Intuition, rather than hard facts, more often guides mine operators in predicting the adverse effects blasts will have on mine excavations. But predicting blast-induced damage is slowly coming into its own as a reliable science. In much the same way that rock mechanics engineers can design stable openings using known variables, such as rock quality and strength indexes, engineers are beginning to rely on blast vibration monitoring to develop meaningful data to minimize damage. Furthermore, blasting problems, which often lead to rock damage, can be investigated with suitable instrumentation. With appropriate quantitative assessment tools, blast patterns, charge concentrations and delay sequence designs can be modified and reassessed for improvement. Sophisticated data acquisition and analysis allow blast designers to obtain positive graphical feedback on how their blast design functions in practice and how it affects the surroundings.
Blast monitoring serves several purposes. In open pit mines it can be used to determine the magnitude and effect of ground vibrations and air over-pressures on buildings and their occupants. This type of monitoring is called “far-field” monitoring; it is also done for environmental reasons. The term “far-field” applies generally to blasting operations situated between 50 and 100 metres or more from such structures. Instruments for monitoring blast compliance can protect the public from blasting effects and, not incidentally, mining companies from litigation.
“Near-field” blast monitoring is generally part of a program to optimize underground blasting. Monitoring is usually within 50 meters of the blast. An analytical tool (but not a compliance tool), it enables blast designers to see in quantitative terms how their blasts perform. It is analogous to high-speed photography of open cut blasts. Actual charge delay times can be assessed and the occurrence of possible blast malfunctions or misfires can be identified. Unless monitoring is carried out, there is usually very little evidence left to indicate the source of the problem. “Near-field” monitoring often reveals problems that would otherwise go unnoticed. For ground control engineers, monitoring vibration levels reveals the effect blasting has on instability and sets limits for blast design variables. Blast monitoring costs are justified by improved blast economies and reduced blast-induced damage that can lead to costly ore dilution and increased support costs.
Monitoring blasts is an excellent educational tool for less experienced blast designers. It illustrates graphically the theory of delayed blasting and how blasting affects surrounding rock structures such as pillars, highwalls or tunnels.
A blast produces a large amount of energy in the form of ground vibration over a broad range of frequencies. However, high frequencies attenuate quickly over distance from the blast. For far-field monitoring the concern is for structural damage and human response sensitivities, both of which respond to frequencies of fewer than 100 hertz. These lower frequencies propagate well over large distances. For blast analysis, high frequencies contain important information on what actually happens during the blast. So monitoring must be done at close range. At these distances, total energy is much greater, of course, and requires sensors capable of handling these levels. For the most part these requirements eliminate standard blast monitoring seismographs, in which neither the sensors nor the recording equipment has the frequency response to capture this information.
Researchers at Noranda Inc. have found useful blast information in frequencies up to about 3,000 hertz. Obviously, sensors that respond to at least this range are needed. Modern monitoring equipment, recording digitally, will require sampling rates high enough to capture this data accurately (in this case better than 8,000 samples per second). The higher the sampling rate, the better the fidelity of the recording — not much different from the “oversampling” done with today’s compact disc players.*
Noranda’s accelerometer and the OmniProbe 1200 data acquisition and analysis unit, manufactured by Ottawa-based Instantel Inc., were designed to meet these requirements (Figure 1). (Blastronics Pty. Ltd. of Australia and Slope Indicator Co. of Seattle, Wash., also manufacture data acquisition units capable of “near-field” blast monitoring. The Blastronics units are marketed and serviced in Canada by blm Mincon of Sudbury, Ont.) The sensors, in either uniaxial or triaxial configurations, were developed specifically for near-field monitoring. They can be placed inside boreholes or mounted on the surface. The determining factor here is how competent the rock structure is near the surface to transmit adequately the vibration energy, especially the high frequencies. Highly fractured surfaces may not constitute an efficient medium. Uniaxial sensors will suffice, if placed appropriately. In near-field monitoring, the energy generated can be highly directional, necessitating some care when orienting the sensors. Triaxials gather information from all three axis, which is important for assessing blasting effects on rock stability. For blast analysis, the timing relationships in the waveform provide the most information. The peak amplitudes from each charge help determine whether each charge has released the same vibration levels in proportion to the others. For rock mechanics work, accurate amplitudes are mandatory because they provide a measure of the stress to which the rock mass was subjected.
