured to each other in different ways to form the rock bolt assembly. Further examples are shown in Figures 5 to 10.

In the example of Figures 5 and 6, a first bolt 52 and a second bolt 54 are connected to each other by a roof bolt coupling 50 as described in South African Registered Design No. F2013/00339. A hole extends centrally and longitudinally through the coupling 50 and a pair of internally threaded zones 58, 60 are defined in the hole. Each zone 58, 60 tapers towards a central region of the coupling 50. The zones 58, 60 are respectively configured to receive a tapered, externally threaded end of the bolt 52 and a tapered, externally threaded end of the bolt 54. This is shown in Figure 6.

Due to the taper in the end of each bolt 52, 54 and the tapered threading inside of the coupling 50, the coupling 50 can have substantially the same maximum diameter as the bolts 52, 54, i.e. it does not stand proud of the bolts 52, 54.

In the example of Figures 7 to 9, a first bolt 70 has a male tapered end 78 and a second bolt 72 has a female tapered end 80. The tapered ends 78, 80 are complementally threaded 74, 74, so that the two bolts 70, 72 may be screwed together to form a single combination roof bolt. This coupling system is described in South African Patent No. 2014/08984. The example of Figure 10 is a conventional external parallel coupling. A cylindrical coupler 90 has an internally threaded opening 96, 98 at each end region thereof and the rods 92, 94 are externally threaded so that they may be screwed into the coupler 90. The threaded parts do not have tapered diameters and, as a result, the coupler 90 has a diameter which is greater than the diameters of the rods 92, 94.

A rock bolt assembly according to the invention may be advantageous in that it can better accommodate the requirements of certain installations, particularly large-hole installations, and may be more cost-effective than rock bolts made exclusively from TRIP/TWIP high manganese steel. The invention provides for the combination of a first bolt, e.g. a carbon steel bolt with a larger diameter to conform to the required hole size, with a second bolt which may have a different diameter and which may provide the assembly with desired performance characteristics. In this way, the rock bolt installer has greater flexibility to select a combination of bolts which may balance the installation requirements necessitated by the underground conditions and/or operating environment with the appropriate load bearing, deformation, energy-absorbing and/or rock supporting characteristics.

The rock bolt may be referred to as a “hybrid-diameter yielding rock bolt” based on the yielding characteristics and embodiments in which its two sections have different diameters.

Embodiments of the present invention have been tested and some of the test results are described below with reference to Figures 11 to 37.

Summary of certain test results/findings

Prototypes were developed with a particular (but not exclusive) focus to meet performance and application requirements of Mponeng gold mine in South Africa. The low yield requirement dictated a bolt with a small diameter (less than 20 mm) while resin grouting in 38-42 mm drill holes dictated a diameter of nearly 30 mm. These conflicting requirements were met by creating a rock bolt assembly with a resin mixing and anchoring section with an effective diameter of 31 mm coupled to a yielding section of manganese alloy steel with a diameter of 18 mm. Typical specifications are shown in Table 1. Table 1 : Typical rock bolt assembly specifications

The assembly in Table 1 also includes “Secura” deformations/paddles which makes the effective anchoring diameter (paddle diameter) about 31-32 mm. The rock bolt assembly was tested with the following parameters:

• Anchored with “Lokset” resin capsules (as provided by Minova Africa) in 40 mm diameter pipes

• Static loading rate 90mm/minute

• Dynamic impact speed 6m/sec

Typical rock bolt assembly performance is summarised in Table 2.

Table 2: Rock bolt performance Mode of action:

• When subjected to an axial load, the assembly initially provides a stiff reaction force comparable to a conventional rock bolt.

• At 95-110 kN (between approximately 373 MPa and 432 MPa) the yielding section yields and continues to elongate with a typical stiffness of 0.48 kN/mm (for a 1 .6 m yielding section) until failure at around 200-220 kN load (i.e. stress of around 785 MPa to 864 MPa).

• Total static elongation to failure on a 1.6 m yielding section is 300 mm to 400 mm. This constitutes a high ductility - elongation of between 18% and 25% in terms of the “high ductility” definition provided above.

• All the yield takes place in the yielding section and the anchoring section remains firmly anchored in the mixed resin.

Some advantages identified from testing include:

• The assembly provides consistent high kinetic energy absorption and excellent elongation performance. This makes it suitable for use in mines experiencing seismic activity or mines with “squeezing” or swelling rock.

