The Weakest Link - Ground Support Observations at Some Canadian Shield Hard Rock Mines

The design of ground support systems is complicated by the fact that a weak link in the system can cause premature failure. Pull tests and/or dynamic test results may not be representative of how the tendon/plate/retainment system interacts with the rock mass. For example it is relatively common to see rockburst damage where the rock has shattered around the tendons, with only superficial damage to the tendons themselves. The ability of the retainment system to transfer load to the tendon elements may be critical to the ultimate success of the system, especially in bursting ground. Variable stiffness of different components can also lead to premature failure of one element prior to utilising the full capacity of another. Underground observations for both static and dynamic loading are used to discuss potential improvements and limitations of commonly used support elements (friction bolts, resin rebar, mechanically anchored rockbolts, cablebolts, weld mesh screen, plates, shotcrete). The observations are predominantly from Xstrata’s Craig Mine in Sudbury, Ontario, Canada, with some examples from other Xstrata Canadian operations. The minimum support standards at the various mines are more than adequate for most conditions. However the exceptions (where something fails) often provide more insight and the observations are focused on these cases. The mining methods are a mix of post pillar cut and fill and blasthole open stoping, utilising trackless development where tunnel size is 5 x 5 m or larger. For “static” or pseudo static loading many of the observations come from cut and fill mining where breasting through the previous cuts’ back offers an opportunity to see a cross section of both the stress induced fracturing and the previous cuts’ back support. The mines are all hard rock Canadian Shield operations where the horizontal stress is approximately double the vertical (vertical stress being predominantly due to the weight of overburden). Some rockburst damage is shown and a brief recap of some dynamic testing is given to illustrate that at least some of the support design issues can be simulated, even if the detailed response of a particular rockburst cannot be practically duplicated.