Effects of faults and stress on open stope design

Abstract

The thesis is focused on evaluating the effects of faults on the stability of open stopes. Faults are often the source for open stope overbreak and dilution. How faults affect stope stability is still not well understood. This thesis identifies and evaluates the fault characteristics that adversely affect stope stability. The modified stability graph method (Mathews et al. 1980, Potvin 1988) is an empirical design tool that is popular for open stope design. However, this design method does not include a factor to account for the presence of faults or soft zones on excavation stability. Substantial effort was devoted in this thesis to improve the applicability of the stability graph method.

The main two goals of the thesis are to: - Determine how faults affect the stability of stopes and to provide a method for estimating overbreak due to faults, and - Develop a fault factor for incorporation into the stability graph to improve the method.

The secondary objective are to: - Verify the applicability of the stability graph method in a geological environment outside Canada, with poorer rockmass qualities - Present statistical tools for optimally defining the boundaries between stability zones in the stability graph, and determine the reliability of a stability graph, and - Provide alternative likelihood-based stability graphs, and stability graphs based on a modified definition of span.

Characteristics of faults that adversely affect the stability of stopes are given. These characteristics are considered in numerical models to examine their influence on the effect of faults on stope stability. A procedure for estimating overbreak in open stopes due to faults, using numerical models, is presented and applied to case histories from Kidd Mine, Ontario, Canada.

To overcome the weakness in the stability graph method resulting from faults, a methodology for determining a fault factor for incorporation into an empirical design procedure for open stopes, using the stability graph method is developed. The method is applied to case histories from Kidd Mine in Canada. For large faults and shear zones intersecting stope surfaces, the rockmass quality of the soft zone is important for the stability of a stope. A procedure for accounting for soft zones in the stability number N' is given. The approach is applied to case histories from Ashanti Goldfields Mine in Ghana, Africa. The Kidd database contains 112 case histories, and the Ashanti database has 130 case histories.

Ashanti Goldfields is in a different geological settin from which the stability graph was developed, and is characterized by much weaker rocks. The stability graph method is used at Ashanti Goldfields Mine to verify its applicability.

Statistical tools are presented based on the likelihood ratio statistic to optimally define the boundaries between stability zones in the stability graph, and to determine the reliability of a stability graph. Alternative likelihood-based, stability graphs and a stability raph based on a modified definition of span are introduced. Procedures are given for determination of the possible errors in design when the stability graph is used, and methods are presented for defining and optimizing the boundaries between zones in the stability graph.

A stability graph method based on the use of a modified span rather than hydraulic radius is presented as an alternative tool to the conventional stability graph.

The main conclusions of the thesis are as follows: - For stopes with nearby faults, the faults affect the stability of the stopes by their tendency to increase the zone of low stress or relaxation near the stopes. - For a stope with a nearby fault, the angle between the fault and the stope surface and the relative position of the fault are the most important factors controlling the severity of fault-related overbreak. - The effects of faults on open stope stability are influenced by stope geometry, distance of fault from stope surfaces and in situ stress. - Numerical models can be used to estimate the extent of overbreak due to a fault near a stope surface, and the parameter for measuring fault-related overbreak is defined as the increase in overbreak for a stope with a fault compared to one without a fault. The method can be used in estimating active support lengths and capacity requirements. A fault factor can be determined based on overbreak due to the fault, that can be used to revise the stability number N' to account for the presence of faults near a stope surface. The effect of soft fault gouge can be accounted for in the stability number by relating the width of the soft zone to that of the surface it intersects. - The stability graph method is applicable to weaker orebodies, than those from which it was originally developed are; however the stope backs require support to prevent progressive caving of the backs in such cases. - The likelihood ratio is a suitable statistical tool for use in interpreting the stability graph because it can be used to: -- Optimally define the boundaries between zones in the stability graph based on misclassification cost and inequality of data subgroups, -- Prove overlap of the stability graph zones and that the zones defined as stable, unstable and cave are not absolute, -- Determine the magnitude of errors that can be made in predicting stope performance when the stability graph is used, and -- Develop likelihood-based stability graphs that are useful for new mines and on-the-spot design. - A stability index graph that includes rockmass parameters omitted in the conventional stability graph, and based on modified span, is presented as an alternative open stope design method.

The application of the results of this thesis in the mining industry should lead to improved understanding of fault effects on open stope stability and increased confidence in the use of the stability graph as a robust empirical design tool.