Chapter 32: Blast Evaluation

Blast design is an ever changing process for which success or failure depends on its outcome. The designer has to determine or adjust a number of controllable parameters to achieve a certain objective. In the process, the designer has to consider several uncontrollable parameters (geology and structure, presence of water etc.), which must be quantified in order to predict the results of the blast. A simplified chart of the design process is shown in figure 32.1.

Clearly, blast design requires knowledge of the conditions prior to the blast, determination and control of performance parameters during the blast and assessment of the final outcome. Then, on the basis of the final outcome, modification of the parameters of the blast design may be attempted to change the result. Thus, prior to designing or performing a blast, the blaster must have good quality information on the material of the blast and the neighboring geologic materials, which may be affected by the blast or through which vibrations waves will travel. The geometry of the blast is also important, since vibration, air blast and flyrock may be related to it.

The blast itself is a very fast series of events, detailed knowledge of which can connect the outcomes of the blast to their root cause. An explosive that exhibits unusually low velocity of detonation may be related to poor fragmentation or out of sequence detonations may result in high vibration. Furthermore, supplies and blasting accessories used in the blast are related to the quality and consistency of the blasting output.

A blast has several outcomes, including fragmentation, throw, damage, vibration and air blast, which need to be quantified in order to improve blast design and achieve the objective of optimum cost under compliance and other restrictions. A blast may have several expressed requirements such as a certain contour in the case of tunneling, a certain maximum amount of dilution in a mine or certain minimum and minimum particle sizes for downstream operations. Unfortunately, satisfying all requirements means that all parameters of the blast design as well as results must be accurately known.

Thus, blast design today is closely associated with measurements used to quantify the inputs, outputs and performance parameters of the blast. It is convenient to separate the measurements used in blast evaluation prior to the blast, measurements during the blast and measurements after the blast. Measurements prior to the blast assess the environment of the blast, while measurements during and after the blast are used to evaluate blast performance and outcomes; such measurements may be used as inputs to modify the next blast design or as quality controls of the blast.

PREBLAST ASSESSMENT

Preblast assessment aims at collecting adequate information on the geologic properties, the structure of the geologic formations and the geometry of the blast.

Geology

The general geology of a blast is typically known from previous work using core drilling or surface mapping techniques. From its knowledge one may estimate parameters that affect blast results and come up with a preliminary blast design. However, prediction of fragmentation or damage are impossible, without detailed information. This is often overlooked in practice, where detailed geologic information is considered after problems arise; however there is less uncertainty in selecting blast design parameters, if detailed geologic information is available before the first blast takes place.

Geologic information, necessary in blast design, is the rock type, density, strength, weathering and structure. Rock type provides basic understanding of the material and the field conditions. Rock density gives a first estimate on how easy it is to blast rock; higher density is associated with rocks harder to blast. Rock strength information can be in the form of a general classification, allowing the rock to be placed in a broad category, or a classification using detailed information, i.e. numerical strength properties, such as compressive strength, tensile strength and Young's modulus. One may characterize the rock as stable, very weak, weak, moderately strong or strong, based on its approximate compressive strength to follow the classification by Deere and Miller, which makes use of the compressive strength and Young's modulus (Ormand, 2002). Other classification systems may be appropriate as well and offer more information. For example the Geomechanics Classification System (RMR) incorporates the uniaxial compressive strength of the intact rock, the rock quality designation, the spacing and condition of the discontinuities present, their orientation, the presence of water as well as the weathering of the rock. Similarly, the Q system is based on RQD, number of discontinuity sets, roughness of discontinuities, degree of alteration, water inflow and stress condition. Although the last two systems were developed for estimating support in tunnels, they provide a relative measure of rock structure. As such, they are not directly related to the blasting result and most current blasting models do not explicitly use them. However fragmentation predictions are often linked to a "blastability index," where rock strength as well as rock structure are required parameters. Rock structure information is critical, since many blasting problems are associated with the frequency and orientation of fractures in the rock mass. Rock structure determines the in situ fragment size of the rock, which has to be reduced by blasting, in order to provide the required blast fragmentation. The burden, spacing and the borehole diameter of the blast have to be selected so that blasting can modify the in situ size. Rock structure properties may change markedly throughout the mine or project. Thus, localized and detailed information is critical, often necessitating frequent changes to the blast design parameters. Measurements while drilling, observations with borehole cameras and detailed rock structure characterization, using optical techniques or simple fracture mapping, can provide detailed information on the material of the blast.

Measurements While Drilling

Of particular interest is information collected while drilling, which can assist in the characterization of the rock mass and the optimization of blast design. This technique is commonly referred to as "measurement while drilling" (MWD) and although it has been routinely used in oil and gas drilling since 1970s, it has become popular in mining only recently, mainly due to the use of computers and software for the interpretation of data. Typically several parameters, such as depth, rotation speed, penetration rate, torque, vibration and air pressure are compiled together at a borehole or drilled. These parameters include both performance parameters of the drill and drill settings and they cannot be considered in isolation but as an interlinked set of data, which are interpreted using appropriate algorithms (Segui and Higgins, 2002). The result can be recognition of different rock types as well as knowledge of the structures in the rock. MWD parameters are considered either independent or dependent as described in table 32.1.

MWD Parameters

Table 32.1 – MWD parameters.

MWD parameters are considered (1) independent, if they are controlled by the drill operator; or (2) dependent, when they are controlled by the rock. Typical examples of the former are rotary speed and penetration rate, yet pullback force is an important contribution to the drilling rate.

Calculated parameters are those, which can be calculated from other MWD parameters. Of particular interest to blasting are the specific energy and the blastability index. Typically the specific energy is calculated on the basis of rotation, rotation speed, torque and penetration rate, while blastability index is based on the rock mass description, joints and their orientation, rock density and brittleness. Integrity rates manufacturers of drill monitoring systems offer a calculation of the blastability index based on MWD, however the formulas used are proprietary.

Figure 32.2 shows an example of MWD (Monfared, 2007) that clearly demonstrates the close correlation between vibration, torque and penetration rate while thrust and rotary speed were kept constant.

Mozaffari (2007) has evaluated the effectiveness of using drill performance parameters to understand the variability of the rock vertically (along the borehole) and horizontally (between boreholes). In his study at Aitik mine in Sweden, he found that penetration rate provides the most accurate information, while torque, specific energy and vibration were better for assessing rock conditions vertically than horizontally. He suggested that more consistent information is recovered by keeping thrust and rotary speed constant.

Thus MWD systems can provide a more accurate description of the rock mass variations in the blast block so that loading of individual holes or powder factors may be adjusted. However MWD systems need to be calibrated for the mine site of the application.

Borehole Camera

A borehole camera can be used for visual inspection of lithology and structure. Technological advances of electronic components and sensors have allowed the development of inexpensive units, which can be used under a variety of conditions. Depending on the amount of information required, sample cameras, using small focal length lenses mounted off a stiff cable or pole to control rotation, or sophisticated units that can provide high-resolution 360-degree views and complete analysis of features providing dip, strike, fracture aperture and frequency, can be employed. Such systems employ high quality optics, orientation devices and dedicated software for analysis. Often, it is important to have hole centralizers to be able to maintain clear focus around the borehole. A borehole camera can provide information listed in table 32.2.

Information Provided By a Borehole Camera

Table 32.2 – Information provided by a borehole camera.

Fracture Mapping

There are many techniques that can be used for fracture mapping. The conventional ones are line sampling, semi-trace length sampling and cell mapping. These have been described in the literature and synthesized by Walrych -Copeland and Lilly (2002).

Line mapping consists of stretching a measuring tape along an exposed face and recording the characteristics of the discontinuities that intersect the tape. Three orthogonal sampling directions of equal length are recommended if rock faces are available. In the semi-trace length technique only the discontinuity length above the trace line is counted. In cell mapping, rectangular cells are defined on the exposed faces and discontinuities. Families of discontinuities with similar characteristics are recognized in each cell and their characteristics are recorded. These techniques are labor intensive, require access to the rock exposure, and can be hindered by production constraints.

Thus, these techniques are rarely used for blast analysis. However, techniques based on image analysis have been developed to provide geological face mapping, faster, safer and at a fraction of the cost. Such techniques use a monoscopic or stereoscopic approach (Lemy and Hadjigeorgiou, 2004) depending on whether one or two images of the same face are used to extract information on two or three dimensions. The latter is subsumed under the heading "stereophotogrammetry." The former involves manual digitization of images of faces or automatic recognition of features using algorithms for image processing. Franklin et al. (1997) have used general purpose image processing techniques, such as grey scale thresholding, smoothing and detecting. Reid and Harrison (2000) consider the image as a surface, whose elevation is a function of image brightness, so that the pixels in the ravines indicate discontinuities, and Lemy and Hadjigeorgiou (2004) have used neural networks to improve the reliability of the recognition of features.

