1. Introduction
Chapter 2 is primarily concerned with underground mining. The application of rock mechanics to open pits (slope stability) is not pursued. “Mining is defined as the art of working mines and a mine is defined as an excavation out of which minerals are dug. The latter definition is not complete. Implicit in the word mine is the need to insure that the mine excavation (or excavation cavity) is safe to work in.
“The miner wants to plan development and mining so that the adverse effects of rock failure will be minimized. This is the place for rock mechanics, the science that studies the strength and failure characteristics of rocks.
“The early approach to rock mechanics was to treat rock as an elastic body, not unlike the way civil engineers treat concrete. The elementary concept is that a unit of rock underground is stressed by the weight of rock above it and constrained by other rock around it, thus inducing a horizontal stress, which is a function of Poisson’s ratio.
“The subsequent study of rock masses and the effects of geology has produced more realistic concepts to which the irregular geometry of an ore body gives challenge.” Alex Ignatieff 1970
“Routine rock mechanics may be used to design temporary and permanent ground support designs in elastic or quasi-elastic ground (i.e. hard rock). Nevertheless, such systems fail as the condition of the rock mass approaches the Yield State in which condition plastic or viscous response components become dominant – and any design based on or extrapolated from elastic behavior is void.” John D. Morton 1990
By definition, hard rocks are elastic and hence “routine rock mechanics” (civil engineering concepts of stress and strain) ought to apply. In fact, they do apply; however, the properties of hard rocks are greatly altered by the inevitable presence of joints (the compressive strength of the rock mass is compromised and tensile strength is reduced to mere friction). This alteration must be taken into account. In addition, the integrity of the typical underground hard rock mine and the facility to mine within it, may rely on a structural (cemented) backfill which itself does not exhibit an elastic response. Moreover, hard rock mines that run deep (and some that are shallow) are threatened with rock bursts caused by elastic instability.
2. Rules of Thumb
Ground Stress
• The vertical stress may be calculated on the basis of depth of overburden with an accuracy of ± 20%. This is sufficient for engineering purposes. Source: Z.T. Bieniawski
• Discs occur in the core of diamond drill holes when the radial ground stresses are in excess of half the compressive rock strength. Source: Obert and Stephenson
• The width of the zone of relaxed stress around a circular shaft that is sunk by a drill and blast method is approximately equal to one-third the radius of the shaft excavation. Source: J. F. Abel
Ground Control
• The length of a rock bolt should be one-half to one-third the heading width. Mont Blanc Tunnel Rule (c.1965)
• In hard rock mining, the ratio of bolt length to pattern spacing is normally 1½:1. In fractured rock, it should be at least 2:1. (In civil tunnels and coal mines, it is typically 2:1.) Source: Lang and Bischoff (1982)
• In mining, the bolt length/bolt spacing ratio is acceptable between 1.2:1 and 1.5:1. Source: Z.T. Bieniawski (1992)
• In good ground, the length of a roof bolt can be one-third of the span. The length of a wall bolt can be one-fifth of the wall height. The pattern spacing may be obtained by dividing the rock bolt length by one and one-half. Source: Mike Gray (1999)
• The tension developed in a mechanical rock bolt is increased by approximately 40 Lbs. for each one foot-pound increment of torque applied to it. Source: Lewis and Clarke
• A mechanical rock bolt installed at 30 degrees off the perpendicular may provide only 25% of the tension produced by a bolt equally torqued that is perpendicular to the rock face, unless a spherical washer is employed. Source: MAPAO
• For each foot of friction bolt (split-set) installed, there is 1 ton of anchorage. Source: MAPAO
• The shear strength (dowel strength) of a rock bolt may be assumed equal to one-half its tensile strength. Source: P. M. Dight
• The thickness of the beam (zone of uniform compression) in the back of a bolted heading is approximately equal to the rock bolt length minus the spacing between them. Source: T.A. Lang
• Holes drilled for resin bolts should be ¼ inch larger in diameter than the bolt. If it is increased to 3/8 inch, the pull out load is not affected but the stiffness of the bolt/resin assembly is lowered by more than 80%, besides wasting money on unnecessary resin. Source: Dr. Pierre Choquette
• Holes drilled for cement-grouted bolts should be ½ to 1 inch larger in diameter than the bolt. The larger gap is especially desired in weak ground to increase the bonding area. Source: Dr. Pierre Choquette
Mine Development
• Permanent underground excavations should be designed to be in a state of compression. A minimum safety factor (SF) of 2 is generally recommended for them. Source: Obert and Duval
• The required height of a rock pentice to be used for shaft deepening is equal to the shaft width or diameter plus an allowance of five feet. Source: Jim Redpath
Stope Pillar Design
• A minimum SF of between 1.2 and 1.5 is typically employed for the design of rigid stope pillars in hard rock mines. Various Sources
• For purposes of pillar design in hard rock, the uniaxial compressive strength obtained from core samples should be reduced by 20-25% to obtain a true value underground. The reduced value should be used when calculating pillar strength from formulas relating it to compressive strength, pillar height, and width (i.e. Obert Duval and Hedley formulas). Source: C. L. de Jongh
• The compressive strength of a stope pillar is increased when later firmly confined by backfill because a triaxial condition is created in which s3 is increased 4 to 5 times (by Mohr’s strength theory). Source: Donald Coates
Subsidence
• In block caving mines, it is typical that the cave is vertical until sloughing is initiated after which the angle of draw may approach 70 degrees from the horizontal, particularly at the end of a block. Source: Fleshman and Dale
• Preliminary design of a block cave mine should assume a potential subsidence zone of 45- degrees from bottom of the lowest mining level. Although it is unlikely that actual subsidence will extend to this limit, there is a high probability that tension cracking will result in damage to underground structures (such as a shaft) developed within this zone. Source: Scott McIntosh
• In hard rock mines employing backfill, any subsidence that may occur is always vertical and nothing will promote side sloughing of the cave (even drill and blast). Source: Jack de la Vergne Rockbursts
• 75% of rockbursts occur within 45 minutes after blasting. Source: Swanson and Sines
