By: Guillermo Condori Ceron, Catalina Huanca Sociedad Minera; Raúl Castro Ruiz, Universidad de Chile, and Lenin Arancibia Guevara, BCTEC Ingeniería y Tecnología.

Abstract

Catalina Huanca Sociedad Minera is a polymetallic mine with a production of 2,400 metric tons per day (TPD). Within the mining plans, there is a mineralized zone with 1,016 KTon of mineral resources (Mariela 4, Doña María Piso 6, and Melisa 3) with average grades of 10.8% Zn and 2.6% Pb, for which a mining method had not yet been defined. The RMR of the mineralized zone of the Melisa 3 orebody is in the range of 20-40, which complicates mining with the traditional methods used at the mine (Cut and Fill, Sublevel Stoping), since it requires sacrificing good ore pillars to maintain stope stability. In some cases, stability simulations did not ensure the optimal safety factors for the safe mining of this zone.

Given the mining difficulties in the Melisa 3 zone, a review was initiated of the experiences with the application of the Sublevel Caving (SLC) method. This mining method is not commonly used in Peru; therefore, evaluations were conducted regarding stability, ore recovery, potential surface subsidence, and ore extraction rates, in order to analyze the applicability and benefits of the method at Catalina Huanca.

The simulations and calculations demonstrated that SLC was the best mining alternative for this zone. In order to refine the parameters used in the calculations, a pilot stope was developed in an adjacent area (Mariela 4) with representative characteristics, so that the operation could gain training in the implementation of the method and confirm the design parameters used.

The paper describes the results of the Pilot Stope mined in Mariela 4 and the applicability of the SLC mining method at Catalina Huanca, including design, planning, and stability analysis in the Melisa 3 sector. 

Keywords: Sublevel Caving, RMR 20 to 40, Stability Analysis.

Introduction

Background

The Catalina Huanca (CH) mine is located in the Andes Mountain Range, in the central region of Peru. It belongs to the Canaria district, Víctor Fajardo province, Ayacucho department, at an altitude of 3,500 masl. Figure 1 shows the location of Catalina Huanca.

Catalina Huanca has mined ore bodies hosted in sedimentary rocks (limestones, conglomerates, and sandstones) with RMR values ranging from 40 to 60, which have been exploited using conventional Room and Pillar and Sublevel Stoping methods. In 2021, three semi-horizontal, overlapping ore bodies were discovered, which generated the resources indicated in Table 1 (90% of these resources are classified as measured or indicated).

The main characteristic of these structures is that they are hosted in a structurally weak zone that encompasses three types of rock (limestone, conglomerate, and sandstone). The ore bodies and their walls (hanging wall and foot wall) have RMR values ranging between 20 and 40 (Melisa 3 and Doña María Piso 6), as shown in Figure 2. In past experiences, Catalina Huanca has mined zones with RMR values between 40 and 45 using the Room and Pillar method, but it has not mined ore with lower RMR values using the aforementioned methods.

Problem Statement

There are 1.06 million tons of resources with good zinc and lead grades which, according to initial evaluations, cannot be mined using the conventional methods at Catalina Huanca mine while maintaining stability safety factors and ensuring acceptable ore recovery.

Objective

To implement Sublevel Caving mining in the Melisa 3 zone after the corresponding evaluations and designs.

Scope

For the implementation of the Sublevel Caving mining method at CH, the following stages will be carried out:

ν Conduct a trade-off study of the mining method.

ν Define the Sublevel Caving mine design.

ν Develop the mine plan and economic evaluation.

ν Perform 3D geomechanical numerical modeling to evaluate the stability of the mining method.

ν Recommend ground support for the mining method.

Evaluation of the Sublevel Caving Method

Selection of Mining Method

First, a comparison of mining methods was carried out using Nicholas’ method (1981). This methodology is based on assigning a score according to the following characteristics:

ν Factor A: Ore deposit characteristics. Shape, thickness, dip, and grade distribution.

ν Factor B: Characteristics related to the competence of the ore rock, fracture spacing, and type of fracture infill.

ν Factor C: Characteristics related to the competence of the hanging wall rock, fracture spacing, and type of fracture infill.

ν Factor D: Characteristics related to the competence of the foot wall rock, fracture spacing, and type of fracture infill.

Table 2 shows the ranking of mining methods for the Melisa 3 and Doña María Piso 6 zone.

This analysis, based on Nicholas’ (1981) methodology, shows that the Sublevel Caving mining method is the most feasible to implement due to its geometric, geomechanical, and cost-related characteristics. However, in order to implement the method, it must first be designed, planned, and its stability evaluated in greater detail, prior to the implementation of a Pilot Stope.