Geometry

Blasting applies a constant amount of energy per unit volume of rock to result in the desired fracturing. Typically the burden and spacing control the application of energy. Smaller than desired result results in over fragmentation, and possibility of flyrock and air blast, while large burden results in coarse fragmentation and increased vibrations. However, blast faces are rarely smooth and boreholes may have unwanted deviations, affecting the local burden and powder factor in practice. Prior to loading, it is important for the blaster to know the burden of the blast for every point of every borehole in the pattern, so that remedial action can take place. Geometric information can be obtained using face profiling methods and drilling deviation measurements.

Drilling Deviation

Drilling deviation is a major contributor to blasting problems; deviation from the intended position violates burden and spacing values resulting in low or high powder factors affecting safety, quality and economics of blasting. Drilling deviation may be caused by collaring deviations, caused by wrong position of the drill on the rock surface, angular deviations, caused by wrong initial drilling direction, and trajectory deviations, when the drill hole curves (Persson et al., 1994). The first two are set-up errors related to the positioning, alignment and instability of the rig, while the last is a result of practices and/or unexpected geological conditions. For example, hole deviation tends to increase with increasing feed pressure and with hole length to half diameter ratio. Positioning errors are minimized using Global Positioning System (GPS) technology in open pits so that drills are positioned to within a few centimeters from the desired location. Several drill-monitoring systems have the capability of providing information for accurate position and depth. Thus, set-up error, both in terms of location and angle, can be a minimum.

Trajectory deviations can be large and need to be examined; if in excessive, holes must be redrilled.

Methods of measuring deviations are borrowed from the technology for oil wells. Three types of systems are used for borehole deviation surveys: gyroscopic probes, orientating rods and optical systems are described in table 32.3.

Borehole Deviation Survey Systems

Table 32.3 – Borehole deviation survey systems.

Figure 32.3 shows an application of drilling deviation measurement in conjunction with face profiling. True burden distances can be determined and potential problems due to inadequate or excessive burdens can be recognized and addressed.

Laser Profilers

Laser survey equipment is used for the assessment of faces. It does not require any reflectors, making it easy to take accurate measurements of inaccessible locations and eliminate errors in measurements. A number of laser systems have been introduced to enable accurate three-dimensional representation of faces and fines, accurate determination of burden conditions. In the form of cavity scanners they can be used in determining the exact geometry of inaccessible openings.

There are many types of laser equipment. The simplest provide distance and vertical angle, which means that they are capable of producing two-dimensional profiles only. More versatile units measure distance, horizontal and vertical angles, thus locating the target in the three dimensional space. Motorized scanning units allow measurement of many points almost automatically.

Laser survey equipment sends pulses of low power laser light, which are reflected on the surface of the rock. The coordinates of a point can be calculated if the distance from a known location (where the equipment is located) and two angles are known. The unit it takes for the pulse to travel the distance and back is measured and is converted to distance, once the speed of the pulse is known. Vertical angles are measured by inclinometers or encoders while horizontal angles can be measured using magnetic devices or encoders.

The method of operation in the case of a quarry face is shown in figure 32.4 (MDL, 2009). The face of the blast can be scanned at various resolutions, depending on the required detail. The positions of the boreholes can be scanned by placing a marker at the top of each hole.

Figure 32.5 face profiling (MDL, 2009) shows the surfaces obtained using laser profiling. The burden can be checked against specifications and regions of excessive burden may be blasted using a higher bulk strength explosive, while, burden is insufficient a lower bulk strength explosive or decking may be utilized.

The accuracy of commercial laser profilers depends on the distance of the units from the face as well as on accurate knowledge of the angles. The distance from the face controls the footprint area of the laser target. As distance increases the footprint becomes larger, resulting in less accurate measurements. Typically distance is not critical, as long as the beam can be reflected on the target surface. It is recommended that the user always check manufacturers' specifications. Angles are however affected by how accurate the set up is. According to Thomas (2003) encoders will provide an accuracy of ± 0.02 degrees, resulting in an error of 0.1 meter over a distance of 300 meters. However if the error in leveling were 0.5° the error would be 2.6 meters, which is significant.

In a different configuration laser profilers can be used as cavity or void scanners underground. These units are made for underground application but the principle of operation is the same as before. The principle of measurement is shown in figure 32.6 (MDL, 2009).

Profilers and scanners are versatile tools, which can be used in a variety of applications. Clearly the benefit of knowing the free face condition and burden distance are important in flyrock, air blast and vibration control.

Global Navigation Satellite System

Global Navigation Satellite System (GNSS) includes the US GPS system, the Russian GLONASS system and the European Galileo system. The system has seen rapid development in the last few years. Availability of special frequencies, once limited to military receivers, and development of dual frequency receivers have allowed centimeter accuracy work in surveying applications. GNSS receivers can be used to establish measurement points in open cast mines and quarries. Stand alone GPS units for navigation have an accuracy of 5 meters but this can be further improved if of a reference station, with known coordinates, which uses the same satellites, is used for correction. High precision geodetic quality systems, known as Real Time Kinematic (RTK), which use one of the revolutions signal as a carrier and it is transmitted by the satellites and is re-broadcasted by the reference station, have a practical accuracy of about 2 centimeters.

Such systems can be very useful in surveying for blast design applications as well as in the navigation of drills to each location, drilling to the desired depth and at the desired angle of inclination. Surveying staff drill monitoring systems have incorporated GNSS/GPS in an effort to improve the quality of the drilling operations. GPS data can be used in parallel with laser scanners to provide the location of the base or as stand-alone systems for surveying work.

GNSS/GPS technology requires that the antenna of each receiver must be able to see at least four satellites. Obviously this is not possible for underground applications and it is increasingly difficult in deep open pit mines. Various technical solutions have been developed to address this problem. These include satellite augmentation, use of pseudo-satellites, use of lasers with GPS, blending of control data with GPS data (Zyetal, Vest, King, and Shrole, 2005). The details of these methods are beyond the scope of this manual.

Stereo Photogrammetry System

Stereo photogrammetric imaging can be used to provide precise information on blast geometry in a time efficient manner. Traditional photography suppresses the third dimension. However stereoscopic photogrammetry can recover three-dimensional information by acquiring spatial information from two images showing the same object but taken from different angles (Moser, Gauster, and Gaich, 2007). The principle is shown in figure 32.7.

Traditionally the technique required specialized camera equipment, two cameras and rigid set-up. With modern advances in computer vision the set up is simplified, requiring only one camera, while the computer software determines the relative orientation of the images taken. The main activities of such a system are shown in table 32.4.

Main Activities of a Stereo Photogrammetry System

Table 32.4 – Main activities of a stereo photogrammetry system (Moser et al, 2007).

Most of the above activities are taken care of by the computer software. The user however has to provide a reference line, to which the pattern is connected, a scale and a baseline on the ground plane and two photographs of the face, a known distance apart. Accuracy depends on the scale, the quality of the camera calibration, the quality of the image matching and the relative orientation of the cameras. Moser et al. (2007), have achieved an error of 0.2 cm for an imaging distance of 5 m in the lab. In the field they have suggested that the errors are much larger.

Rock Mass Fracture Characterization

To determine rock mass characterization one needs detailed information about rock structure and features. Field mapping is often a method employed to collect such data; however there can be hazards associated with this operation due to poor accessibility. New surveying technologies involving GPS and laser profilers or photography, as described in the previous sections can be employed. The technique is known as ground based 3-D mapping and involves laser scanning (commonly called LIDAR) or high-resolution photography. Millions of points can be collected to represent the details of the surface in 3-D. Techniques, such as texture or overlay mapping can be used to overlay high-resolution information from digital images on the 3-d surface. Thus structural discontinuity analysis can be performed. The technique has been investigated by Kennedy, Harwood and Turner (2006) and Lato, Diederichs, Hutchison and Harrap (2009). It has been pointed out that the success of the technique depends on the instrument, the size resolution and the processing of the resulting point cloud.

For laser scanners, distance accuracy, position accuracy and beam diameter vary with distance. Manufacturers state these at a certain distance or they provide a formula for their variation with distance. Another issue is the maximum/minimum of the instrument that depends on the reflectivity of the target. According to Kemeny (2008) most systems suitable for use in blasting have a distance recommended by the manufacturer.

Resolution has to be high, although today's laser scanners can collect data at rates over 20,000 points per second with a position accuracy of less than 5 mm at distances up to 800 m (Kemeny, 2008). The points collected are typically filtered on unwanted features (i.e. vegetation) are eliminated. Then the point cloud is presented in the form of a surface by developing a mesh. Surface mesh density is important, since a coarse mesh will not represent detail, so minor edges are lost to smoothing. However a very fine mesh includes a lot of noise. The concept is shown in figure 32.8 where Split-FX software (Split Eng., 2009) was used for mesh generation and feature analysis. Further processing of the mesh is required to extract geological features.

It is important to define limits of angles in which features can be considered co-planar and a maximum deviation between normals of adjacent triangles must be specified by the user (Lato, Diederichs, Hutchison and Harrap, 2009). Finally there needs to be a minimum number of mesh triangles to be able to define a single discontinuity. As the number becomes larger, sets may be lost, while as number becomes smaller, noise dominates the results.

Explosives Loading

Explosives loading is of critical importance. Knowledge of the geometry of the blast is of no value without accurate control of loading since the essence of a proper blast design is to provide the correct amount of explosive for the burden available. The equipment used for borehole loading offers several controls to calculate mass and volume of material used and conduct basic quality control checks.