• Seismic events may be the result of the reactivation of old faults by a new stress regime. By Mohr-Coulomb analysis, faults dipping at 30 degrees are the most susceptible; near vertical faults are the safest. Source: Asmis and Lee
• In burst prone ground, top sills are advanced simultaneously in a chevron (‘V’) pattern. Outboard sills are advanced in the stress shadow of the leading sill with a lag distance of 24 feet. Source: Luc Beauchamp
3. Tricks of the Trade
• For development openings (drifts cross cuts, pump stations, etc.), excessive stress concentrations are to be avoided. Hence, brows should be arched to the shape they would otherwise tend to work to. Source: Coates and Dickhout
• When discs occur in the core of diamond drill holes, the thinner the discs, the higher the ground stress. Source: Allan Moss
• Where discing occurs in diamond drill core, there should be no RQD penalty. Discing is a stress phenomenon unrelated to rock fabric. Source: Phil Oliver
• Caution should prevail when proposing a central rib pillar to provide integral support for a stoping area. The rib pillar may provide a home for wayward stresses and cause serious problems. Source: Fritz Prugger
• In hard rock mines, the most effective ground support systems are those that are installed as soon as possible after excavation of the opening. Source: Kevin Cassidy
• The application of empirical rules requires rounding to obtain a standard length and pattern for rock bolts. If the calculated length is rounded up to the nearest standard length, the calculated spacing should also be rounded up to a standard pattern, and vice-versa. Source: Mike Gray
• It is extremely difficult to fully spin-in an eight-foot (2.4m) resin rock bolt with a jackleg or stoper drill. Where a jumbo drill cannot be employed, consideration should be given to specifying shorter bolts on a tighter pattern. Source: Doug McWhirter
• At intersections of wide headings and wide back man-entry stopes, long cable bolts on a wide pattern support the span while shorter rock bolts installed between them prevent raveling and wedge development. The empirical rules used for length and pattern of rock bolts may also be applied to the cable bolts. Source: Mike Gray
• In slabby ground, straps may be a more effective support than screen. These straps should be placed between rock bolts and run across the layering. Straps placed parallel to the plane of weakness are generally a waste of money. Source: Hoek and Wood
• The long axis of a rectangular shaft should be oriented perpendicular (normal) to the strike of the ore body. Source: Ron Haflidson
• The long axis of a vertical rectangular shaft should be oriented perpendicular (normal) to the bedding planes or pronounced schistocity, if they are near vertical. Source: R. K G. Morrison
• The long axis of a rectangular shaft should be oriented normal to regional tectonic stress and/or rock foliation. Source: Jack Morris
• When designing a large opening underground it may be better to have it wider than longer, i.e. equi-dimensional, because (theoretically) the compressive ring stress around a spherical opening (in an elastic medium) is only three-quarters that around a cylindrical opening. Source: Jack Spalding
• Reinforcing steel applied to draw points in a block caving operation has been found to be a detriment rather than help. Failure results sooner than with plain concrete. The reason is believed to be that percussion caused by secondary blasting causes resonance within the reinforcement that tends to spring it free from the concrete. The stress concentrations encountered are mainly compressive and steel is not an economical material to withstand compression. Source: Hannifan and Hill
• The fundamental distinction between underground mining methods is between those that employ pillars in ore and those that seek complete extraction in the first pass. In burst-prone ground the latter is essential to reduce the incidence of rockbursts. Source: RKG Morrison
• In deep mines, the aim of all stope planning should be complete recovery. Source: Jack Spalding
• The first goal of research and development is surely to achieve success in the prevention of rockbursts, not their prediction. Source: A. van Z. Brick
• In burst-prone stopes, it can take up to five hours to reach stress equilibrium after a blast. For these conditions, stoping is carried out on a single shift basis. Source: E.L. Corp
• In burst-prone ground, if failure of a rib pillar occurs in the first or second cut, it can be violent, whereas if it occurs after the third cut it is generally a gentle, yielding-type failure. Source: Singh and Hedley
• When sinking a shaft full-face, the best way to reduce bottom heave in burst-prone ground is to approach the ideal hemispherical shape by making the bottom dish-shaped. Once the required shape is initiated, it is easily preserved in subsequent rounds. Source: Jack Spalding
• Inclusions, such as dykes or sills are typically more brittle (stiffer) than the country rock. They attract stress and, as has been well documented, can be the source of more and bigger rockbursts. Conversely, a soft formation or fault sheds its stress and driving a heading or mining towards it can cause the face to suddenly explode. Source: Coates and Dickhout
• When it is required to drive a lateral heading to a completed shaft or crusher room, it is preferable to start driving from the shaft or room towards the advancing drift rather than hole the drift directly into it. In this manner, the safest intersection of all is made – the straight line holing. In bad ground, both headings should have extra ground support as the intersection is approached. Source: Jack Spalding
• When new loose soon develops on ground that was previously scaled to solid, it is a sign of high stress and an omen for rock bursts. Source: Merv Dickhout
• In burst-prone ground, big loose should be blasted down. The loose may be restraining rock behind that is already strained to the limit and will burst. Really big loose found anywhere in a mine should be either pinned in place or blasted down, not scaled or barred down. Source: Bob Dengler
• To help avoid rockbursts, headings and faces should be advanced continuously and the ventilation and air temperature should be kept constant, even during shutdowns necessary for holidays, etc. Source: Jack Spalding
4. The Role of Rock Mechanics
“In this case (Thames Tunnel), an excavation method to suit the ground conditions is what is required. Altering the ground to satisfy a method is not practical here.” Sir Marc Brunel 1830
A conflict remains between a “traditional” and a “progressive” concept of rock mechanics. The traditional concept says that rock mechanics should facilitate practiced mining procedures (after the fact). The progressive concept says rock mechanics should guide determination of all the procedures in the first place. Based on the traditional (fragmented) concept, the role of rock mechanics may be categorized as follows.
