Measurement and Modeling of Blast Movement for Open Pit Mines

Contact: Ian R Firth. +1 775 784 4185

ianfirth@powernet.net

One of the most frequently encountered requirements of mining operations in which rock is broken using bench blasting techniques, is the need to produce a well fragmented muckpile which will "dig" easily. "Diggability" is a function of various aspects of the blast design, notably timing and powder factor. A "hard digging" muckpile will cause poor productivity in the loading fleet, excessive machine wear and will adversely effect areas as diverse as mining plan conformance and mill through-put. The most common approach to the provision of a muckpile which is "easy" to dig is the combination of long inter row delay and relatively high powder factor, promoting extensive lateral movement and vertical heave within the rock mass.

In direct contradiction of this approach is the desire to minimize dilution of ore zones and reduce mixing of material types within the muckpile. In practice this is achieved by the use of "choke" blasting techniques. While choke blasting is accepted as being successful in this regard, there currently exists no method of quantifying the degree of movement and mixing experienced within a given muckpile, and relating this mixing to other aspects of the blast such as degree of fragmentation and grade recovery.

The Mackay School of Mines, Department of Mining Engineering, has now completed four investigations regarding the categorization and measurement of blast movement at precious metal mines in Nevada. The focus of the work has been the development of a methodology which will allow the operator to characterize blast movement for a given blast design and in a given geologic setting, prior to the beginning of the loading cycle. This would allow more accurate placement of relevant material boundaries, with the concomitant result of improved value recovery and reduced unit cost. Techniques trialed have included marker bags (Zhang 1994) and magnetometry (Harris 1997). While these studies have demonstrated considerable success in terms of their respective research objectives, neither system has been developed to a level at which it will provide the necessary information in a manner suitable for use on a regular basis within the operational constraints of a mine site.

Numerical modeling of bench blasting may be considered the key to a fundamental understanding of the blast movement and fragmentation process. To date no code is commercially available which is capable of realistically modeling the complex rock mass and gas flow dynamics required to accurately predict blast movement and fragmentation. The Orica Explosives group have developed the Discrete Motion Code, (DMC), which is capable of the full sequence of blast movement modeling, but this code is proprietary and its use is limited to Orica personnel. Other significant models include Sabrex (Jorgensen and Chung 1987), also a proprietary product written by Orica explosives and 3x3 Pro, a commercial model written by the JKMRC. These models tend to allow only very limited understanding of the mechanics of the blast and are by no means accurate. The combination of physical measurement and numerical modeling when integrated with a mine planning software suite, would enable accurate digging limits to be set in the field as a routine aspect of the daily mining plan.

The primary objective of the project is to develop a method for the measurement of blast movement in open pit mines. The information needs to be provided to the mine planning engineer and ore control geologist within a suitable time frame after the blast, such that this information can be used in determining the correct post blast location of ore blocks in a given muckpile.

In addition to the development of a practicable method of blast movement measurement, the project aims to integrate this with the ability to model open pit bench blast and allow the results to be displayed using standard mine planning software.

The first investigation into the possibility of measuring blast movement in open pit mines used marker bags. Sections of wet hole liner bag approximately 1.0 m in length were filled with drill cuttings and marked according to the hole number and depth from collar. The bags were placed in extra, un-charged holes drilled in between the normal blast pattern.

After the blast, during the loading cycle, the position of the marker bags was surveyed as they became visible in the digging face. The result is a number of discrete vectors describing the movement of the bag. The method is simple and relatively inexpensive, however, bag recovery is very low and the post blast survey is time consuming.

The next development of the method was to use a magnetic target and a magnetometer to determine its position in 3D post blast. The magnetic targets are approximately 0.5m long and are constructed of a steel core with permanent magnets attached to it at either end. Design considerations include the diameter of the blast hole in which the target will be placed and sufficient magnetic "strength" to allow easy detection to depths of 10m.

The targets are again placed in un-charged holes. The location of the target is calculated from the collar location and a measured depth from collar, assuming a vertical blast hole. After the blast, a magnetic survey is conducted over the muckpile surface. The resulting data is processed using various geophysical data processing algorithms, to produce the new position of the target in 3D.

The method has the significant advantage over marker bags that it requires only approximately 1 hour of survey time post blast. The new target locations can be determined before the muckpile is loaded.

Several workers have developed blast movement models which to some degree allow the user to predict the heave and lateral movement expected for a given blast design in a given rock mass. Perhaps the most advanced in terms of practical application is the Distinct Motion Code, developed by Dale Preece at Sandia National Laboratories in conjunction with Orica Explosives. This code is proprietary and not widely available.

In general, the objective of a bench blast model is to allow the user to design a blast and make some quantitative judgement of the results, as compared with other designs, without the need for field trials. The model must be capable of allowing variation of both rock mass and blast parameters.

Gilbride 1995 attempted to develop a 2D blast model using the Universal Distinct Element Code by Itasca. While a model was developed that allowed real rock mass and blast parameters to be used, the model suffers from very long run time and some problems are encountered in the physical interaction of rock blocks. Even with the significant improvement in processor speed available in todays desk top computers, the model is too slow to be of practical use.

Since it is given that a functional model exists (DMC), the project is now considering the use of output data from numerical models as input data to mine planning software. The general approach is to use the block centroid locations generated by the numerical model for pre and post blast sates, and assign grade and tonnage information to them using a mine planning package such as Surpac. This allows the predicted movement of the muckpile to be used in the calculation of post blast material limits.

Gilbride LJ. 1995. Blast Induced Rock Movement Modeling for Bench Blasting in Nevada Open Pit Mines. M.S. Thesis, University of Nevada Reno Dept Mining Engineering.

Harris GW. 1997. Measurement of Blast Induced Rock Movement in Surface Mines Using Magnetic Geophysics. M.S. Thesis, University of Nevada Reno Dept Mining Engineering.

Zhang S. 1994. Rock Movement due to Blasting and its Impact on Ore Grade Control in Nevada Open Pit Gold Mines. M.S. Thesis, University of Nevada Reno Dept Mining Engineering.