A density check is a quick test to make sure the explosive is within specifications. Density can be measured by weighing a known volume of material. The volume of any container (i.e. a measuring cup) can be easily measured by weighing the amount of water that fills the container and dividing it with the specific gravity of water (1 gram/centimeter³). Please note that several measurements are necessary in order to obtain meaningful data. A simple statistical analysis can provide the mean value and the standard deviation of the measurement. These can be used to monitor the consistency of the explosive composition.

The volume of explosive delivered to the borehole is also important. Divided by the area of the borehole it provides the length of the charge. Differences between actual and calculated values, assuming that delivery equipment is calibrated and functional, signify the presence of irregularities (gaps in loading or fissures) or diameter changes. Therefore, the depth of each hole, the collar height and the length of each deck should be verified by the blaster and checked against calculations from the loading unit data. Simple measurements at the site and recording of these measurements are critical in ascertaining the quality and consistency of the blast and back analyzing a blast that went wrong.

DURING BLAST MONITORING

During blast monitoring implies measurements taken during the time in which the blast is performed. Such measurements relate to explosives performance parameters, shock and stress wave propagation, and projectile motion.

Velocity of Detonation Monitoring

Velocity of detonation (VOD) monitoring is normally performed to assess explosive detonation parameters in the laboratory. However, in the field, it provides the effect of field conditions on explosives performance. Explosives, when loaded in boreholes, can be affected by environmental conditions (such as water, mud and stemming contamination), blast geometry factors (such as depth of boreholes and proximity of other charges), and operational conditions (such as sleep times, loading method, priming, and quality control), since many explosive mixes may be prepared just prior to loading in a borehole. Velocity of detonation recording may also provide blast diagnostics. Cutoffs, gaps in loading or insufficient priming can be identified. In the case in which multiple holes are monitored, VOD monitoring can provide information about timing.

For measuring velocities of detonation there are two distinct categories of measurement systems: (1) the continuous ones, in which VOD is measured at many points of the explosive column, and (2) the discontinuous ones, in which VOD is measured at a few points.

Continuous VOD Systems

The general principle of a continuous system is illustrated in figure 32.9. Its main advantage over the discontinuous systems is that velocity of detonation (VOD) is measured at many points of the charge allowing intimate knowledge of changes of the detonation. For example, in priming, one is interested in run up distances, while, in the case of decking, a gradual decrease of the velocity of detonation may signify desensitization issues.

The system consists of the measuring probe, a power supply and a data collector, which can be an oscilloscope, a dedicated data acquisition system or a computer with a data acquisition card. The measuring probe type depends on the principle of operation. In any case it will be in the form of a cable that should be inserted in the borehole straight without bends and loops. The power supply and the probe account for the difference between the various systems in use today. Typically the power supply is an integral part of the system and the principle of operation is transparent to the user. The various principles of operation are outlined in the following discussions.

The Continuous Probe Method

The power supply uses a constant current source and the probe a resistance shorted at one end. The resistance probe consists of a resistance wire inserted into a small diameter brass tube in the case of a laboratory system or a resistance wire forming the axis of a co-axial cable. The resistance wire is a nichrome wire having an accurately known linear resistance. In the former case it has a typical resistance around 300 ohms/meter while in the latter the required resistance is around 10 ohms/meter. At detonation, the wire resistance probe is consumed. However the circuit remains closed due to the fact that the detonation head is sufficiently ionized. The current follows Ohm's law.

Therefore, since current is constant, the voltage changes with time is proportional to the resistance. Knowing the full voltage drop across the probe and the length of the probe, the voltage drop can be converted to distance along the charge. Therefore the velocity of detonation can be calculated by interpolating the voltage drop-time record provided by the data acquisition system. A record for a borehole loaded with an AND/Emulsion blend at the bottom and straight ANFO at the top is shown in figure 32.10. One can obtain the velocity of detonation at various locations along the axis of the borehole and examine priming effectiveness, product performance in situ and effect of loading practices. The system can collect data at the maximum rate available by the recording system. Typical recording rates are between 1 MHz and 10 MHz. Since its principle of operation is based on the resistance of the shock wave, this system is only suitable for detonations waves. This type of system cannot measure shock waves in inert materials. Another limitation of the system is the length of the cable that can be used per channel. Although this is typically at the hundreds of meters, it may limit the number of boreholes that can be connected in one channel of VOD recording.

SLIFER System

The SLIFER system was originally developed by Sandia National Laboratories to measure the propagation of shock waves from nuclear explosions. SLIFER stands for Shorted Location Indicator by Frequency of Electrical Resonance. The system utilizes the fact that a shorted length of coaxial cable becomes part of an oscillator circuit, the frequency of which depends on the length of the cable. A radiofrequency oscillator, a frequency-to-voltage converter and a voltage recorder are required for the measurement. Typically the probe is inserted in the borehole and for blasthole applications the shorted end of a 50 ohm or 75 ohm coaxial cable is attached to the primer, which is loaded in the hole. The system can pulse at one (1) megahertz, collecting one point per centimeter. In this way boreholes up to 64 meters in length can be monitored, and since the cable is destroyed by the detonation and not the shock wave, measurements in air and rock can be recorded per channel of data acquisition.

Time Domain Reflectometry

This has been used by Ouchterlony et al. (1996) and Brent and Smith (1999). According to the technique, pressure transducers are placed in monitoring boreholes behind the blast and each hole is plugged at the collar. Pressure transducer reads a few atmospheres with durations up to a few seconds. Figure 32.20 provides a schematic of the measurement.

During the blast the pressure pulses are recorded and interpreted. Positive pulses (overpressure) suggest gas penetration in the borehole. Negative pulses (decreasing pressure) indicate the cracks are opened by the stress wave so that air is sucked into the fractures. Thus, the reason for damage (stress wave, gas) may be identified together with the limits of the damage zone. In some cases, these techniques have been combined with borehole camera assessment to visualize new cracks (Brent and Smith, 1999).

CORRTEX System

The CORRTEX (continuous Reflectometry for Radius versus Time Experiments) system was originally developed by Los Alamos National Laboratories for the measurement of nuclear yield. This system is also suitable for blasting and commercial systems have been developed on the same principle. The probe used is 50 ohm or 75 ohm coaxial cable. The technology on which it is based, Time Domain Reflectometry (TDR), is well known and used in a variety of fields. The system transmits fast rise time electrical pulses, which are reflected at points where discontinuities occur or the impedance of the cable changes. The time between the sending of the pulse and its return is accurately measured. Knowing the velocity of propagation of electromagnetic waves in the cable, this time can be converted to distance. During detonation, the shock wave crushes the cable, producing discontinuities so what the pulse can be reflected. The system has the advantage that it does not need a detonation wave; any shock wave that will crush the cable is sufficient to produce a record. As a result the system can measure shock velocities even in inert materials. Currently the system pulses at a maximum rate of 20 MHz, which provides a limited number of measurements for observation of changes in performance of short charges or close to interfaces.

Discontinuous (Start-Stop) Systems

The general principle of a discontinuous system is shown in figure 32.11. Clearly the distance between the measuring points must be known accurately. The measuring points consist of sensors which can be contact pins, piezoelectric pins, emission pins, fiber optics etc. The systems which have been used today and in the past are described in the following paragraphs.

Start-Stop method

Two probes are placed at a known distance apart in the explosive. Each probe consists of two wires placed in close proximity (contact probes) connected to a voltage source. When the detonation wave contacts each probe, it shorts the circuit by bringing the two wires in contact. By measuring the signals obtained by either a counter or an oscilloscope or a data acquisition system, one can measure the detonation velocity.

Ionization system

A typical ionization system is a multichannel high-speed timer that records time intervals between channels which can be converted to detonation velocity. It relies on the ionized plasma generated inside the detonation front. This plasma sequentially short circuits a series of wire pairs terminating at known positions in the explosive charge. The principle of operation is shown in figure 32.12.

Fiber Optic System

In the fiber optic systems the time intervals for the detonation wave to pass from various locations in the charge are measured from the illumination of successive optical probes. The optical signal is converted into voltage in the unit.

Examples

An example of VOD monitoring in the case of decked charges is shown in figure 32.13. The horizontal axis offers a time scale while the vertical axis is a distance scale. The velocity of detonation of the ANFO charges is within expectations for the 125 millimeter diameter charges. However the delay time of approximately 7 milliseconds between detonations falls short of the 25 millisecond delay used. Apparently a malfunction occurred, resulting in a violation of delay time specifications for this blast. As a result, vibration and possibly air blast levels may be higher.

Pressure Monitoring

Measurement of dynamic pressure is required to characterize the pulse generated by the detonation or the pulse which is propagating inside undisturbed explosives due to the detonation of neighboring charges. The former is related to the performance of the explosive, while the latter is related to malfunction of explosives and accessories. Occasionally, pressure measurements are needed in marine environments if blasting takes place close to them.