• Location, support, and protective pillars for mine entries such as shafts.
• Dimension, support, and geometry of mine development headings.
• Size, pattern, and orientation of stopes and stope pillars.
• Sequence and timing of extraction.
• Prediction of backfill performance.
• Procedures at mine closure.
The subsequent contents of this chapter are intended to facilitate understanding with a progressive (total) approach. For this purpose, rock mechanics is discussed under the following headings.
• Ground Stress (Section 2.5)
• Ground Control (Section 2.6)
• Stability of Excavations (Section 2.7)
• Rockbursts (Section 2.8)
(Backfill is dealt with separately in Chapter 21 of this handbook.)
5. Ground Stress
Ground stress is the pressure to which a rock mass is subjected. All types of ground stress may be described, but each may be characterized as either a virgin stress or an induced stress.
5.1 Virgin (in situ) Ground Stress
The natural stress that exists in a rock mass before it is subjected to mining is referred to as a “virgin,” “primitive,” or “in situ” stress. For purposes of analysis, the vertical and horizontal components are considered separately. The vertical stress component is simply calculated by elastic analysis for any particular depth of proposed mining, while the horizontal stress is not.
Vertical Stress
In the hard rocks of the world, the vertical stress is typically a straight-line function of the weight of the column of rock lying above the depth in question. The following formula is true for typical quartz and feldspar rich hard rocks with a specific gravity of 2.65.
Average vertical stress gradient = 2.65/102 = 0.0260 MPa/m of depth, or
= 2.65 x 62.44/144 = 1.15 psi/foot of depth
Source: Dr. G. Herget
In rock mechanics literature, this vertical stress may be referred to as the “gravitational stress,” “lithological stress,” or “overburden stress.”
Horizontal Stress
Determining the natural horizontal stress is more difficult. Horizontal stress is larger than predicted by simple elastic analysis, and it is not equal in each direction. Furthermore, horizontal stress is site specific. Generally, it is considered that the horizontal stress is a maximum in one direction and decreases to a minimum in the direction orientated at 90 degrees. The maximum is referred to in the literature as the “principal” or “major” stress (s1), and its orientation as the “major axis.” The minimum, normal (90 degrees) to the maximum, is typically referred to as the “minor” stress, (s2).
Residual Stress
The intensity and orientation of virgin stress may be significantly altered when the mine is situated near a major fault line in the earth’s crust. An example is the former Lucky Friday Mine in Idaho located near the Osburn Fault, which is reported to have a displacement (slip) of 15 miles. This type of virgin stress is classed as one sort of tectonic stress. If the cause of a tectonic stress is later relieved by force of nature, a portion of the stress remains in the rock and is referred to as residual stress.
Industry Notes
The following notes (accredited to experts) are selected and arranged in a sequence designed facilitate understanding of the horizontal stress phenomenon. The notes start with general observations, then discuss a large regional district (Canadian Shield), and finally refer to a specific mining location within the district.
• By elastic theory, the horizontal stress ought to be near one-third the vertical stress. In fact, it is always much higher than this. In the absence of in-situ stress measurements or other indications of a higher horizontal stress, the most reasonable assumption (for open pit slope analysis) is that the horizontal stress is equal to the vertical stress. Source: Richard Call
• High horizontal stresses are a worldwide phenomenon. At a depth of 1,500 feet (450m) in the earth’s crust, horizontal stresses exceed the vertical stress. Source: Z.T. Bieniawski
• The maximum thickness of the North American glacial ice sheets exceeded 3 kilometers. It can be demonstrated that if 20% of the vertical strain is viscoelastic, then 60% of the horizontal stress may remain. Thus, with each advance and retreat of the glaciers (Nebraskan, Kansan, Illinoian and Wisconsin), residual stresses accumulated. This, along with the previous (Jurassic Age) 3.5 kilometers of uplift and erosion, accounts for the today’s high horizontal stresses. (A similar scenario may be responsible for very high horizontal stresses found elsewhere in the temperate zones of both the North and South hemisphere.) Source: Asmis and Lee
• The azimuth (orientation) of the principal stress is not necessarily uniform with depth; frequently, it has been found to rotate horizontally. One reason that oval or elliptical shafts should not be considered is the horizontal rotation; two others are cost and schedule. Source: Jim Redpath
• Some stress measurements taken underground indicate a very high ratio between the principal stress and minor stress. Invariably, it is found that these measurements were taken (by necessity) near mining operations, the induced stresses from which make the readings suspect. In addition, it is likely found that measurements in different directions were not taken at the same location underground. Sources: Wilson Blake and S. G. A. Bergman
• In the typical rocks of the Canadian Shield, the horizontal stresses exceed the vertical for any depths of mining now contemplated. The ratio decreases with depth until at approximately 750m (2,300 feet) the ratio is approximately 1:1.5, and at a depth of 3,000m (10,000 feet) the average horizontal stress is equal to the vertical stress. Source: Dr. G. Herget
• In the typical Precambrian rocks of the Canadian Shield, the principal stress (maximum horizontal component) exceeds the minor stress (minimum horizontal component) by 20 - 30%. Various Sources, including Asmis, Lee, Herget, Pahl, Oliver, Haimson, Sbar and Sykes
• In the hard rocks of the Shield, the azimuth of the principal stress typically trends between East and NNE. Source: anonymous
• The average horizontal stress in the Shield can be determined as a function of depth by two straight-line relationships with a break at 900m, as follows:
Average horizontal stress gradient (0-900m) = 9.86 MPa + 0.0371 MPa/m
Average horizontal stress gradient (900-2200m) = 33.41 MPa + 0.011 MPa/m
Here it should be noted that the horizontal stresses measured at Timmins and Sudbury in Ontario are significantly higher than the average for the Shield. For example, the stress gradients in the Sudbury basin have been described as follows:
Average vertical stress gradient = 0.0294 MPa/m (1.30 psi/foot)
Major horizontal stress gradient = 10.86 MPa +0.0419 MPa/m (1575 psi +1.85 psi/foot)
Minor horizontal stress gradient = 70% of major horizontal stress gradient
Source: Dr. G. Herget and Allan Moss
5.2 Induced Ground Stress
Two separate classifications of induced ground stress exist. The first is static, similar to virgin ground stress in that it is relatively stationary. The second is dynamic and moves through the rock mass at the speed of sound. Dynamic stress is normally referred to as a seismic stress.
