24-MMP-A2 Underground Mining Methods and Design · December 2017
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
09-MMP-A2 Underground Mining Methods and Design — National Exam, December 2017. Compulsory Question 1 (Section A, 40 marks) plus three optional questions (two from Section B, one from Section C) constitute a graded 100-mark paper; every optional question (2–6) is answered in full below as a complete study resource.
Reference texts: Hartman, H. & Mutmansky, J., Introductory Mining Engineering, 2nd ed., Wiley (2002); Hartman, H. (ed.), SME Mining Engineering Handbook, 2nd/3rd ed., SME; Hartman, H., Mutmansky, J., Ramani, R. & Yang, Y., Mine Ventilation and Air Conditioning, 3rd ed., Wiley (1991) — the three texts named on the exam's own reference line.
Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.
Backfill becomes essential rather than optional wherever a mining method removes ore in a sequence that leaves large, permanently open voids the surrounding rock cannot safely span on its own. Two conditions drive this: the MINING METHOD itself — cut-and-fill (each lift needs a filled floor for the next lift's equipment and ground control before the next slice is taken), sub-level/post-pillar cut-and-fill and paste-fill room-and-pillar (fill substitutes for the pillars being recovered), and any method sequencing multiple adjacent stopes where a neighbouring void must be filled before the next primary stope can be safely extracted (primary-secondary stoping sequences) — and the HOST-ROCK CHARACTERISTICS: weak, closely jointed, or high-stress ground that cannot stand an unsupported span for the mine's planned stope life, ground prone to squeezing/rock-burst where the fill's passive confinement measurably reduces convergence and burst risk, and orebodies close enough to surface or to other workings that unsupported subsidence would be unacceptable. Where the host rock is very competent and the method (e.g. sub-level open stoping into a large permanent void, or block caving where the intent IS controlled collapse) does not require the void to be closed, backfill is not essential and is often omitted for cost.
The mill produces backfill from its own reject/tailings stream by three broadly distinct routes. Classification (hydraulic fill): whole-tailings slurry is passed through hydrocyclones to remove the finest slime fraction (which drains too slowly and reduces strength), leaving a coarser, faster-draining sand product pumped underground as a slurry. Thickening (paste fill): the FULL tailings stream (including the fines the cyclone would reject) is dewatered in a high-rate/deep-cone thickener to a non-segregating paste consistency (70–85% solids), then mixed with a metered cement/binder addition in an in-line mixer immediately before pumping. Crushing/screening (rock fill): waste rock from mine development, or purpose-quarried aggregate, is crushed and screened to a design gradation, then either cemented in place underground (grout added as a separate pour) or delivered pre-mixed as cemented rock fill. All three routes are typically also augmented with a metered cement/binder dosing system (silo, weigh-feeder, in-line mixer) common to whichever base material is chosen.
The mill's tailings/fill product is generated essentially continuously at a near-constant rate tied to the daily milling throughput, while the mine's fill demand is inherently batch and intermittent — only when a specific stope has finished its ore extraction and is ready to receive fill, on a schedule set by the mine plan rather than the mill's production rate. The mismatch is absorbed with surge/buffer capacity: a surface fill plant with agitated storage tanks or a thickener underflow surge bin holds several hours to a day's production so filling can be switched on and off to match whichever stope is ready, without ever stopping or restarting the mill's continuous reject stream; underground, a distribution manifold with multiple valved branch lines lets the fill crew route the (still essentially continuous) pumped stream to whichever stope is currently being filled, closing that branch and opening the next when one stope reaches capacity. Where fill demand genuinely cannot keep pace with continuous mill production (e.g. during a period when few stopes are ready), the surplus reverts to the surface tailings storage facility rather than backing up the plant.
Cement (ordinary Portland or a blended/slag cement) is the standard binder, giving the most predictable, fastest-curing strength gain but at the highest per-tonne cost, which matters acutely in remote mines where cement must be trucked, barged or flown in and can be the single largest fill operating cost. Fly ash and (blast-furnace) slag are industrial by-products that can replace a substantial fraction of the cement (often 30–70% by mass) at lower cost where a nearby power plant or smelter makes them available; they cure more slowly than pure cement (pozzolanic/latent-hydraulic reaction) but can match or exceed long-term (90–180 day) strength, a good trade in remote mines with the transport logistics to source them but not full-cement quantities affordably. Ground glass (recycled, pozzolanic when finely ground) is a niche, regionally available substitute in the same family as fly ash/slag. Flocculants are not strength additives at all but dewatering aids, dosed into the thickener feed to accelerate settling and raise the underflow solids content, indirectly improving fill strength by allowing a lower binder-water ratio at the same pumpability. Typical mixes of 8 parts tailings : 1 part cement (a relatively rich, roughly 11% cement content) are used for structural/pillar-replacement fill in sub-level or post-pillar mining where the fill itself must carry significant load, while the leaner 32 parts tailings : 1 part cement (≈3%) mix is used for simple void-filling/ground-support fill (cut-and-fill floors, backfilled secondary stopes) where only enough cohesion to stand a vertical exposed face is needed.
The two broad types of mill-produced backfill are hydraulic (classified sand) fill, the cycloned, coarser fraction of tailings delivered as a free-draining slurry, and paste fill, the full, dewatered tailings stream delivered as a non-segregating paste. Both are commonly augmented with gravel from local pits when a coarser, higher-strength, more permeable fill is wanted for a specific application (e.g. a drainage layer at the base of a fill mass, or bulking out a fill volume where tailings alone would be too fine-grained/slow-draining) — the gravel is blended in at the surface fill plant or, less commonly, trucked directly into the stope.