• Predictable static and dynamic performance

• Dynamic energy absorption capacity demonstrated to 40 KJ without failure

• Static elongation of over 300 mm

• Low initial yield load, reducing rock disintegration between bolts

• Performs similar to a conventional rock bolt with load capacity of 200 kN

• Can be used successfully in “large” holes up to 42 mm in diameter

• The rock bolt assembly has been found suitable for application in Mponeng gold mine, with their requirements of 38-40 mm holes, about 80-100 kN yield point, energy absorption of at least 30 kJ, within deformation of 300 mm, while the rock bolt assembly is able to function as a conventional rock bolt with ultimate strength of over 200 kN.

First set of tests

A prototype was tested. The purposes of this set of tests were to test anchoring of the anchoring section by pull testing a bolt secured in a 39 mm pipe (simulating a hole in rock) by Lokset capsule resin, and to assess resin capsule mixing and bolt encapsulation by installing a full-length bolt of 2.9 m in a steel pipe and then cutting sections away to observe the internal condition. Insertion rate was pre-set at 3 m per minute and approximately 280 rpm. Anchor test

The anchor section, with a short length of 18 mm diameter manganese rod screwed in, was installed in a 39 mm (internal diameter) x 690 mm long thick-walled steel pipe. The pipe was not roughened internally. The anchor was grouted with a 32 mm diameter x 60 second set Lokset resin capsule. After installation, the end of the pipe was cut off to reveal the end of the bolt. Good resin mixing was observed in both specimens.

One hour after installation, the anchor was tested by tensioning with a hydraulic ram, manually pumped. Measurements were taken of bolt diameter and full (external) deformation of the bolt. Deformation between two index marks on the smooth yielding portion of the bolt was also measured (“Internal” measurements). Measurements were made using a digital Vernier calliper.

Mixing test

The pipe was filled with 32 mm diameter Lokset resin capsules; the toe-end had a 500 mm x 60 second capsule and the balance was filled with 5-10 minute resin. The bolt was thrust in while being rotated for the full length of the pipe.

The installation was completed within the recommended spin times of the resin capsules. After the resin had set the pipes were cut away in a number of places to expose the resin for assessment of filling and mixing. The bolt was cut through completely to expose the full cross-section.

Results of anchor test

Load rose to 15 tonnes and kept that level until the test was stopped. The measurements are shown in the table below.

Table 3: Anchor test results

Observation of the cut-away toe ends of the bolts showed no significant movement of the anchor end. Deformation was therefore of the 18 mm steel, not pull-out of the anchoring section.

Results of mixing test

The pipe was cut open at 4 places, to examine resin grout condition. Figure 11 shows position 1 , 500 mm from the toe-end. The resin is well-mixed and hard 60 second resin is visible around the anchor section. A small amount of film was visible. Figure 12 shows position 2, 1100 mm from the toe-end. The resin is well-mixed and hard 5- 10 minute resin is visible. Figure 13 shows position 3, 1700 mm from the toe-end. The resin is well-mixed and a small amount of film was visible. Figure 14 shows position 3, 2000 mm from the toe-end. Well-mixed resin is visible. Figure 15 shows a cross- section of the pipe and bolt, illustrating well-mixed resin.

The following conclusions could be drawn:

• The anchoring section provides more than 15 tonnes of anchorage.

• This is more than the design yield of the yielding section, ensuring that the yielding properties of the steel are fully mobilised/utilised.

• Resin mixing and bolt encapsulation were good.

Second set of tests

Bars of 20 mm made of the manganese alloy steel described above were tensile tested to destruction. The objectives of the tests were to establish the static yielding characteristics and to verify that static energy absorption capacity of bolts made from the Mn-alloy can be scaled according to the volume of steel subject to loading.

Three bars were tested. Specifications were: · Overall length: 3.1 m

• Thread length: 150 mm at each end

• Test length: 2.8 mm (nominal)

• Diameter: 20 mm (nominal) The bars were passed through a platen at each end and secured by double nuts. Diameters were measured close to each end of the bar and near the centre. Tensile load was applied, by elongation at the rate of 90 mm/minute. Loading was stopped and the load maintained at 80 kN, 120 kN, 160 kN and 200 kN for re-measurement of diameters. Loading was then continued until rupture. Diameters were re-measured, post-rupture. Load and deformation were measured, recorded and displayed electronically. Diameters were measured with a digital Vernier calliper.