In normal blasting applications it is rare to measure detonation pressures inside a borehole. It requires transducers and connecting cable that will survive the event. Consequently, measurements may only be taken at the explosive/stemming interface. Transducers that are normally used for this purpose are polyvinylidene fluoride (PVDF) gauges. PVDF gauges are thin piezoelectric film, which can record pressures higher than 40 gigapascals (GPa). They have a very fast response time, making them ideal for detonation pressure measurements. However they are expensive, must be used with high data acquisition rate instruments and they need high quality connecting lines to maintain signal degradation. PVDF gauges have been used inside boreholes to measure pressure profiles of detonating explosives (Davies, Smith and La Cruz, 1997). However the use of such instrumentation is not common in operating mines. It is rather reserved for research environments.

Knowledge of pressures produced by neighboring charges is of significance to operations suffering from malfunctions of explosives and accessories. Although measurements cannot be considered routine, they are more common than the previous case. Pressures of interest in the case of malfunction are typically below 1 GPa. In this pressure range Carbon Composition Resistor offers an inexpensive way of conducting measurements. The working principle is that the resistance decrease (or conductance increase) is related to the applied pressure. Several equations relating resistance change to pressure have been proposed, depending on the pressure expected. The two best known calibration equations for 470 ohm resistors are equations 32.1 and 32.2.

Weiland's equation (1987) for pressures below 0.1 gigapascals is shown as equation 32.1.

$$P = \left(\frac{R_0}{R}\right)^{1.71}$$ <!-- VERIFIED -->

Equation 32.1

Where:

• P = Pressure (kilobar)
• R₀ = Original resistance of the gauge (ohms)
• R = Resistance at pressure P (ohms)

EXAMPLE 32.1

Calculate the pressure corresponding to a change of resistance from 470 ohms to a final resistance of 430 ohms in the case of a commonly used 470 ohm resistor using equation 32.1.

$$P = \left(\frac{R_0}{R}\right)^{1.71} = \left(\frac{470}{430}\right)^{1.71}$$ <!-- VERIFIED -->

$$P = \left(\frac{470}{430}\right)^{1.71} \times \left(\frac{1}{1}\right)$$ <!-- VERIFIED -->

P = 0.465

The pressure corresponding to this change of resistance is 0.465 kilobars (0.0465 gigapascals), which is in the range of equation 32.1.

The equation by Ginsberg and Asay (1991) for pressures below 5 gigapascals is stated as equation 32.2.

$$P = 2.89 \times 10^{-4} \times \left(\frac{R_0}{R} - 1\right) + 0.945 \times \left(\frac{R_0}{R} - 1\right)^{1.6} \times 10^{-11}$$ <!-- VERIFIED -->

Equation 32.2

Where:

• P = Pressure (gigapascals)
• R₀ = Original resistance of the gauge (ohms)
• R = Resistance at pressure P (ohms)

EXAMPLE 32.2

Calculate the pressure corresponding to a change of resistance for a change of resistance of 200 ohms (from 470 ohms to a final resistance of 270 ohms) using equation 32.2.

$$P = 2.89 \times 10^{-4} \times \left(\frac{470}{270} - 1\right) + 0.945 \times \left(\frac{470}{270} - 1\right)^{1.6} \times 10^{-11}$$ <!-- VERIFIED -->

$$P = 2.89 \times \left(\frac{470 - 270}{270}\right) + 0.945 \times \left(\frac{470 - 270}{270}\right)^{1.6}$$ <!-- VERIFIED -->

P = 0.6

The pressure corresponding to this change of resistance is 0.41 gigapascals or 4.1 kilobars.

Caution Equation 32.2 cannot be used for pressures below 0.5 gigapascals (5 kilobars), while Equation 32.1 is suitable for pressures below 1 gigapascals (1 kilobar).

For this reason an equation has been developed by Katsabanis, which works well for pressures between 0.1 gigapascals and 0.3 gigapascals (1 kilobar to 3 kilobars).

$$P = 0.644 \times \left(\frac{R_0}{R}\right)^{1.6} - 0.6 \times \left(\frac{R_0}{R}\right)^{0.5}$$ <!-- VERIFIED -->

Equation 32.3

Where:

• P = Pressure (megapascals)
• R₀ = Original resistance of the gauge (ohms)
• R = Resistance at pressure P (ohms)

EXAMPLE 32.3

Calculate the pressure corresponding to a change of resistance of 100 ohms (from 470 ohms to a final resistance to 370 ohms) using equation 32.3.

$$P = 0.644 \times \left(\frac{470}{370}\right)^{1.6} - 0.6 \times \left(\frac{470}{370}\right)^{0.5}$$ <!-- VERIFIED -->

$$P = 0.644 \times \left(\frac{470 - 370}{370}\right)^{1.6} + 0.6 \times \left(\frac{470 - 370}{370}\right)^{0.5}$$ <!-- VERIFIED -->

P = 261

The pressure corresponding to this change of resistance is 261 megapascals (0.261 gigapascals or 7.61 kilobars).

Pressure measurements with carbon resistors have been described by Ginsberg and Asay (1991) and Weiland (1993) for the case of measurement to assess the malfunction of commercial explosives in underground coal mines (Liu and Lien et al. (1995) for the case of malfunction of explosives in hard rock.

Pressure measurements in aquatic environments are typically easier since gauges are placed at significant distances away from the blast and can be recovered. In this case commercial piezoelectric gauges are used. Quartz or tourmaline gauges are typically used in these applications.

Stress Wave Measurement

Stress wave measurements are not very common in blast evaluation. Occasionally, there is a need to record stress waves, to associate them to a blasting theory or to examine the effect of initiation methods and timing, to measure the effects of blasting on infrastructure components or to understand damage caused by explosive charges. Stress wave propagation measurement is associated with the measurements of strain or acceleration.

Strain Measurement

Strain can be measured using strain sensors or accelerometers. Strain gauges are based on the change in resistance of a conductor when it is compressed or elongated. Strain gauges are typically foil types, available in a variety of configurations using special adhesive the foils are bonded on the material whose strain is to be measured. The strain gauge is connected to a Wheatstone bridge circuit using a combination of active gauges. The Wheatstone bridge is excited with a DC supply and is zeroed when no stress is applied to the material. When stress is applied, the circuit is unbalanced and the resistance change can be measured and converted into strain. Figure 32.14 shows a uniaxial strain gauge while figure 32.15 provides a schematic of the Wheatstone bridge configuration.

Strain gauge technology is beyond the scope of this Handbook. However, if strain gauges are contemplated for dynamic measurements, the issues highlighted in table 32.5 must be resolved.

Strain Gauge Issues

Table 32.5 – Strain gauge issues.

Accelerometers

High frequency accelerometers have been used to measure stress wave parameters close to the blast. Strains can be calculated from accelerations according to equation 32.4 (This equation is provided as a reference only).

$$a = \frac{dv}{dt} = \frac{d^2s}{dt^2}$$ <!-- VERIFIED -->

Equation 32.4

Where:

• a = Strain (dimensionless)
• s = displacement (millimeters) (inches)
• v = velocity (millimeters/second) (inches/second)
• t = time (seconds)
• $c_p$ = P wave velocity (millimeters) (inch)

Velocity can be estimated by integrating the acceleration record.

High frequency accelerometers must be selected with the application in mind. Close to the borehole frequencies can exceed 10 kilohertz and accelerations can be higher than 10,000g. Depending on the proximity and the size of charge, higher or lower values may be anticipated. In any case monitoring equipment should be able to record at a frequency more than twice the highest frequency of the wave to avoid aliasing.

Accelerometers should not be placed too close to the blast, since this will result in damage to the gauges and destruction of the cables. However measurement should be placed close enough to sample the stress wave before coalescence of waves. A study demonstrating the use of high frequency accelerometers to understand blast damage is presented by Yang et al. (1993) where accelerometers were placed 2 m (6.6 feet), 4 m (13.1 feet), 6 m (19.7 feet) 10 m and 12.3 feet) away from a 100 mm (3.9 inch) hole in indochinite, 2.4 m (7.9 feet) long) containing charge in granite. A typical record is shown in figure 32.16 (Yang et al, 1993).

The wave characteristics at the various locations recorded and the calculated strain and strain rates are shown in table 32.6 (Yang, 1993).

Wave Characteristics at Various Locations and the Calculated Strain and Strain Rates

Table 32.6 – Wave characteristics at various locations and the calculated strain and strain rates.

Vibration Monitoring

Vibration monitoring is another way of measuring wave parameters. Measurement for compliance is dealt with in chapter 26. Clearly a blaster needs to produce a blast design which satisfies the requirements of regulatory bodies and minimizes complaints. However amplitudes measurement, especially measurements close to the blast, can provide information about blast performance. Vibrations close to the blast has relatively high frequencies while far field vibrations are of lower frequency. This means that the recording close to the blast has to be at a sufficiently high frequency. Thus, high frequency geophones have to be employed and the recording has to be at much higher frequencies than far field compliance recordings.