Static Stress
When the miner excavates an opening underground, the stresses in the rock surrounding the void increase due to the fact that he has put a hole in the virgin stress field. The increase is called an “induced” stress. The zone of increased stress in the rock surrounding the cavity is often termed a “ring” stress.
The induced stress is best demonstrated by considering the case of a vertical shaft. By elastic analysis, the ring stress at the skin of the shaft wall is double the horizontal ground stress that existed before the shaft was sunk, and then decays exponentially into the wall rock.
The magnitude of the additional induced stress (Qi) at the skin and any distance into the wall rock
may be calculated by simple elastic analysis with the Spalding formula.
Qi = Q[r2/(r + y)2]
Q is the original virgin ground stress, r is the radius of the shaft, and y is the distance into the wall rock (at the skin, y =0, therefore, Qi = Q).
This theory is close to the truth if the shaft is round and perfectly smooth (i.e. a drilled shaft), regardless of the shaft diameter. It is also virtually true for a shaft pilot hole that was diamond drilled, even though its diameter is many times smaller than the shaft. Additionally, the theory is the case for the blastholes in the shaft bottom, whether drilled by a plugger or jumbo.
In hard rock, the magnitude of this stress at the skin of the shaft wall is actually closer to zero if the shaft is conventionally drilled and blasted. In this case, the horizontal stress is only increased by about 50%, and this maximum increase in stress is found at some distance into the wall rock (i.e. in the shaft pillar). The distance to this maximum stress depends on and is roughly proportional to the shaft diameter. The shaft wall between the skin and the zone of maximum stress is referred to in the literature as “ the relaxed zone.” In this area, the differential stress between the skin and the zone of maximum stress is believed to be the cause of the cracks that tend to form in a ring around the shaft perimeter. These cracks may be entirely independent of the rock formation and can be found to cut across natural slips, joints, and contacts between different rock types. The cracks may be visible when a cap lamp is shone into a rock bolt hole drilled into the shaft wall beneath the concrete lining.
The induced stresses around shafts can be illustrated as follows.
Figure 2-1 Induced Stresses around Shafts
Seismic Stress
An event, such as a round being shot or a fault slip, produces a seismic response that is an alteration in pressure, stress, particle displacement, and particle velocity, propagated in waves through an elastic medium. In rock mechanics, the rock mass is the medium and the significant parameters of a seismic response are stress alteration and noise.
A hammer blow to the end of an intact piece of drill core demonstrates the classic illustration of a seismic response. The blow generates waves, the principle of which is the P wave that can be described as a compressive stress that travels longitudinally to the opposite end of the core where a small portion escapes to create an airwave of noise. The greater portion of the stress is reflected backwards as a tensile stress wave. Since the rock core is strong in compression, the initial wave has no physical effect, but the core is likely to crack or break near the far end at the first point where its tensile strength is exceeded by the returning wave.
Miners tend to associate the word “seismic” with earthquakes. There is a similarity between the seismic response from an earthquake and a seismic event in a mine, but today they are considered to represent different phenomena as will be described later in the text.
6. Ground Control
Ground is controlled in the first instance by addressing attention to both stress and strain. In a hard rock mine, strain (movement resulting from stress) is minimal (there are exceptions) and so control in the first instance is primarily concerned about dealing with stress. Six methods exist to manage stress and accomplish control.
• Avoidance (heading location and alignment)
• Alteration (tensile to compressive)
• Reduction (protective pillars)
• Resistance (ground support)
• Displacement (“chase it away”)
• Isolation (“keep it away”)
Some ground control techniques serve more than one of the above functions. For example, a rock bolt may provide for alteration of, and resistance to, ground stress.
Avoidance
Stress is avoided in the first place by aligning entries, headings, and boreholes to miss treacherous fault zones, dykes, sills, old workings, and zones of subsidence by a wide margin. When a problem fault must be traversed, the heading is aligned to meet it at near a right angle, rather than obliquely. Stress concentration is avoided by rounding the corners in a rectangular heading.