The load-deformation curves for Bolts 1 , 2 and 3 are shown in Figures 16, 17 and 18, respectively. The tables below show a summary of the overall results and a summary of diameter measurements.

Table 4: Summary of overall results

Table 5: Summary of diameter measurements for the three tested bolts

The following observations were made:

• The three bolts showed consistent behaviour.

• Each bolt fractured within the threaded portion, which implies that the ultimate load and deformation capacity of the effective/smooth yielding section of the Mn-steel is higher than presented here.

• The uniform deformation along the length of the bars allows energy absorption capacity to be scaled by length.

• None of the dynamic tests performed produced fracturing of the steel.

Third set of tests The bolts used for this set of tests had an anchoring section with 27 mm diameter and a length of 650 mm with “Secura” deformations, giving an overall mixing diameter of 31 mm. The anchor section was thread-coupled to an 1800 mm long, 18 mm diameter smooth-bar yielding section made from the Mn-alloy steel described above. The yielding sections had been produced by turning-down of 20 mm rods. The free end of each yielding section was threaded for a length of 150 mm. Three test series were performed:

• Series A: Direct tension on the bolts, without embedment in pipes

• Series B: Tensile testing of bolts embedded in resin in steel pipes; pipes split

• Series C: Tensile testing of bolts embedded in resin in steel pipes

The specimens were prepared using a suitable test rig. The specimens were tested in steel pipes 2.4 m in length and with a 40.3 mm diameter (internal). One end of the steel pipe is closed with a welded-on steel cap. Lokset resin capsules were used (35 mm diameter with 85/15 resin mastic/catalyst paste ratio; 1 x 500 mm x 60 sec for the anchoring section; 2 x 900 mm x 5/10 m for the yielding section).

The bolts were spun through the resin capsules in the pipes, with left hand rotation. Full-depth penetration took 20 seconds and spinning was continued for 8 seconds to ensure good mixing of the anchoring section. All bolts inserted easily, with no stalling of either thrust or rotation. After insertion each specimen was left undisturbed for 6 minutes to allow the 5/10 minute resin to set, then removed from the rig.

Specimens 1 to 5 (see Table 6) had the pipes split circumferentially at 900 mm from the collar (approximately halfway along the yielding section).

Tensile testing of specimens was carried out using a 500-tonne Mohr & Federhaff tensile testing machine. Deformation was controlled at 90 mm/minute.

For specimens A to E, only the yielding portions were tested. The bars were secured in the machine by double nuts on the threaded portion at each end. The length of bar subjected to load was 1630 mm.

Series B testing (Tensile testing of bolts in split pipes) For specimens 1 to 5, the closed end of each pipe was gripped with “V” blocks. A thick plate was placed between the open end of the pipe and the start of the threaded section of the bolt. Double nuts were tightened up against the plate, which was then loaded by the testing machine.

For specimens 6 to 10, the specimens were secured in the testing machine in the same way as the Series B specimens. Specimen 6 failed anomalously at low load, just above the yield point of the steel. The test was stopped and re-started to gain information on post-failure behaviour (Test 6a).

Test results

The results are summarised in Tables 6 and 7 below and sample load-deformation curves are included in Figures 19 to 21 . For illustrative purposes, specimens “E”, “4” and “7” are shown in Figures 19 to 21 , respectively.

Table 6: Observations in Series A, B and C tests (third set of tests)

In Table 6, for tests 1 and 9, the figures in parentheses are the deformations at which the loads started to rise; deformation prior to that point is bedding-in of the specimen in the machine.

Table 7: Analysis of Series A, B and C test results (third set of tests)

Discussion of results

The results are consistent except for Specimen 6, which failed at low load, possibly as a result of incorrect loading. After testing, the closed end of the pipe was cut off to examine the condition of the anchor section of the bolt. The anchor had not moved, ruling out resin-anchor failure as cause of the premature failure. The discussion below excludes Specimen 6.

Failure load and mode:

Tensile stresses at failure (UTS) were between 740 - 750 MPa. In each case, failure was in the transition between the threads and the undisturbed steel. This indicates that the failure loads were governed by the machining of the threads, rather than the properties of the material itself.

