Why the study matters
Mechanical cutting offers a possible route towards more continuous hard-rock mining, but cutting tools and models developed for relatively homogeneous rock cannot simply be assumed to work in platinum-bearing UG2 reef. UG2 changes markedly over millimetre-scale distances and between mines. The dissertation therefore asks two connected questions: how can its cutting response be measured repeatably in the laboratory, and how reliably can a numerical model reproduce that response?
Experimental platform and method
A laboratory linear cutting machine was designed, manufactured and commissioned as part of the research. It moves an instrumented conical pick through a rock sample while recording normal and drag forces. Sandstone trials established that the machine produced credible, repeatable measurements before the programme moved to heterogeneous UG2 samples.
The experiments varied cutting depth and cut spacing. Force histories, specific energy and fragment characteristics were compared across materials and operating points. This produced the calibration evidence for a continuous surface cap material model in LS-DYNA.
What the cutting tests showed
Cutting depth had a particularly strong influence on UG2 response. At a depth of 4 mm, normal force, drag force and specific energy tended to level off once the spacing-to-depth ratio became sufficiently large. Tests at 2 mm produced powder-like fines in both sandstone and UG2. Increasing the sandstone cut to 4 mm produced larger chips, but UG2 continued to fragment finely.
That difference has an operational consequence: even if the rock can be cut, very fine material may be difficult to clean efficiently from a narrow stope. The work therefore identifies fragmentation and material handling—not only pick force—as a central feasibility question.
Frequency-domain analysis was useful when the sandstone signal contained regular, repeating structure. It was less suitable for the non-periodic response of heterogeneous UG2, demonstrating why signal-processing methods must be matched to the physical character of the material.
Numerical modelling result
The LS-DYNA model could be calibrated to represent one experimental cutting condition. The same parameter set did not, however, retain its accuracy when cutting depth changed. The result is an important boundary on interpretation: the model is useful for studying a calibrated case, but it is not yet a general predictive tool for UG2 cutting.
Engineering implications
The dissertation delivers an experimental platform, a carefully characterised UG2 dataset and a clearer statement of the modelling challenge. It recommends broader cutting trials, a model that generalises across operating conditions and focused investigation of fine-fragment removal. Together, these are practical next steps towards assessing whether mechanical cutting can support continuous, safer platinum mining.