A straight vertical pipe run from a surface fill plant to a working level 1,000 m deep would expose the LOWER pipe sections to close to the full static head of a 1,000 m fill-slurry column (roughly 9–10 MPa), far beyond conventional steel or HDPE pipeline pressure ratings. Two established solutions are used, singly or together: a reticulated/telescoped borehole system, in which the fill is dropped down a large-diameter, near-vertical borehole (unlined or steel-cased) rather than a pressurised pipe — the slurry falls essentially in free flow/spray rather than under confined pressure, so the borehole itself sees little more than its own weight of falling material, and pressure only builds up again once the fill is picked up in a shorter, lower-pressure pipe run at depth; and friction/reduction fittings in the pipe string (energy-dissipating orifice plates, reducers, or a deliberately non-straight, multiply-bent pipe route) that convert a controlled fraction of the static head into friction loss over the pipe's length, so the pressure at any point in the string stays within the pipe's rating even though the total elevation drop is large. In practice a mine commonly combines the two: a borehole for most of the vertical drop, transitioning to a shorter conventional pressure-rated pipeline (with reducers/bends as needed) for the final horizontal run out to the stope, exactly as sketched in Fig. 4.6.
Production. The mill's full tailings stream (not just the coarse cycloned fraction) is dewatered in a high-rate or deep-cone paste thickener to a non-segregating consistency (typically 70–85% solids by weight, well above the ≈40–50% of ordinary hydraulic fill slurry), then metered cement/binder is blended in an in-line or batch mixer immediately before delivery. Transport. Because paste behaves as a non-Newtonian, plug-flow material rather than a free-flowing slurry, it CANNOT be delivered by gravity/hydraulic pipeline the way classified sand fill can — it is pumped through the pipeline under positive pressure by heavy-duty positive-displacement (piston) pumps, which can push paste both vertically down and, where needed, back up or long horizontal distances that a gravity system could never reach. Placement. The paste is discharged directly into the stope through the pipeline outlet or a short distribution boom, and because it does not segregate or bleed significant water, it can be placed against a much simpler, more permeable barricade than hydraulic fill needs (no requirement to hold back a large drained-water volume).
Advantages over conventional fill. Consuming the ENTIRE tailings stream (not just the coarse fraction) both minimises the tonnage that must go to the surface tailings storage facility — a major cost, environmental-footprint and long-term closure-liability benefit — and, because no fines are rejected, gives more consistent, controllable strength; the non-segregating paste also achieves useable strength with far less mixing water than hydraulic fill, and because it does not bleed large volumes of water it can be used in sequences (e.g. rapid successive lifts) that would drown a bleed-water-heavy hydraulic fill barricade. Difficulties. Achieving and holding the correct water content is the central challenge: too wet and the paste behaves like a slurry (segregates, cannot be pumped efficiently, loses strength); too dry and pump pressures/pipeline wear rise sharply and blockages become a real risk — so the plant needs tight, continuous density control on a highly variable tailings feed. Size distribution matters similarly: too coarse a gradation causes the paste to segregate and behave like ordinary hydraulic fill despite the high solids content, while too fine (excess ultrafine clay-sized particles) increases water demand and can make the paste unpumpably viscous at any reasonable solids content, so the mill must actively manage (not simply accept) the grind/classification circuit's product gradation for paste-fill service.
Fill protects against rock pressure, blasting shock and seismic/earthquake loading through several distinct mechanisms depending on type and placement. Passive confinement: any fill (even unconsolidated hydraulic sand) placed tightly against a stope wall limits the wall rock's ability to dilate/bulk outward under stress, which measurably increases the confined rock mass's own residual strength (a Mohr-Coulomb effect — confining pressure raises the shear strength the rock can sustain) even without the fill itself being load-bearing. Structural (pillar-replacement) support: cemented fill (rock fill or paste, at the higher binder contents discussed in 4.4) is placed specifically to CARRY load, replacing a rib or sill pillar that has been or will be mined out — this is what makes post-pillar and sub-level cut-and-fill recovery of pillars possible at all. Regional stiffness/energy absorption: in burst-prone or highly stressed ground, a continuous fill mass (even moderate strength) stiffens the overall stope/pillar system, reducing the span over which sudden pillar/wall failure can propagate and absorbing some of the kinetic energy of a seismic event or rockburst rather than letting it transmit unimpeded into open voids — the denser, better-consolidated the fill (paste and cemented rock fill more than loose hydraulic sand), the more effective this energy-absorbing role is.
Saturation reduces backfill strength through several linked mechanisms: excess pore water reduces the effective stress carried by the cemented matrix (the classical soil-mechanics effective-stress principle, σ' = σ−u, applies directly to a cemented fill mass), a saturated fill has essentially no residual internal friction contribution from particle-to-particle contact (pore pressure carries load that would otherwise go through the grain skeleton), and prolonged saturation of a young, still-curing cemented fill can leach cement hydration products before they have developed full strength, permanently degrading the design strength the fill would otherwise reach. The resulting damage risk is twofold: a saturated fill mass can simply FAIL under its own weight or the load it was meant to carry (loss of an intended pillar-replacement or wall-confinement function, potentially triggering a wall collapse or pillar failure it was placed to prevent), and, in the acute case, an inadequately drained, over-saturated hydraulic or paste fill mass behind a barricade can LIQUEFY and flow, a barricade-failure/fill-flow hazard that has caused fatal inundations in mining history — which is why fill-system design always includes deliberate drainage provision (barricade permeability, underdrainage) to prevent saturation building up in the first place, not just adequate binder content.