Diameter reduction of approximately 2 mm was seen. This was sufficient to cause full de-bonding of the yielding section from the surrounding resin. De-bonding is an essential characteristic to prevent local stress concentrations in the bar; the manganese alloy steel de-bonds without needing a de-bonding layer. After failure, the bars could be moved by hand within the resin.

Yield loads: The yield load for each specimen in Table 7 is given as a range, as transition was not instantaneous. Nevertheless, the ranges are narrow, giving the bolts a consistent behaviour. Comparison of results from specimens A - E versus 1 - 10 shows that resin encapsulation slightly raises the yield load.

Energy absorption:

The strain energy absorbed by each specimen is presented in Table 7. Energy absorbed was calculated as the area under the load-deformation curve, using a trapezoidal numerical integration formula. Where yield to failure exceeded 300 mm, energy absorbed up to 300 mm is also given, as 300 mm is widely considered as the maximum practical deformation acceptable in a mine tunnel. All of the specimens comfortably exceeded the design requirement of 30 KJ, with a mean of 45 KJ.

The third set of tests described above successfully qualified the hybrid-design yielding rock bolt assembly according to the invention as a “static/convergence” bolt for use with anchoring compositions such as resin.

Fourth set of tests

For the fourth set of tests, the bolt assemblies used had an anchoring section of 27 mm diameter and 650 mm length, with “Secura” deformations, giving an overall mixing diameter of 31 mm. The anchor section is thread-coupled to a 1750 mm long and 18 mm diameter smooth-bar yielding section made from mill-rolled Mn-alloy steel (with the manganese content as described above). The free end of each yielding section is threaded for 150 mm.

The objectives of these tests were to qualify the bolt assemblies made from rolled steel in terms of static yield load, ultimate strength, amount of yield and energy absorption, and to quantify the differences in performance between rolled and machined-down Mn-steel bolts (with machined bolts having been used in the third set of tests above).

Three test series were performed: • Series A: Direct tension on the bolts, without embedment in pipes

• Series B: Tensile testing of bolts embedded in resin in steel pipes

• Series C: Tensile testing of bolts embedded in resin in steel pipes; pipes split

The specimens were prepared using a suitable test rig. The specimens were tested in steel pipes 2.4 m in length and with a 40.3 mm diameter. One end of the steel pipe is closed with a welded-on steel cap. Lokset resin capsules were used (35 mm diameter with 85/15 resin mastic/catalyst paste ratio; 1 x 500 mm x 60 sec for the anchoring section; 2 x 900 mm x 5/10 m for the yielding section).

The bolts were spun through the resin capsules in the pipes, with left hand rotation. Full-depth penetration took 20 seconds and spinning was continued for 8 seconds to ensure good mixing of the anchoring section. All bolts inserted easily, with no stalling of either thrust or rotation. After insertion each specimen was left undisturbed for 6 minutes to allow the 5/10 minute resin to set, then removed from the rig.

Specimens 6 to 9 (see Table 8) had the pipes split circumferentially at 800 mm from the collar (approximately halfway along the yielding section).

Tensile testing of specimens was carried out using a 500-tonne Mohr & Federhaff tensile testing machine. Deformation was controlled at 90 mm/minute.

Specimens A to E: only the yielding portions were tested. The bars were secured in the machine by double nuts on the threaded portion at each end. The length of bar subjected to load was 1600 mm.

Series B testing (Tensile testing of bolts in continuous pipes) Specimens 1 to 5: the closed end of each pipe was gripped with “V” blocks. A thick plate was placed between the open end of the pipe and the start of the threaded section of the bolt. Double nuts were tightened up against the plate, which was then loaded by the testing machine.

Series C testing (Tensile testing of bolts in split pipes)

Specimens 6 to 9: the specimens were secured in the testing machine in the same way as the Series B specimens.

Test results

The results are summarised in Tables 8 to 10 below and sample load-deformation curves are included in Figures 22 to 24. For illustrative purposes, specimens “1”, “6” and “A” from Table 9 are shown in Figures 22 to 24, respectively.