Figure 32.17 shows a typical recording of an underground blast in the form of particle velocity vs. time. One can easily obtain timing information for charges of the blast and compare it against design specifications. It is also possible to associate amplitudes of vibration with sympathetic detonation, use of improper detonators and lack of confinement during the blast that could result in charges flying together will produce high amplitude vibration, while charges firing with reduced burdens, as well as desensitized charges will produce low amplitude vibrations. Vibration amplitude can also be important in assessing damage to the host rock. If it exceeds thresholds related to rock conditions. Typical thresholds are 1,000 millimeters/second inches/second) for hard massive rock and 250 millimeters/second (9.84 inches/second) for rock containing frequent discontinuities.

Air Overpressure Monitoring

Like vibration monitoring, air overpressure monitoring is normally used for compliance reasons. Monitoring can also be used to evaluate blast performance. Excessive air blast may indicate the presence of small burdens, soft seams or inadequate stemming. The overpressure-time histories may also provide information on the reason for high air overpressure measurements. Peaks at regular intervals may indicate stemming ejections while night sharp peaks may indicate blowouts.

High-Speed Video

A production blast lasts only a few seconds. For each detonation there is a shock and stress wave propagation phase, a fragmentation phase and a throw phase. High-speed video information can provide invaluable information, however there are practical limitations on what can be recorded. Frame rate is an important consideration in high-speed imaging. It depends on the duration of the event, the speed of the features that need to be captured and the size of the area of interest. For example if the event occurs in 2 milliseconds and the recording rate is 500 frames per second, not enough information will be recorded. If the recording rate is set higher, then an adequate number of frames may be captured. Many manufacturers of high-speed video equipment allow high filming rates at the expense of resolution. If the aim is to examine full scale blasts and provide general troubleshooting and capture throw, a filming rate of 250 frames/second to 1000 frames/second is adequate. In the case where the shock and stress wave propagation needs to be captured, the filming rate has to be in the order of many thousand frames per second. In such cases one needs to concentrate on a very small area of the blast and may provide an adequate lighting due to the resulting short exposure time. Exposure time has to be adequate to record good quality images but short enough to avoid blurring. The letter depends on the subject's velocity and the amount of lens magnification. Figure 32.18 shows a high-speed image frame where little blurring is evident as the flying fragments.

For blasting diagnostics dedicated high-speed video equipment with related software can be used to provide visual and quantitative information. It is worth noting that some of today's off the shelf video recorders have the capability of recording images at approximate rates of 300 frames/second for adequate time, making them capable of providing blast diagnostics. One limiting factor is available memory. Although continuing development of memory cards has resulted in affordable strategic storage, high speed video recordings are especially demanding and have a duration limited to a few seconds.

To be able to do quantitative analysis, dimensional as well as accurate timing controls should be used. According to Chiappetta (1998) the coordinates of at least five (4) non-collinear points are needed to calculate the calibration constants for 2-D analysis. In the case of 3-D analysis the minimum number of points needed is six. Calibration is typically done by dedicated software, which allows the user to place proper timing controls by triggering the camera by the desired event. Some programs enable data acquisition from sensors (accelerometers, pressure transducers etc.) at the same time as video is being captured.

In blasting, high-speed video can be used for items listed in table 32.7.

Uses Of High-Speed Video

Table 32.7 – Uses of high-speed video.

POST BLAST ASSESSMENT

The term post blast assessment refers to measurements taken after the blast is completed, to evaluate the effect of the blast. Typically these measurements are directly related to the result of the blast. They include fragmentation analysis, damage analysis and muck pile geometry. Measurement of fragmentation or assessment of diggability of the muck pile relate to the size distribution of fragments and the ease by which loading equipment digs into the muck pile. Some of these topics have been dealt with in other chapters of this book and they will not be repeated here.

Fragmentation Analysis

Fragmentation is the gauge of success or failure of a blast. It is usually the main objective of the blast design. Blasting costs, productivity, equipment performance, downstream costs and productivity have been associated with fragmentation. However it is only recently that fragmentation measurements have been available. Fragmentation measurement analysis implies a method, which is quick, representative and inexpensive and repeatable. Furthermore the method has to be flexible to allow measurements at a variety of points of interest. Such points are muckpiles, for the assessment of blast design parameters, trucks, for the assessment of fill factors, crushers, to optimize their performance and conveyor belts to order particle size with continuous effort and costs.

Qualitative Visual Observation

A common but unreliable approach is the method of qualitative visual observation, in which fragmentation is characterized on the basis of experience. Obviously the method is subjective and unreliable, as the human eye cannot estimate the difference in sizes between various similar piles.

Screening

Screening offers an accurate method to estimate fragment size distribution. However it is time consuming and can only tests used as experimental, small blasts. For large scale blasts, sampling a problem, since it is impossible to screen the entire muckpile. It is estimated that 1 to 2 of the muckpile needs to be screened for reliable measurements.

Photographic Techniques

Photographic techniques are the predominant of image analysis, which is widely used today. They ranged from visual comparison of a muckpile with photographs of standard muckpiles prepared in the laboratory (Cunningham, 1987) to manual digitization of fragment from pictures of muckpiles (Ching and Ladberg, 1992, Klishin and Cameron, 1996). It is worth noting that photographic techniques reduce the three-dimensional reality of the fragment and muckpiles into two dimensions; the third dimension is hidden, so photographs present fragments as surfaces, whose areas are representative of their true size. Furthermore, only the surface of the muckpile is visible, meaning that several photographs, taken as the muckpile is excavated are needed to have full picture.

Photogrammetry

Photogrammetric techniques are accurate and provide three-dimensional analysis; however they are time consuming or computer intensive and they are used only for problem feature analysis as well as boulder analysis.

High-Speed Photography

High-speed photography is not used to obtain blast fragmentation distribution. It is however used to shed light in fragmentation related problems, such as to examine the source of large boulders or the source of energy losses in a blast.

Equipment Monitoring

Equipment monitoring offers an indirect method to quantify fragmentation. Intuitively fragmentation affects equipment performance in downstream operations. Shovel and dragline monitoring have been used to evaluate the diggability of muckpiles produced by blasting, while delays at the crusher or boulder count can be related to fragmentation. Hendricks et al. (1990) have monitored performance parameters of a shovel, such as crowd armature current and voltage, hoist armature voltage and current, hoist field current, hoist rope position and crowd arm extension, to understand how these variables vary with fragmentation. They found that there is a link between equipment performance parameters, as well as operator practices and fragmentation and developed techniques to consider the influence of different operator practices. Terrance and Baldwin (1990) have presented a case of dragline monitoring to evaluate the effects of blast design.

Image Analysis

Image analysis as a means of fragmentation analysis has flourished in the last 20 years. A workshop under FRAGBLAST-5 in Montreal (Franklin and Katsabanis, 1996) provided the opportunity for scientists involved in image analysis, users of the systems and engineers in search of solutions to fragmentation measurements, to exchange ideas and discuss progress and limitations. The presentations, outlining techniques and issues, are still of value to the interested reader.

In the years since its commercial introduction, image analysis has enjoyed success in many applications inside and outside the industry. Notably, the aggregate and mineral processing industries use this method for on-line quality control and automated process control.

Image analysis starts from the acquisition of an image, which will be analyzed by a computer program to extract fragment sizes. Consequently, image analysis has to wrestle through the same issues as conventional photography. It needs to identify fragments, it has to measure their size from 2-D images and it has to produce a size distribution from the available measurements. Image analysis is not a new field; it has been used extensively in machine vision, microscopy and medicine. In fact, some of the early image analysis codes were written for use in medicine. However, muckpiles present new challenges for image interpretation and require specific fragmentation software and techniques. These challenges are related to the muckpiles (three dimensional in nature), the fragments (nonuniform texture, unfiltered(facted) nature), image quality (view locations introducing perspective errors, not optimum illumination) and the environment of operation (dust, humidity) (Bolter et al. 1996). The technique involves the following steps: capture of image with a scale, processing and reporting. Since a certain number of images will be analyzed, the application must also consider sampling techniques in order to obtain meaningful results. Images can be acquired by a variety of digital sources.

Image Processing

It is important that the images are clear and undistorted and contain a standard scale. Since images may be taken at oblique angles, it is recommended that a telephoto lens, which compresses the depth of the image, be used. The image should contain relative information. Since the purpose is to analyze fragmentation, a number between 500 and 1,500 of fragments per image is recommended. A smaller number will result in statistical errors while a large number of fragments per image may result in particles lacking the necessary resolution to be recognized by the system.

The image is processed using several operations, performed by a computer program. Typically this includes use of filters to remove speckles and enhance contrast, determination of threshold values to enable edge detection, and use of techniques to delineate blocks, even if they are partially outlined, and disregard spurious edges that are not part of a contour. At this point the user may intervene and improve the fidelity of the fragment net, delineated by the computer. Commercial image analysis packages offer a variety of interactive editing tools. It must however be noticed that manual editing is a time consuming process thus few corrections are realistically possible.