Pilot Stope

With the objective of evaluating the suitability of the mining method, Catalina Huanca carried out a Pilot Stope in Mariela 4. This stope was mined, and it was recorded that the open area, with a hydraulic radius of 6 m, was sufficient to generate caving, which is consistent with the behavior reflected in the numerical modeling, as shown in Figure 3.

The operation conducted strict control of drilling, blasting, and ore extraction to monitor stability and the propagation of the caving, avoiding possible airblast risks. Figure 4 shows part of the operational control carried out for the proper extraction of the Pilot Stope.

Definition of Sublevel Caving Mine Design 

The definition of the mine design is based on a benchmark carried out in SLC operations similar to Catalina Huanca and on an analysis of ore flow in SLC.

To define the mine design, the following aspects are established:

ν Type of design.

ν Distance between levels.

ν Distance between drifts.

ν Number of levels.

ν Number of drifts at the same level.

Type of Design

A fan-shaped design is defined. This design has a lower height and seeks to achieve complete interaction of ore flow at an extraction level, as shown in Figure 5.

Distance Between Levels

The Kvapil (1982) methodology is used to determine the distance between levels, establishing the following relationship:

Where:

Ht: Maximum total drilling height plus drift height. Considering a drilling equipment diameter of 76 mm at CH, a maximum drilling length of 22.5 m is achieved (field data).

Hg: Drift height: 4m x 4m, according to the benchmark conducted for mines with similar conditions.

Therefore:

Distance between levels

Distance Between Drifts

Based on Kvapil’s (1982) methodology, the distance between drifts at the same level considers the following:

ν Geometry of the extraction point.

ν Expected fragmentation.

ν Ensuring the interaction of movement zones.

It is determined that, in terms of gravitational flow, the recommended drift spacing may range between 10 and 14 m. A comparative analysis was conducted between both designs, and the option of 14 m between drifts was selected, as it ensures good gravitational flow interaction and allows for larger pillars, which in turn provides greater stability under fair to poor rock conditions.

Number of Levels and Drifts per Level

In an initial approach, based on the mineralization distribution, 6 levels were established: one in Mariela 4 (Stope 520) and five in the Melissa 3 and Doña María Piso 6 orebodies, as shown in Figure 6. Figure 6 shows pilot stopes 1, 2, and 3, which are SLC method test extraction units that indicated that the rock had good caveability and confirmed the applicability of the method.

In Table 3, the data by level in the preliminary design are shown.

Mining Method for Catalina Huanca Based on the Mine Design

Figure 7 shows a schematic of the mining method for the Melisa 3 sector in Catalina Huanca. Figure 7 illustrates two moments of the mining operation: on the left, the period where only extraction from the Mariela 4 level is considered; and on the right, a more advanced stage where extraction is already taking place from level 2,763, which implies a larger caving volume both in the basal area covered and in the vertical extent reached, connecting with the caving generated by extraction from level 2,854 of Mariela 4.

Production Plan Evaluation

FlowSim SLC

At this stage, in addition to defining the production plan, the mine design is refined in terms of the number of levels and drifts per level. To perform this analysis, FlowSim SLC is used.

FlowSim is a gravity flow simulator applied to underground operations mined by Sublevel Caving methods, developed by BCTEC. It has been validated through laboratory experiments and mine-scale applications (Valencia, 2013; Fuentes, 2015; Castro et al., 2006, 2018). FlowSim SLC considers a sequence of extraction fans or rings, which constitute the basic mining unit in Sublevel Caving. It enables the creation of production plans by incorporating planning criteria. 

Figure 8 shows a gravity flow simulation in FlowSim SLC for Catalina Huanca in Melisa 3, where the evolution of the flow can be observed along with the production plan for the extraction of different levels, thus allowing the estimation of extractable tonnage, extractable grades, and mining recovery. This type of visualization makes it possible to study the evolution of the flow and to identify potential zones of remnant ore. 

Mine design refinement and production plan development

In order to develop a production plan, certain geomechanical and planning criteria must be defined and tailored to the mining method to be implemented. In this case, the criteria applied are as follows:

ν Extraction is carried out level by level, in a downward sequence. Within the same level, the fan-shaped extraction sequence follows a V-pattern, starting from the central drift and advancing toward the bypass, as shown in Figure 9.

ν The estimated tonnage per fan is 1,250 tons, with a minimum extraction of 400 tons.