Alteration
Tensile and bending stresses are altered to compressive stresses when the back of a heading is arched. The same is true of a shaft or raise that is changed from a rectangular to a circular crosssection. Conversely, the ability of the rock mass to resist shear, tensile, and bending stress is altered for the better when a cable bolt is tensioned because the friction in joints and fractures is increased. It is altered for the worse when an ungrouted exploration diamond drill hole provides a conduit for ground water to reach a rock mass that was otherwise dry. The water lubricates joints and fractures in the rock and may destroy cohesion at lamination contacts. Friction in joints and fractures may be restored if boreholes divert the flow and drain wet ground or injection grouting cuts off the flow.
Reduction
The ground stress around one heading arising from its proximity to another opening is reduced by a protective pillar (safe distance) between them. The magnitude of the ring stress is reduced (and displaced) if a circular shaft or raise is advanced by drilling and blasting instead of raiseboring. Controlled (“smooth wall”) blasting techniques are used to minimize over break and crack propagation; however, their introduction to highly stressed ground may have another, negative effect (ring stress concentration). To reduce stress in deep shaft sinking, it is typical that smooth wall blasting is abandoned near the horizon where discs were first observed in the pilot hole drill core.
Many rules of thumb relate to the safe distance between excavations underground. The rules generally relate to the equivalent diameter of one of the two headings (i.e. 1½ diameters, 2 diameters, or 3 diameters). A logical way to determine the safe distance for good ground in a hard rock mine is to consider the calculated decay of the induced stress around the opening by simple elastic analysis. For a typical application in good ground, it can be assumed that the minimum distance is that where the induced stress from one hole is significantly reduced at the other. When these calculations are completed for a variety of situations, they demonstrate first that the larger sized heading or room governs the required distance (pillar dimension) and secondly that no rule of thumb is applicable, except by coincidence. (If one of the excavations is an ore pass, waste pass, or bin, the distance calculated should be increased to account for wear and sloughing.)
Resistance
Stresses are resisted with ground support. The support may consist of sets (wood or steel), rock bolts, cable bolts, shotcrete, screen, strapping, or concrete. Ground support is commonly evaluated for comparison purposes by the average pressure that it is calculated to exert against the rock face. Poorly blocked sets provide a resistance that is near zero. Properly blocked sets of wood or steel provide resistance in the order of 10 psi (70 kPa). So do pattern rock bolts if fully tensioned by slight deformation of the rock. Even a thin layer of shotcrete takes advantage of atmospheric pressure to provide approximately 15 psi (100 kPa) at the slightest dilation of the rock mass. The normal concrete lining in a circular shaft has the strength to provide in the order of 300 psi (2 MPa) of resistance; however, its role in hard rock is normally a passive one and the resistance that concrete lining is required to provide is typically near zero. (Refer to Chapter 9 - Shaft Design.) Screen and strapping also provide minimal resistance; their role is to control loose. When a screen becomes filled with loose, it forms a catenary that exerts “back pressure” to inhibit further raveling.
Table 2-1 compares resistance values (maximum theoretical ground support in a direction normal to the face) and other data for some different ground support mechanisms in a typical mining application.
Table 2-1 Resistance Value Comparison
The support tool most employed in hard rock mines is the rock bolt. The traditional mechanical rock bolt (“point anchor” with an expansion shell) has been largely replaced in hard rock mines with resin bonded steel bolts for permanent headings and friction (split-set) bolts for temporary (6+ months) support. Other types of rock bolts see occasional application and cement grouts or
cartridges have been used instead of resin cartridges.
The traditional mechanical bolt was torqued at installation to 50-60% of the tensile strength of the bolt and it was considered that this tension had to be maintained for the bolt to remain effective. These bolts were sometimes very difficult to install properly. Over a period of time, the mechanical anchor was subject to creep in the hole resulting in loss of tension and a maintenance chore to retorque or replace bolts.
Problems with mechanical point anchors in weak rock led to the development of the resin rock bolt. Initially, a single resin cartridge was used to end anchor the bolt. But as the “beam building” theory gained acceptance, the resin rock bolt was fully bonded with several cartridges. In beam theory, the high shear resistance of a fully bonded anchor is an asset. By using resin cartridges of different setting speeds, a resin rock bolt may be tensioned and still maintain the benefits of full column bonding.
This tensioning is not always necessary. A simple calculation demonstrates that a ground dilation (strain) in the order of one-eighth of an inch (3 mm) is more than is required to fully tension a resin bolt. Current belief is that the strongest steel is the most important aspect of a resin rock bolt. For this purpose, the steel bolt now employed is a high-strength rebar or ultra-high-tensile rope thread steel thread bar (Dywidag@). A resin bolt also facilitates mechanized installation. For these and other reasons, the resin bolt has become a mainstay in hard rock mines.
Resin bolts are sometimes even specified for temporary applications when a friction bolt would be less expensive and simpler to install. The friction bolt (“split-set”) was invented by a rock mechanic (Dr. James Scott) who slit one side of a thin-shelled tube so that it could be driven into a drilled hole of slightly less diameter to obtain a “bond” by friction and without the necessity of resin or cement.
Unfortunately, the thin shell is subject to corrosion in the long term and the friction bond is only a fraction of that obtained by a resin bolt. Other types of thin-shell friction bolts (elliptical and Swellex®) later developed suffer the same problems and are not as popular in North America. Any type of rock bolt employed will be most effective if installed soon after new ground is exposed (before the rock has opportunity to fully dilate).
Displacement
A pyramid (Christmas tree) or inverted ‘V’ sequence of stoping can displace stresses from zones weakened by mining activity. First developed for narrow vein mining in burst prone ground, it is now regularly employed in bulk mining of massive ore bodies.
“De-stressing” displaces stress away from the walls of an entry or heading and into country rock. When properly executed, de-stressing creates a failure envelope that shunts stress to the far field stress regime. It is a valuable weapon for combating rockbursts. The evolution of de-stressing is described in the following four quotations.