The last part is the reconstruction of the three-dimensional reality from the available surfaces. This is done using statistical techniques and stereology. The interested reader can find information in Maerz (2007) and Kennedy et al. (1997). Finally, fragmentation analysis is provided in terms of standard distribution curves and histograms. Of interest is what is defined as particle size, since this is a different process than sieving. Bedar et al. (1996) provides the following definition for fragment size in their image analysis procedure. "Assuming a fragment contour form a regular, enclosed, closed curve, then the fragment size can be characterized by the length of both its major and minor axes, where the major axis is defined as the longest euclidean distance between two extreme points of the fragment contour, and the minor axis is the sum of the maximum orthogonal distances between points of the contour and the major axis on both its sides." Image analysis techniques vary on their definition of size. They may use the major and minor axes of the 2-d particle, defined as before, or determined after fitting the shape of the image of a fragment into an ellipse (Girdner et al. 1996), or the diameter of an equivalent sphere (Maerz 1996), defined as a sphere with the same projected area as the particle.

Current Commercial Systems' Sources of Error

Several commercial systems analyzing fragmentation on the basis of image analysis are in use today. They provide value to blast analysis; however the user has to be aware of the following sources of error, which include sampling, image quality, delineation of fragments, and fines (Maerz et al., 1996).

Sampling is an issue in studies of particle size of fragmented rock. The muckpile is not uniform. Large fragments may be concentrated at the surface or covered by fines. The surface of the muckpile may not be representative and imaging may have to take place after some part of the muckpile has been excavated. Furthermore a sampling strategy must be implemented to avoid bias. Previous biased images may pay more attention to where they see a problem, or to areas which they consider representative. It is important to eliminate personal bias when collecting images for the analysis of fragmentation. For these reasons automated image acquisition is preferred. Since most material ends up on a truck or on a conveyor it follows that each vehicle load is a discrete package of fragmentation information. Trucks can be photographed in transit or when they dump at a primary crusher (Fig. 32.19) revealing not only the surface layer but underlying material sizes. To characterize the entire blast simply merge all the load data to give a statistically valid and consistent measurement of blast fragmentation obtained in a controlled, non-disruptive setting (Palamsjo, 2008).

Image quality is important in all photographic work. Clear undistorted images with uniform lighting and proper contrast are essential for proper recognition of edges of fragments. Nonuniform lighting, when using a single red light, creates bright and dark areas in different parts of the same image, resulting in difficulty in image analysis. Contrast is what image analysis systems use to discriminating the boundaries of fragments. If the contrast is not sufficient, many fragments will be erroneously recognized as a single block, while, if the contrast is too high, features on each fragment will be supplied and falsely recognized as edges.

Poor delineation of fragments is obviously a significant source of error. It can be caused by poor image quality or it can be the result of the edge detection software due to inadequate or poor calibration (Edan and Franklin, 1996). Two types of edge detection errors are observed: "fusion," when boundaries between fragments are not recognized and many fragments are recognized as a single block, and "disintegration," when surface features of a single rock are recognized as edges and the block is broken into many fragments. Edan and Franklin (1996) have suggested that disintegration increases with the presence of larger fragments and fusion with the presence of small fragments.

Fines may be too small to be delineated, or they may be hidden by larger fragments and may affect the results of image analysis (Katsabanis, 1999). Various techniques have been proposed by developers of image analysis systems to rectify the problem. Advances in imaging technology and the availability of high resolution digital cameras have helped to resolve some of the "Fines" issues but it is recommended that, if the measurement of fines is critical, the success of the proposed techniques be evaluated by appropriate testing.

Image analysis systems have been employed for a number of years in mines and quarries. From their initial versions, as aids to perform fragmentation analysis at single, independent locations, they have evolved into systems to assist in the quality control of the comminution process. It is common to see automated image analysis at a variety of critical points during the comminution process, enabling recording of fragmentation information at the face, on trucks and loaders, before and after crushing. Since it is critical to know the source of the fragments, in order to associate performance of ore handling and comminution systems to blast design parameters, modern image analysis software records vehicle identification and equipment tracking information and integrates with the most popular mine dispatch systems.

Damage Analysis

Blast induced damage must be assessed and related to controllable parameters. The following provides a summary of the main methods used.

Visual Observation

Visual observation using laser profiling can provide assessment of back break. However this provides only surface observation and not information on the extent of damage, which may run deeper. Measurement of overbreak, backbreak, rockmaterial and radial fracturing associated with geological features, may provide some information on the extent of damage; however some measurements may be affected by the diligence of the personnel involved, as minor cracks and preexisting cracks can be hard to detect. In-hole observation with a borehole camera and mapping can provide information on damage, as long as preblast and post blast assessments are conducted. It is important that such surveys are done systematically using a regular pattern of observation holes.

Half-cast holes after a wall control blast are indicative of the quality of the operation and can indicate the extent of blast damage. The half cast factor is the total length of half casts divided by total length of perimeter holes.

Vibration Measurements Combined With Various Damage Criteria

Damage can be observed once the ground vibration exceeds a certain level. Various investigators have associated damage to vibration level close to the blast. Langefors and Kihlstrom (1973) proposed a peak particle velocity (PPV) threshold of 610 millimeters/second (24.0 inches/second) for the formation of new cracks in rock. Similarly Oriard (1982) proposed a threshold of 635 millimeters/second (25 inches/second). Persson et al. (1994) have suggested a PPV threshold related to rock strength according to equation 32.5.

$$PPV_{max} = \frac{\sigma_t}{E} \times V_p$$ <!-- VERIFIED -->

Equation 32.5

Where:

• PPV(max) = peak particle velocity (millimeters/second) (inches/second)
• σt = dynamic tensile strength of the rock (metric units) (U.S. units)
• Vp = p wave velocity in the rock (metric units) (U.S. units)
• E = Young's modulus

For hard rock the peak particle velocity for the onset of cracking is close to 1,000 millimeters/second (39.37 inches/second). In the case of fractured rock masses the maximum PPV is estimated to be a quarter of that is calculated for intact rock (Calder and Larocque, 1973).

EXAMPLE 32.4

Calculate the peak particle velocity assuming a dynamic tensile strength of 20 megapascals for a rock with Young's modulus 80 gigapascals and p wave velocity 5,200 meters/second using equation 32.5.

$$PPV_{max} = \frac{\sigma_t}{E} \times V_p = \frac{20}{80000} \times 5200$$ <!-- VERIFIED -->

$$PPV_{max} = 1.3$$ <!-- VERIFIED -->

The maximum peak particle velocity is 1.3 millimeters/second.

EXAMPLE 32.5

Calculate the peak particle velocity assuming a dynamic tensile strength of 2,900 pounds/inch² for a rock with Young's modulus 11.6 x10⁶ pounds/inch and p wave velocity 17,000 feet/second using equation 32.5.

$$PPV_{max} = \frac{\sigma_t}{E} \times V_p = \frac{2900}{11.6 \times 10^6} \times 17000$$ <!-- VERIFIED -->

$$PPV_{max} = 4.25$$ <!-- VERIFIED -->

The maximum peak particle velocity is 4.25 inches/second.

Due to the high amplitude of vibrations, accelerometers are typically used for close to blast monitoring while geophones are preferred in the medium to far field ranges. As discussed previously, the frequency of the expected wave is important in selecting monitoring equipment. Data acquisition should take place at a minimum frequency twice the expected one to avoid aliasing.

Monitoring vibrations close to the blast provides an indirect method of assessing damage. Strictly speaking, the above formula, which was developed for plane conditions, does not apply in the common, nonplanar waves, radiating from cylindrical boreholes. However it provides a rather simple method to evaluate the damage potential of different blasts and has found some industry. The technique is widespread and many examples of its use can be found in the literature. The reader may wish to review relevant papers by Ouchterlony et al. (1996), Liu et al. (1998), McKenzie and Holley (2004), and Villanueva et al. (2008).

Monitoring of Pressure-Time History in Monitoring Boreholes

This has been used by Ouchterlony et al. (1996) and Brent and Smith (1999). According to the technique, pressure transducers are placed in monitoring boreholes behind the blast and each hole is plugged at the collar. Pressure transducer reads a few atmospheres with durations up to a few seconds. Figure 32.20 provides a schematic of the measurement.

During the blast the pressure pulses are recorded and interpreted. Positive pulses (overpressure) suggest gas penetration in the borehole. Negative pulses (decreasing pressure) indicate the cracks are opened by the stress wave so that air is sucked into the fractures. Thus, the reason for damage (stress wave, gas) may be identified together with the limits of the damage zone. In some cases, these techniques have been combined with borehole camera assessment to visualize new cracks (Brent and Smith, 1999).

Time Domain Reflectometry

Coaxial cables, grouted in boreholes can be monitored using time domain reflectometry (CDR) cable testers, providing locations where the cable is damaged. For the technique to work, a large enough displacement of the rock is required. The technique has been used by Leblane, Ryan and Heilig (1995) to monitor damage around a confined charge.

Cross-Borehole Seismic Monitoring

This relates changes in the characteristics of the transmitted pulses before and after a blast to changes in the geotechnical properties of the rock. Often the P-wave velocity is used (Recoque, 1992); however the P-wave velocity is not a sensitive indicator of damage. Closure of joints and changes in the water table affect the recorded P-wave velocity making interpretation difficult or impossible. For this reason the rock quality factor Q, which is related to the loss of energy during a stress-strain cycle in the rock, may be used (Kennard et al., 1996).