Table 8: Observations in Series A, B and C tests (fourth set of tests)

Table 9: Summary of results of Series A, B and C tests (fourth set of tests) Table 10: Energy absorption observed in Series A, B and C tests (fourth set of tests)

Discussion of results

Tensile stresses at failure were between 760 - 825 MPa. In each case, failure was in the transition between the threads and the undisturbed steel. This indicates that the failure loads were governed by the machining of the threads, rather than the properties of the material itself. None of the specimens showed “necking down” prior to failure. In all cases the strength of the resin anchor exceeded the strength of the bolt.

In specimens 6 - 9, the resin exposed at the pipe split was examined. The resin was homogeneous and without any voids. This indicates that mixing by the anchoring section of the bolt is good and a solid resin column is created along the entire length of the bolt.

Post-test measurements of bolt diameter showed diameter reductions in the range 1 .58 - 2.00 mm. This was sufficient to cause full de-bonding of the yielding section of the bolt from the surrounding resin. De-bonding is an essential characteristic to prevent local stress concentrations in the bar; the Mn-alloy steel de-bonds without needing a de-bonding layer. After failure, the bars could be moved by hand within the resin.

Yield loads and behaviour:

In Table 9 the range of yield loads is narrow, giving the bolts a consistent behaviour. Comparison of results from specimens A - E versus 1 - 9 shows that resin encapsulation slightly raises the yield load. This is consistent with previous test series. There was no significant difference in yield loads between the solid pipe and split pipe specimens. Comparison of Figures 22 and 23 (resin-encapsulated bolts) with Figure 24 (plain bolt) shows that for encapsulated bolts, over the first +/- 80 mm post-yield, the load increase is low, then the stiffness increases. For plain bolts, the flatter curve zone is either absent or much shorter (+/- 20 mm). The same behaviour is present in all the specimens in their respective groups.

Energy absorption:

Strain energy absorptions are summarized T able 10. Energy absorbed was calculated as the area under the load-deformation curve, using a trapezoidal numerical integration formula. Where yield to failure exceeded 300 mm, energy absorbed up to 300 mm is also given, as 300 mm is widely considered as the maximum practical deformation acceptable in a mine tunnel. All the specimens comfortably exceeded the design requirement of 30 KJ. The encapsulated bolts gave a mean of 35.7 KJ at 300 mm deformation. The plain bolts showed very consistent energy absorptions of 42 - 43 KJ at 300 mm deformation. Specific energy absorption capacity to 300 mm of extension was calculated as 0.106 KJ/mm ³ .

Table 11 summarises and compares the key load-deformation characteristics for the 18 mm machined bolts (from the third set of tests described above) and 18 mm as- rolled bolts (from this fourth set of tests). Each number in Table 11 is the mean of the results for like specimens.

Table 11: Comparison between machined and rolled bolts

For purposes of Table 11 , specimen 6 from the third set of tests was excluded. The numeral (1) in Table 11 is a note used to indicate that most specimens did not achieve 300 mm extension. The rolled steel shows higher load bearing capacity, higher extension to failure and higher ultimate energy absorption capacity. Yield load was not significantly different. Energy absorption to 300 mm appears lower for the rolled steel, but all specimens achieved 300 of extension whereas only 7 out of 14 machined specimens achieved 300 mm of extension.

Fifth set of tests A prototype dynamic yielding bolt assembly was manufactured for use at Mponeng gold mine. The bolt assembly incorporates a 27 mm diameter x 650 mm long anchoring/bonding section coupled to an 18 mm diameter x 1750 mm long yielding section. The objective of this set of tests was to determine if the bolts would buckle during insertion and to establish that the installation rig would be able to provide sufficient thrust and torque to insert the bolts.

Bolts: As described above. The drive nuts used had shear pins with 110 - 120 Nm breakout torque rating.

Pipes: 39 mm (internal diameter) x 2400 mm long steel pipes, to simulate hole dimensions on Mponeng gold mine. The pipes were not roughened internally.

Resin capsules: Lokset two-speed (60 s and 5/10 minutes) two-speed, double clip resin capsules (30 mm diameter x 2400 mm long).

Drill rig: FIPE hydropower drill rig, as used on Mponeng gold mine. For the tests, the boom was horizontal. The rig has a thrust rating of 4000 N and torque rating of 60 - 80 Nm.