“When we were planning to mine a shaft pillar at the Lakeshore, it was decided to drill eight holes into the shaft walls. They were loaded with powder and blasted in the hope that they would relieve the stress. No burst occurred (when the pillar was subsequently mined out).”
Source: T. Ramsay (1939)
“The skin stresses on a mine opening can be appreciably reduced by inducing above normal compressive stresses at moderate depth into the walls, and directed tangentially to the walls. The method might well receive its first application in the prevention of rockbursts.” Source: Dr. John Reed (1956)
“De-stress holes for the shaft sinking in burst-prone ground were drilled ahead, into the walls in a direction normal to the axis of the principal stress, on each side of the shaft. Then, the bottom half of each was loaded with powder and sprung. We tried one hole on each side of the shaft and it didn’t help. Then, we tried two holes on each side. It worked!” Source: Phil Oliver (1982)
“At a depth of 7,500 feet, a pillarless mining sequence is employed where no permanent sill/crown pillars are created. A narrow pillar width is employed for bottom sill development and these temporary pillars are de-stressed with the advance of the headings that are also de-stressed (ahead). A de-stress core is established on each level. After it is in place, panel widths may increase to a set maximum length. Center-out mining is employed in an attempt to disperse stresses from the mining area to the far field regime. This also allows mining to proceed in the stress shadow of the previous stope (overhead). Experience and numerical models are used to establish bock dimensions. The maximum sill pillar width is 20 feet for 15 feet wide sills. The
maximum panel strike length is 50 feet and the maximum panel width is 35 feet for the de-stress core, and 70 feet once the core is established.”
Denis Thibodeau (1999)
As may be inferred from the last quote, is not only applied to shaft sinking. It is used regularly to advance lateral headings in deep mines. The holes are “looked out” from the face and drilled to a length greater than the next round to be taken. It is also used to de-stress temporary pillars in burstprone ground.
Isolation
In deep mining, perimeter headings may first be driven around a stoping block to avoid wrongful stress transfer and minimize stress buildup in stope ends. At the current South Deep project in South Africa, the shaft pillar at the reef horizon was deliberately mined out before shaft sinking could reach it. It has been proposed (W. F. Bawden) that a ring heading around an existing shaft will isolate it from stresses induced by future mining in the near vicinity.
7. Stability of Excavations
The application of rock mechanics (along with the advent of remote control mucking) has enabled the introduction of open stoping to ore bodies that were previously only mined by tedious cut-andfill methods. This has been accomplished by increasing the hanging wall span that can be exposed in an open stope while controlling the increased tendency for dilution. This important evolution has enabled greater mechanization and made economic recovery possible from ore zones that might
otherwise have been abandoned.
The requirement for large spans has been met with an empirical analysis of structural stability that is dependent upon rock classification systems that consider strength, joint spacing, joint inclination, moisture conditions, etc. The classification systems used today largely evolved from a crude system proposed by Karl Terzaghi in 1946. Classification systems are continually undergoing modifications to make them more effective. The two most commonly employed up until a few years ago were modifications of Bieniawski’s “RMR” (rock mass rating) system and Barton’s “Q” system. Typically, values calculated by both systems were listed in geotechnical reports.
Both the RMR and Q systems had shortcomings. The RMR system did not consider a value for the rock stress as a parameter and the Q system did not consider joint orientation or rock strength / stress. Dr. Ken Matthews developed a modified system in which Q was simplified (ESR eliminated) and became a sub-classification called Q prime (Q’). He added a factor to account for stress condition and another to account for joint orientation that were included in the computation for his new “stability number,” N (that replaced Q). Instead of correlating it to span, he used shape (hydraulic radius) and a plot of results was termed a stability graph.
Later, after analyzing data from 175 mines, Dr. Yves Potvin modified the N system to the N Prime (N’) system and published a modified stability graph in 1988. His work took better account of joint orientation, eliminated Jw (most stopes are dry) and introduced a sliding mode of failure. The sliding concept has since been modified and re-interpreted (Pakalnis, Clark, Bawden, Hadjigeorgiou, Leclair, Potvin, Neumann, and others) to better reflect a gravity condition. The N Prime stability graph has become popular partly because it is readily adaptable to cable bolting and software is available to facilitate its use. Mines using N Prime can also plot conditions actually encountered.
This evolves into to a customized graph specific to the mine. It may be used to accurately determine safe openings for a variety of shapes and conditions found anywhere underground.
RQD System
One component of the Q and N systems and their modifications is the RQD of a core sample. This is a simple rating first derived by D. U. Deere in 1964.
RQD = S lengths of intact pieces of core > 4 inches/ Length of core advance
Example
Determine the RQD in the following case.
Facts:
1. A diamond drill advance is measured at 60 inches
2. Core lengths of pieces 4 inches or greater total 36 inches
Solution: RQD = 36/60 = 0.6
Note
Dr. Deere meant the RQD to be determined from a core about 2 inches (50 mm) in diameter. A value calculated from AQ or BQ core from a particular rock may be lower than one measured from typical (NQ or CHD 76) core from the same location.
Q System
Following are formulas used in this system.
SRF =0.5 (low stress), SRF = 1.0 (medium stress), SRF =2.0 (high stress)
ESR =±4 for temporary mine openings, ESR = 1.6 for permanent mine openings
De = 2.32Q0.37
Example
Find the maximum unsupported span for a permanent underground excavation with the following characteristics (Table 2-2).