The technique, in the form of P-wave velocity measurement has been used by Rocque (1992) and Leblane, Heilig and Ryan (1995). The latter intensive nature of the technique makes it essential for monitoring mine development.

Geotomography

Geotomography is a technique in which seismic waves are used to penetrate the rock mass and their travel velocities are analyzed to reveal areas of different velocities. The technique is similar to three different planes (Rocque, 1992). Rays connecting the transmitter and receivers are computed as well as the velocities. During the analysis velocities in each cell form a set of recovered velocities. These can be accelerometers, geophones or hydrophones, which must have the necessary frequency range to capture the wave at the locations of interest. During the recording step the arrival times of P-waves are determined and various techniques are used to invert and interpret the data.

All phases require substantial effort and cost; thus geotomography should not be considered a routine exercise. The concept is shown in figure 32.21 for the case of an asymmetric experiment in granite. Several holes were drilled in a square configuration around the charge and served for the placement of the seismic receivers (detonators) and the receivers (hydrophones). Seismic surveys were conducted in three different planes perpendicular to the axis of the charge before and after the blast and tomography interpreted the difference as wave velocities. Another study of the damage in the case of underground pillars has been presented by Trentanek, Kepniak and Thengold (1995). Seismic sources were placed by hammer impacts on one side of the pillar, while twelve geophones on the other side were used as receivers. Velocities below an assumed level were associated with blast damage.

Extensometers

As their name indicates, they are used to measure the extension of their longitudinal axis. The value of such measurements is in the extent of preconditioning of the next bench as well as the influence of blasting on permanent rock structures. According to the measurement medium, extensometers can be divided into wire, rod and probe types. They can also be divided into single point or multipoint extensometers, depending on whether one measuring section or several measuring sections are considered. Extensometers use is not as common in blasting as in civil engineering applications. The problem is that they do not provide dynamic data and anchoring may be problematic close to the blast. However, regular measurement before and after a blast can provide a good record of relative displacements. A study of blast-induced damages using extensometers, among other instrumentation, is given by Villanueva et al. (2004). Multi-anchor extensometers were used to assess rock mass deterioration due to blasting. Effects of the extraction of neighboring stopes could also be observed. Figure 32.21 provides an example of an extensometer record from this study.

Muckpile Profile

Muck piles are not generally accessible after a blast. However, reflection laser profiling equipment can be used to obtain geometrical information, the same way it is used for face profiles. Otherwise profiles can be generated in a less rigorous manner from the muck surface. Areas are visible will have to be scanned from a different direction and the profiles will have to be merged by software. Figure 32.22 shows cross sections of muck piles using laser technology.

Generation of a geometrical model of a muck pile allows the calculations of displacement, volume, swell and cast efficiency in cast operations.

CONCLUSION

In order for blast design to move away from intuition and empiricism, measurements of critical parameters are necessary. A variety of measurements aimed at knowing the properties of the rock, the geometry of the blast the properties of the consumables used and the outcomes of the blast can assist the blaster in evaluating a blast design and attempt improvements. Optimization of blasting is only possible through accurate measurements and their interpretation. According to the material of this Handbook it is clear that accurate prediction of the blast result from fundamental principles is still an elusive target; thus means to lack of adequate and high quality data to clearly identify the role of various parameters as well as the complexity of the processes involved.

Today, available instrumentation and technology improvements allow the blaster-in-charge to collect the necessary information, which can be used in calibrating engineering models or formulas or, alternatively, in a step approach in which the blast design is modified on the basis of the blast outputs until a practical optimum is achieved. Not all of the previous methods will be used in every blast. However the blaster can select methods that provide the necessary information for the problem at hand. Instrumentation is costly and monitoring requires time and effort; however the benefit of accurate measurements should outweigh the expense. In all cases, accurate documentation of the conditions of the blast, as well as observation of experienced personnel are critical. They increase the possibility of success, augmenting the information recorded by instrumentation and providing articulable detail.

REFERENCES

Barnaba, F., L.D. Dunckerhorf, and C.F.B. Hendricks. 1996. Comparative performance of a novel seismograms for identification of stuck pile fragment boundaries. Rock Fragmentation by Blasting (FRAGBLAST) Proceedings of the 5th International Symposium on Rock Fragmentation by Blasting. Montreal, Quebec, Canada, Aug. 23 – 24, pp. 127-132. A.A. Balkema, Rotterdam, Netherlands.

Bedar et al. 1996

Brent, G.F. and G.E Smith. 1999. The detection of blast damage by borehole pressure measurement. Rock Fragmentation by Blasting (FRAGBLAST) Proceedings of the 6th Conference FRAGBLAST, pp. 39 – 52. Johannesburg, South Africa.

Calder, P. and G. Larocque. 1977. Perimeter blasting. pit slope manual, chapter 7. Centre for mineral and energy technology, Ottawa, Canada.

Chiappetta, R.F. 1998. Blast monitoring instrumentation and analysis techniques, with an emphasis on field applications. FRAGBLAST – International Journal of Blasting and Fragmentation, Vol. 2, No. 1, pp. 79-122. BALKEMA, A.A. Rotterdam, Netherlands.

Ching, H. B. and G. Ludwig. 1992. Semi-automated fragmentation assessment. International Society of Explosives Engineers (ISEE) Research Proceedings of 8th 8th Annual Conference on Explosives and Blasting Technique, pp. 135-139, Orlando, Fl. ISEE, Cleveland, OH.

Cunningham, C.V.B. 1987. Fragmentation estimations and the Kuz-Ram model - four years on. 2nd International Symposium on Rock Fragmentation by Blasting, August SEM, pp. 475 – 487, Keystone, CO.

Dautrith Process Instrumentation. 2009. Volos 1000 A Velocity of Detonation Measuring Instrument. http://dautrith.com.

Davies, E. W., L.A. Smith, and C. De La Cruz. 1997. The measurement of Detonation Waves as a Comparative Explosive. International Society of Explosives Engineers (ISEE) Research Proceedings of the 23rd Annual Conference on Explosives and Blasting Technique, pp. 145-156. ISEE, Cleveland, OH.

Edan, D.D. and J.A. Franklin. 1996. Discontinuity, fines and edge set fidelity. Measurement of Blast Fragmentation. Rock Fragmentation by Blasting (FRAGBLAST) Proceedings FRAGBLAST, Workshop on measurement of blast fragmentation, August 23 – 24, pp. 127-132. Montreal, Quebec, Canada, A.A. Balkema, Rotterdam, Netherlands.

Franklin, J. and T. Katsabanis. 1996. Measurement of Blast Fragmentation. Rock Fragmentation by Blasting (FRAGBLAST) Proceedings of the 5th Conference FRAGBLAST, Workshop on Measurement of Blast Fragmentation, Workshop on Measurement of Blast Fragmentation, pp. 157 – 166, August. Montreal, Quebec, Canada, Aug. 23-24, pp. 31, A.A.A. Balkema, Rotterdam, Netherlands.

Franklin, J.A., N. H. Maerz, and C.P. Bennett. 1988. Rock mass characterization using photoanalysis. International Journal of Mining and Geological Engineering. Vol. 6, pp. 97–112.

Ginsberg, M.J. and B.W. Asay. 1991. Commercial carbon composition resistors as dynamic stress gauges in difficult environment. Review of Scientific Instruments 62(9), Sep. , pp. 2218-2227. American Institute of Physics. Melville, NY.

Girdner, K.K., J. M. Kemeny, A. Srikant, and R. McGill. 1996. The split system for analyzing the size distribution of fragmented rock Rock Fragmentation by Blasting (FRAGBLAST) Proceedings of the 5th FRAGBLAST Conference, Measurement of Blast Fragmentation, Montreal, Quebec, Canada, Aug. 23-24, pp. 101-108. A.A. Balkema, Rotterdam, Netherlands.

Hendricks, C., J. Peck, and M. Scoble, M. 1990. Integrated drill and shovel performance monitoring towards blast optimization. 3rd International Symposium on Rock Fragmentation By Blasting, Aug. 26-31, pp. 9-30. Brisbane, Australia.

Hendricks, C., M. Scoble, J. Peck, and T. Szynalowski. 1991. Blasthole deviation: Measurement, mechanisms and impact on dilation. International Society of Explosives Engineers (ISEE) Proceedings of the 17th Annual Conference on Explosives and Blasting Technique. Vol. 1, pp. 11-21, Las Vegas, NV. ISEE, Cleveland, OH.

Katsabanis, P.D. 1999. Comparison Between Images Analysis and Serving Using Laboratory Scale Model Muckpiles. 25th Annual Conference on Explosives and Blasting Technique. Nashville, TS. Feb. 7-10, Vol. 2, pp. 325-332.

Kemeny, A., G.P.J. Chalivour, C.K. McKenzie, B.J. and Yang. R.T. 1993. A model of seismic pulse propagation and its application to Determine Q for a rock mass. International Journal of Rock Mechanics and Mining and Geomechanics Vol. 27, No. 1, pp. 33 – 41. 1990. Elsevier. International Journal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts.