The OmniProbe 1200 field collector and analysis software provide the needed recording instrument and blast analysis capabilities. Up to 12 channels of data can be collected, with record times of up to 10 seconds and sample rates of 16,000 samples per second per channel. Virtually any sensor can be easily programmed — accelerometers, geophones, etc. Other channels could be connected to high pressure air sensors, for example, if air overpressures are causing damage to underground ventilation systems or equipment.
Usually, a wire loop placed in the first hole triggers the recording instrument. This provides important “time zero” information for calculating propagation velocities or synchronizing cap delay times on the waveform traces. Alternatively, ground vibration levels at the sensor can set off the monitoring. Because recording monitors are usually set up several hours before the shot, extraneous vibrations from machinery must not prematurely trigger the monitor. If this is possible, wire triggering is the preferred method. The continuous mode selection of the data collector also allows for the capture of more than one event. Depending on the duration and the distance from the shot, record times should be set to ensure the entire event is captured. It is always best to err on the side of longer record times, especially if air overpressures are being monitored.
A standard compatible 286 DOS-based computer can transfer data to diskettes. This medium can display, analyze, and print or plot data. The disks are good for long-term future reference.
Two examples of blast recordings are shown in Figures 2 and 3. In Figure 2, the recorded traces from a multiple hole blast contain separate vibration packets corresponding to the delayed explosive charges in the blast sequence. Analyzing this data begins with determining which packet represents each nominal delay so that actual and theoretical delay times are compared. Such a comparison often reveals considerable discrepancy in the actual intervals and may indicate possible blast malfunctions. The longhole production blast shown in Figure 2 indicates good charge separations and sequence design. Note, however, that two holes delayed at 410 milliseconds are not visible on the trace, suggesting a cutoff or a misfire occurred.
`Sympathetic’ Detonations
Near-field blast monitoring can reveal undetonated charges in rock when they do not appear on the vibration trace. It can also highlight explosive charge disturbances caused by
“sympathetic” detonations (detonations caused by vibration induced by resonance), explosive or detonator desensitization, and overbreak. Once these potential problems are identified, fundamental blast engineering can solve the problem. Follow-up monitoring, and the use of other instrumentation (such as vod probes), can be used to check solutions. Naturally, regular monitoring provides early warning of potential problems.
Figure 3 illustrates a vibration trace from an underground drop raise blast. This record indicates that of the 12 delayed charges in the blast, only three fired as expected (the larger traces), two fired simultaneously (at 0.8 seconds), five malfunctioned in some way (the very small packets), and two were misfires or cutoffs (not seen at all). The first three packets in the trace were produced by equal quantities of explosive placed in three separate “cut” holes. Evidently, two of the charges were affected in some way as indicated by their very small amplitudes in proportion to the first packet. Further testing proved that indeed only one of the cut holes needed to be loaded to provide the equivalent blast results. Reductions in the quantity of explosive and the number of blast holes were subsequently made with confidence.
Further detailed analysis of the vibration waveforms obtains the frequency content of the events and the peak ground particle motions. Analysis of the frequency attenuation over time in a particular rock domain reveals significant changes. Figure 4 shows a typical fast Fourier transform (fft) for a selected packet from Figure 2. This information can show increased fracture frequency in the rock mass and, thus, how the blast has “damaged” or altered the surrounding rock. Usually this data must be correlated with other rock mechanics instrumentation, such as extensometers or borehole cameras. In a similar fashion, determining particle accelerations, velocities, or displacements correlates quantitative values with visual changes in the rock mass. These values are easily determined with modern analysis software such as the OmniProbe 1200, which includes built-in routines for integrating or differentiating waveforms, performing peak particle summations, and running ffts on selected portions of the data.
Although there is much to learn regarding near-field blast damage to rock structures, only by using suitable instrumentation can we gather the necessary information to begin. Eventually damage criteria may be established for rock structures as it is for civil structures.
Be the first to comment on "CURBING BLAST DAMAGE"