Table 2-2 Table of Observations (Facts)
Solution: Q = (80 x 3 x 0.33)/(4 x 4 x 1) = 4.95 De = 2.32 x 4.950.37 = 4.2
Maximum unsupported span = De x ESR = 1.6 x 4.2 = 6.7m (22 feet)
N System
N = Q’ x A x B, in which
Q’ = RQD . Jr . Jw/Jn . Ja
A = Strength Factor
B = Joint Orientation Factor
N’ System
N’ = Q” x A x B x C, in which
Q” = RQD . Jr/Jn .Ja
A = Strength Factor
B = Joint Orientation Factor
C = Gravity Influence Factor
8. Rockbursts
Rockbursts refer to a sudden implosion that may occur in highly stressed rock. Rockbursts commonly occur on a small scale (“air blast”) where small particles of wall rock “spit” and larger particles bang (“pop”). Less frequently, but more dangerously, slabs of rock are blown from the wall rock. Occasionally, an entire mining zone suddenly fails (“bumps”) in what is termed a major seismic event. Major failures can be terribly harmful sometimes resulting in multiple fatalities and mine closures.
Today the term “rockburst” is commonly defined as an energy release resulting in more than 5 tonnes of rock coming down in an underground opening. A release of smaller scale is referred to as an event or simply “noise.”
Measure of Stiffness
Rockbursts are believed to be caused by high-ground stress in hard rock. Hard rock is described in literature as “crystalline,” “clastic,” or “elastic” rock [as opposed to “plastic” rock that tends to squeeze (creep) rather than burst when stressed to the yield point]. Hard rock may be described as being brittle or “stiff.” The measure of stiffness is Young’s modulus of elasticity, E.
“All rock at depth is in a state of compression and awaits an opportunity to expand. The pressures encountered in deep mining are so great that the potential energy locked up is enormous. Rock suddenly released of stress shatters itself and this is what makes the failure so explosive and is responsible for the term “rockburst.” Jack Spalding
“Brittle failure is said to occur when the ability of the rock to resist load decreases with increasing deformation. Brittle failure is often associated with little or no deformation before failure and, depending on the test conditions, may occur suddenly and catastrophically. Rockbursts in deep hard rock mines provide graphic illustration of the phenomenon of explosive brittle fracture.” Source: Hoek and Brown
“Larger seismic events (rockbursts) are associated with major known geological structures such as dykes and faults.” Spottiswoode and McGarr
“The surface of the discrete feature (discontinuity such as a dyke intrusion, small local fault, geological contact, etc.) is rough and/or has ‘locking-up’ geometric aspirates, thus resisting slippage and storing large amounts of energy in the process. When slippage occurs (either because the driving stress has become higher than the shear strength of the feature, or the perpendicular clamping stress has been removed), it can instantly release very large amounts of energy.” Simser and Andrieux
These classic descriptions of the cause of rockbursts are not satisfactory when the following four facts are considered.
• Rockbursts are known to occur at stress levels well beneath the yield strength.
• Rockbursts are known to occur in mines where the rock is far from brittle.
• Rockbursts are known to occur without the presence of geological discontinuities.
• Seismic recordings of rockbursts often do not produce the saw-tooth curve that indicates the stick-slip behavior that is characteristic the fault-related earthquake.
Elastic Instability
A plausible solution to the paradox lies in the Strength of Materials science, where it is long recognized that a slender column, plate, or shell may fail by buckling long before yield stress is reached. This phenomenon is not confined to brittle materials; it also occurs in ductile steel. “Local buckling” is the term used when an element of a structure fails in this manner. When a significant portion of the whole structure fails instantly and catastrophically, the term used is “general instability.” General instability failure is characterized by abrupt and violent collapse accompanied by instantaneous release of energy, one component of which is a loud noise resembling a thunderclap. The stress at which failure occurs is not predictable, only the minimum stress level at which it can occur can be determined and used as a criteria for safe design (Timoshenko’s Theory of
Elastic Instability).
Timoshenko’s theory explains how a sliver of wall rock (or fault surface rock) exfoliated by a seismic wave (or tension resulting from cooling) can fail (local buckling) with an “air blast” at a stress level well below yield. It also provides an explanation of why a “bump” (general instability) is unpredictable and why it may even take place in ground that has exhibited significant plastic deformation, such as may occur when mining in salt domes.
The results of micro-seismic tests reported by the USBM in 1945 appear to support the concept of local buckling. Table 2-3 lists the lowest percentage of total compressive strength at which a microseismic event was first detected for some different types of common rocks subjected to compression.
Table 2-3 Compressive Strength
The theory of elastic instability is well understood by specialists in the design of pressure vessels, off-shore drill platforms, and hydrostatic linings for drilled shafts; however, it appears to have escaped serious consideration by geoscientists, in particular the concept of general instability.
Shallow Rockbursts
While rockbursts most often happen deep in a mine, they can occur in shallow workings. For example, rockbursts have occurred the tunnels driven under New York City. They can even occur in open pits, such as the bottom heaves that occurred in the Dumfries Quarry on the Niagara escarpment.
Screen
Just as insulation on wiring will not protect it from a lightning strike, typical ground support (rock bolts, shotcrete, timber, or concrete) is not effective by itself to prevent rockbursts or major seismic events. Screen is valuable because it can contain flying rock from air blasts. The screen is more effective when covered with shotcrete.
Procedures
Seven procedures are now employed to deal with rockbursts (reduce frequency and severity). The first three are directed at ground stress. One is isolation from ground stress, the second is alteration of the stress field to advantage, and the third is designing and cutting excavations to minimize the effects of stress. These three procedures have been described previously in this chapter.
• The fourth is simply to wait for ground to stabilize after a blast before allowing man entry to an advancing stope face or heading.