Kemeny J. and B.Port R. 2003. Estimating three-dimensional rock discontinuity orientation from digital images of fracture traces. Computers and Geosciences, 29(6): 77. Journal of the International Association for Mathematical Geology

Kemeny, J., Harwood, J., & Turner, K. (2006). The Use of Ground based LiDAR for Geotechnical Aspects of Highway Projects. 57th Annual Highway Geology Symposium. 1p. 1831

Kemeny, J., A. Devgan, R., Hagaman, R., and X. Wu. 1993. Analysis of rock fragmentation ming digital image processing. Journal of Geotechnical Engineering., vol. 119, no. 7, pp. 1144-1160.

Kemeny, J., J. Donovan, J. and C.R. Silva. 2006. Application of ground – based lidar for pre-blast rock mass characterization. Rock Fragmentation by Blasting (FRAGBLAST) Proceedings of the 8Th Conference FRAGBLAST, pp. 50 – 54, May 7 – 11, Santiago, Chile.

Klishin T.M. and A.K. Cameron. 1996. Blast fragmentation assessment using GoldSize. Measurement of Blast Fragmentation, Rock Fragmentation by Blasting (FRAGBLAST) Proceedings of the 5th Conference FRAGBLAST, Workshop on Measurement of Blast Fragmentation, August 23 – 24, pp. 83 – 89 Montreal, Quebec, Canada, A.A. Balkema, Rotterdam, Netherlands.

Langefors, U. and B. Kihlstrom. 1973. The Modern Technique of Rock Blasting. Wiley and Sons, New York, NY.

Lato, M., M.S. Diederichs, J. Hutchinson, and R. Harrap. 2009. Optimization of laser scanning and processing for automated structural evaluation of discontinuities in rock masse. Canadian Geotechnical Journal, Vol. 41, pp. 1011-1025. NRC Research Press.

Leblane, M. J. M. Ryan, and J. H. Heilig. 1995. Predicting the envelope of damage from the detonation of a confined charge. International Society of Explosives Engineers (ISEE) Proceedings of the 21st Annual Conference on Explosives and Blasting Technique, pp. 11-35. Nashville, TN. ISEE, Cleveland, OH.

Lemy, F. and J. Hadjigeorgiou. 2004. A digital face mapping study in an underground hard rock mine. Canadian Geotechnical Journal, Vol. 41. pp. 1011-1025. NRC Research Press.

Lin, Q. 2003. Blast vibration and damage evaluation, optimization of blasting operations. Queen's University continuing education course. June.

Lin, Q., P. Tebono, D. Tunaley, and T. Francis, T. 1995. Observations of the shock resistance of emulsion explosives in rock blasting. International Society of Explosives Engineers (ISEE) Research Proceedings of the 11th Annual Conference on Explosives and Blasting Technique, pp. 26-27, February 5 – 9, Nashville, TN. ISEE, Cleveland, OH.

Liu, Q., H. Tran, D. Cosman, and P. Andrieux, P. 1998. A case study of blast damage evaluation in open stooping at Kidd Creek Mine." International Society of Explosives Engineers (ISEE) Proceedings of the 24th Annual Conference on Explosives and Blasting Technique, pp. 731 – 734, February 2 – 5, pp 213-314 New Orleans, LA. ISEE, Cleveland, OH.

Maerz, N.H. 1996. Reconstructing 3-D block size distribution from 2-D measurements on sections. Measurement of blast fragmentation. Rock Fragmentation by Blasting (FRAGBLAST) Proceedings of the 5th FRAGBLAST Conference, Workshop on Measurement of Blast Fragmentation, pp. 39-64, August 23 – 24, Montreal, Quebec, Canada, A.A. Balkema, Rotterdam, Netherlands.

Maerz, N.H., and T.C. Palangio. 2007. Optical sizing analysis of blasted rock: Lessons learned, Proceedings of the European Federation of Explosives Engineers. Vieanna Conference, pp. 73 – 82.

Maxwell, S.C. and R. P. Young. 1993. Seismic imaging of blast damage. International Journal of Rock Mechanics and Mining Science 30, pp. 1435-1440.

McKenzie, C. & Holley, K. (2004). A Study of Damage Profiles Behind Blasts. Proceedings of the 30th Annual Conference on Explosives and Blasting Technique (pp. 203-214), Cleveland: ISEE

MDL. 2009. MDL LaserAce® Pocket Series Manual.

Moser, P., Gauster, M., & Gaich, A. (2005). Experience with and Benefits from the use of 3D Stereophotogrammetry for Blast Design and Control. Proceedings of the 33rd Annual Conference on Explosives and Blasting Technique. Cleveland: ISEE

Mozaffari, S. 2007. Measurement while drilling systems in Aitik Mine M.Sc. thesis, School of applied geosciences and mining. Lulea University of Technology, Lulea, Sweden.

Palangio, T.C. 2008. Digital image analysis – photography in blasting systems, International Society of Explosives Engineers (ISEE) Proceedings of the 34th Annual Conference on Explosives and Blasting Technique, February 27 – 30, New Orleans, LA. ISEE, Cleveland, OH.

Persson et al 1994

Rosado, N. A., K. Chalaturoik, and T.G. Schunnesson. 2009. Damage assessment of basaltic rock mass due to repeated blasting in a railway tunneling project – A case study. Tunneling and Underground Space Technology, Vol.24, Issue 2 pp. 208 – 221. Elsevier Ltd.

Reid, T.R. and J. P. Harrison. 2000. A semi-automated methodology for discontinuity trace detection in digital images of rock mass exposures. International Journal of Rock Mechanics and Mining Sciences 37 pp. 1073-1089.

Rocque, P. 1992. Techniques Utilized for the Detection and Measurement of Blast Induced Damage. M.Sc.(Eng.) Basix, Dept. of Mining Engineering, Queen's University at Kingston, ON, Canada.

Rodgers, J.A., R. Lee, and K. C. Whitaker, K.C. 2008. The origins and effects of inter-deck pressure at decked blasts." Blasting Analytics International, Inc. Proceedings of the 9th High-Tech Seminar, July 24 – 27, Orlando, FL.

Oriard, L.L. 2002. Explosives Engineering, Construction Vibrations and Geotechnology, pp. 17-26. International Society of Explosive Engineers, Cleveland, OH.

Ouchterlony, F., G. L. Nielson, I. A. Daug, U. 1996. Monitoring of large open cut Rounds by VOD, PPV and gas pressure measurements. Rock Fragmentation by Blasting (FRAGBLAST) Proceedings of the 5th Conference FRAGBLAST, Montreal, Quebec, Canada. A.A. Balkema, Rotterdam, Netherlands.

Persson, A. S., H. Holmberg, and J. Lee. 1984. Rock Blasting and Explosives Engineering. CRC Press, Boca.

Segui, J.B. and M. Higgins. 2002. Blast design using measurement while drilling parameters. FRAGBLAST – International Journal of Blasting and Fragmentation, Vol. 6, pp. 287-299, A.A. Balkema, Rotterdam, Netherlands.

Split Engineering. 2009. Split-FX Software. Automated rock mass fracture characterization. http://www.spliteng.com/tutorials/SplitFX Brochure 2009.pdf.

Thomas, C. 2003. Improved blasting control and quantity calculations using reflectorless laser systems. International Society of Explosives Engineers (ISEE) Proceedings of the 29th Annual Conference on Explosives and Blasting (FRAGBLAST), Cleveland, OH.

Trentacoste, A.C. and G. Baldwin. 1990. Blast performance measurement using a dragline monitor. Rock Fragmentation by Blasting (FRAGBLAST) Proceedings of the 3rd Conference on FRAGBLAST, Aug. 26-31, pp. 219-234, Brisbane, Australia.

Villanueva, C., Eguiza, A., and R., Thompiquel, C (1995). Determination of Blast Induced Damaged Zones in Pillars by Seismic Imaging. Explo 95 Conference (pp. 387-303). AusIMM

Villanueva, E., I. 2ubarista, and W. Scott. 2004. Blast induced damage and dynamic behaviour of hangingwalls in bench stoping. Rock Fragmentation by Blasting (FRAGBLAST) Proceedings of the 8TH Conference FRAGBLAST, pp. 220 – 230, May 7 – 11, Santiago, Chile.

Walrych-Copeland E. and P. Lilly. 2002. A review of general considerations for examining rock mass blastability and fragmentation. International Annual Journal for Blasting and Fragmentation, vol. 6. no. 2, pp. 157-168.

Wieland, M. 1987. Instrumentation delay-blast confirmation in underground coal. International Society of Explosives Engineers (ISEE) Research Proceedings of the 9th Annual Conference on Explosives and Blasting Technique, pp. 139 – 250, January 31 – February 4, San Diego, CA. ISEE, Cleveland, OH.

Yang, R.L., P. Rocque, P. Katsabanis, and W.F. Bawden. 1993 Measurement and analysis of near-field blast vibrations and damage, Geotechnical and geological Engineering. 12, 169-182.

Zyural, J, F. Vest, K. King, and I. Shroble, J. 2005. Integrated Inertial GPS positioning system for open pit mine applications. Prepint 05-062, February 28 – March 2, SME Annual Meeting, Salt Lake City, UT.