Method:

Five tests were carried out. For each test, one of the steel pipes was bolted onto to end of the drill rig boom. The resin capsules were then inserted (60 s portion at the inner end of the pipe). A bolt was attached to the drifter and collared in the pipe. The bolt was inserted into the pipe, using the thrust of the boom and the percussion and rotation of the drifter. Percussion and rotation parameters were varied to determine the effect of each and establish an operating envelope for the system.

Results:

The results are summarised in Table 12 below.

Table 12: Summary of results of fifth set of tests

Sixth set of tests

In the sixth set of tests, the bolt assembly was dynamically tested. The objective of the testing was to inspect the resistance of the bolt assembly to dynamic loading at variable values for the load impact energy and velocity of 6 m/s. This was done with two types of Minova Africa Lokset resin capsules, namely “GREEN” which is a fast resin at the anchor end of the bolt assembly and “YELLOW” which is a slow resin at the yielding section of the bolt assembly.

The dynamic tests were carried out at test workstations as shown in Figures 25 and 26. In Figures 25 and 26, reference numerals indicate the following:

(1) - drop mass

(2) - force sensor

(3) - beam for bolt fastening

(4a) - bolt grouted into a split tube (4b) - bolt grouted into a continuous tube

(5) - impact plate

(6) - bolt case and nut

Figure 26 is the same as Figure 25, but for the difference between (4a) and (4b). Figure 25 shows “Load case 1” in which the bolt was grouted into a split tube and Figure 26 shows “Load case 2” in which the bolt was grouted into a continuous tube. Tables 13 and 14 below refer to Load case 1 as “LC1” and Load case 2 as “LC2”.

Referring to Figure 27, the rock bolt assemblies (A) used in the testing were 2400 mm long with 150 mm of thread. The yielding section (C) was 1750 mm long and its diameter was 18 mm. The Secura-deformed (paddled) anchor section (B) was 650 mm long and 27 mm in diameter with 31 -32 mm paddles. All threads were Rd 18 x 1/8” L/H threads according to DIN 405 standards.

Procedure:

The bolt assembly (A) was grouted into a 2.2 m long tube in the workstations of Figure 25 and 26. In Load case 1 , the assembly (A) was grouted into a split tube at a proportion of 0.925 m (tube upper section) and 1 .270 m (tube lower section). In Load case 2, the assembly (B) was grouted into a continuous tube.

Impact energy E (in kJ) and impact velocity v (in m/s) were determined as follows: mgh 1000 v = j2gh where m is drop mass in kg, h is drop height in metres and g is gravitational acceleration constant 9.81 m/s ² .

The procedure involved raising the drop mass to a determined height corresponding to the given impact energy E and load velocity v. The testing involved free fall of the drop mass from the drop height onto (a) the base of the bolt grouted into a continuous tube and (2) the base welded to the tube, 5 cm above its end.

During bolt testing, the measurement data was registered at a sampling rate of 19.2 kHz. The measured factors were the load F imposed on the bolt and the elongation L as a function of time t. Graphs were used to determine the value of the maximum load value Fmax and energy Wd imposed on the bolt. After testing the bolt grouted into a split tube, further measurements inspected the parting length of the gap between the upper and the lower sections of the tube and in case of the bolt grouted into a continuous tube the total elongation was measured between end of the tube and centering sleeve. The force measurements were carried out via strain gauge sensor, while the elongation measurements were carried out via laser sensor. The sensors were connected to an HBM MGCplus-type measuring amplifier, which worked in cooperation with a computer that registered the measurement data.

The results are shown in Figures 28 to 37 and in Tables 13 and 14 below. Figures 28 to 37 correspond to test numbers 1 to 10, respectively, as marked in Tables 13 and 14. This is summarised as follows:

• Figure 28 - Test 1 / LC1

• Figure 29 -Test 2 / LC1

• Figure 30 -Test 3 / LC1

• Figure 31 - Test 4 / LC2

• Figure 32 -Test 5 / LC2

• Figure 33 -Test 6 / LC2

• Figure 34 -Test 7/ LC1

• Figure 35 -Test 8 / LC2

• Figure 36 -Test 9 / LC2

• Figure 37 -Test 10 / LC1

The following is a key for Figures 28 to 37:

_• Fd= load in kN

_• Fdr= relieved load in kN

_• Wd= energy in J

_• Wdr= relieved energy in J

_• Ld = elongation in mm

Table 13: Summary of results for tests 1 to 6

Table 14: Summary of results for tests 7 to 10