• The fifth is to design stope blasts to induce a rockburst simultaneous with the explosion, thus restoring ground stability.
• The sixth is to induce complete failure of a pillar around a heading or a support pillar in a stope, thus rendering it incapable of carrying high loads.
• The seventh is seismic monitoring; an advanced science that is difficult to describe in lay terms.
Monitoring Ground Stress
Seismic tools are valuable to monitor the ground stress regime, but they do not prevent or predict rockbursts. The tools’ role is to provide an essential device to implement the total rock mechanics program aimed at avoiding or mitigating problems with ground stability.
Prediction
Miners know that when spitting and air blasts become frequent events (“the ground starts talking”), and loose develops in ground that was previously scaled to solid, it is time to beware. (The Russians have noticed that drill cuttings much coarser than usual are an omen.)
Seismic Monitoring System
At a hard rock mine with a high stress regime, a seismic monitoring system is installed to apply science to the miners’ observations. The system monitors the distribution of rock noise due to rock failure as a result of stress changes caused by mining activity. High stresses in the rock mass generate micro-fractures and induce movement along geological structures. The resulting “noise” (acoustic impulses) is detected by receptors placed at strategic locations underground. The classic receptors are geophones connected to a (now) whole waveform MP-250 system originally developed by the USBM. Updated versions are capable of detecting acoustic emissions up to 1,000 hertz (cycles per second). This system is limited to counting impulses and source location to within 10m (by timing), but this is enough to enable man re-entry decisions.
Advanced Systems
More elaborate systems have been developed and have won wide acceptance. These are full waveform or digitized systems that convert voltage functions from “accelerometers” (that replace the geophones) triggered by ground motion into “bits” or numbers, which are then stored on computers and analyzed. This allows for on-line accurate event locations and estimates of source parameters (i.e. seismic energy and stress release) over a broad range of magnitudes (from microseismic to macro-seismic events).
The accelerometers are installed in drilled holes 1¼ inches in diameter. They are individually connected to a junction box on each level. Each junction box is hard-wire connected to a centrally located “multiplexor.” The multiplexor gathers all the signals from each level and sends the data to surface to a surface de-multiplexor via a primary mine entry in two-strands of the mine’s fiber optic cable. The data is then transmitted for processing to a PC that is running the micro-seismic monitoring system for processing.
The more popular of these advanced systems downloads the on-line calculated parameters into a Microsoft Access® seismic database and the results may be viewed on a Microsoft Excel® spreadsheet and AutoCAD® for windows. Mine levels, mine sections, and 3-D drawings created in AutoCAD® or imported from other numerical models are used to view event source locations.
Automatic reporting and plotting of events is also provided on a daily basis or whenever necessary. The program can run on-line showing current activity in real time or be used off line to view data collected over longer periods. This system can also provide information on an off-shift basis through a foreman/supervisor terminal set up in the control room.
Case Histories of Rockbursts
• “The No. 5 West sub-vertical shaft is a five compartment, circular, concrete lined shaft, 22 feet in diameter. At 2:20 PM on the April 26, a seismic event (rockburst) occurred, which caused damage from the 32 Level down, a distance of 45 feet. Slabs of concrete were blown from the shaft wall below the station floor. Five buntons were either bent or torn away from the concrete lining. Guides were bent above and below the level. Severe caving took place in the station.” Source: West
Dreifontein Mine
• “The winze (internal shaft) was ellipsoid in section, almost a circle. It had been lined to within 13 feet of the bench. The concrete lining was 30 inches thick, and the sinking had reached a depth of 5,436 feet. At 4:20 AM, on the 5th of July, during the muck cycle, after 31 buckets had been hoisted, a heavy rockburst occurred that buried everyone in the shaft. Four shaftmen were rescued alive. The bodies of nine others were recovered the next day, by which time two of the rescued had succumbed to their injuries, for a total of 11 dead. A portion of the concrete shaft lining measuring 6 feet by 6 feet had been blown away and above this the concrete was shattered and cracked to a height of 7 feet.” Source: John Taylor & Sons
• “At a depth of about 4,000 feet, a shaft was sunk parallel to a vertical fissure. The latter consisted of a two-foot wide gap, loosely filled with Breccia. Crosscuts from the shaft intercepted the fissure at an angle of 45 degrees. When these crosscuts approached within 20 feet of the fissure, they would blow up for a distance of 30 feet and the heading would become full of broken rock. After mucking out and tight lining with heavy timber sets, before the face met the fissure, some of the headings would blow up again, smashing the timber to match sticks. Obviously, all the ring stress became concentrated in the continually narrow band of rock remaining between the face and the fissure.” Source: Jack
Spalding
• “At another mine, the vein was 6 to 12 feet wide and dipped at 45 degrees. When final stoping was initiated in a direction towards the shaft, the shaft timbers started to break. When 15 broke in one night, mining was discontinued. That next night the shaft blew up and was filled tight with broken rock and timber over a distance of 180 feet.” Source: Jack Spalding
• “In another mine at a depth of 4,500 feet, a fire occurred that burned out stope support timbers. Afterwards, the ground in the vicinity of the shaft started to work. This lasted for ten years during which time shaft timbers had to be replaced frequently.
Then it was decided to drive a new heading in the vicinity of the shaft to provide for a rail loop. This drive initiated a rockburst that blew up the shaft for 150 feet and severely damaged the concrete-lined station. Then ground started to work in the crosscut 100 feet away from the shaft. This burst was not the result of a stress ring failure around the shaft (that was secondary), but rather a complete failure of the shaft pillar. The proof is in the fact that the loop drive was completed, some time later, with no trouble at all.” Source: Jack Spalding