ν For the Melissa 3 and Doña María Piso 6 areas, plans were evaluated that allowed extracting up to 150% of the fan tonnage.

ν The extraction rate is 420 tpd during the first six months, increasing to 720 tpd from the seventh month onwards, without exceeding 400 tpd per fan.

ν The cut-off NSR for sending ore to the plant is USD 40/t, while the cut-off NSR for sending ore to stockpile is USD 30/t.

One of the major design changes under evaluation is the development of Level 2,811 (Figure 6), which lies in a geomechanically complex environment. This condition could lead to safety issues, higher costs, or delays in level development or production, among other risks. The second observation that gave rise to the four design alternatives was the assessment of incorporating a new bypass at Level 2,795, located 22 meters behind the current bypass (Figure 10), with the aim of increasing resource recovery at this level.

The combination of these two considerations resulted in the four designs presented in Table 4. For each design, an optimized production plan was generated and compared in terms of economic benefit and resource recovery. It should be noted that the tonnage, dilution, and NSR values were derived from flow simulations performed in FlowSim SLC.

The selected design is D, as it maximizes recovery and tonnage while minimizing the developments to be carried out. The production plan contemplates one level in Mariela 4 (Level 2,854) and four levels in Melisa 3 and Doña María Piso 6, totaling 2,100 meters of development and 561 extraction fans. Figure 11 shows a comparison between the initial design proposal (with 5 levels in Melisa 3) and the final optimized design to which the presented production plan corresponds (with 4 levels in Melisa 3).

Evaluation of Caving Propagation 

One of the key concerns for the implementation of a caving method at Catalina Huanca is to ensure that the caving does not reach the surface and that the crown pillar formed between the caved void and the surface remains stable.

The following methodologies were used to estimate caving propagation:

ν 2D modeling, using the methodology proposed by Flores (2004).

ν Gravitational flow modeling with FlowSim SLC.

ν Evaluation of crown pillar stability using the methodology proposed by Carter (2014), which is recommended by the Peruvian Ministry of Energy and Mines (2017).

Estimation of Caving Propagation 

Caving propagation was estimated using a numerical model (Flores, 2004) and flow simulations with FlowSim SLC, which indicated that caving propagation should occur up to 400–550 m from the first level of Melisa 3 (elevation 2,795). To verify that the assumptions of the caving propagation estimation are met, instrumentation will be installed to measure caving propagation. Caving will be monitored using TDR and seismicity measurements.

Evaluation of Crown Pillar Stability

The methodology of Carter (2014) relates the geometry of the crown pillar (through the concept of scaled thickness) with rock quality. Considering that rock quality improves at shallower depths (closer to the surface), with RMR values ranging from 45 to 55, and that the scaled thickness (Cs) is 2.1 m, the crown pillar would fall into the stable zone with a low probability of collapse, as shown in Figure 12.

3D Numerical Modeling for SLC

After defining the mine design and production plan, the stability of the pillars in the SLC zone was evaluated. To analyze stability, a 3D simulation was conducted using the finite element software RS3 to determine the magnitude of induced stresses, stress influence halos, and safety factors.

Input Data for 3D Numerical Modeling 

For this simulation, the following were considered:

ν In-situ stress conditions, which were estimated during the engineering phase (USACH, 2023).

ν Geological-geotechnical model: CH has conducted at least four drilling campaigns to estimate rock properties. These properties were included in the numerical model.

ν Mine design: the 5-level mine design for Sublevel Caving extraction was considered, as well as critical infrastructure at Catalina Huanca, such as the South Tunnel, the integration ramp 3050, the integration ramp 2750, and the exploration drift, as shown in Figure 13.

ν Production Plan and Sequencing: The production plan was considered along with the extraction sequencing of the Sublevel Caving zone.

3D Numerical Modeling Results 

At Mariela 4, the major principal stresses (σ₁) were estimated, which should not exceed 72 MPa, and in several zones values between 37 and 50 MPa were observed, as shown in Figure 14. It is worth noting that this level is characterized by the presence of conglomerate with a strength of 100 MPa; therefore, the rock’s compressive strength is not being exceeded by the induced stresses. However, ground support control must be rigorous.

At the Melisa 3 levels, the average values of the principal stress (σ₁) are around 40 MPa, with some zones where this value increases to approximately 52 MPa, as shown in Figure 15. It is worth noting that this zone is characterized by the presence of the Sandstone unit, which has a UCS of 78 MPa.

Regarding the safety factor, it is calculated based on the Hoek-Brown failure envelope (Hoek et al., 2002). Figure 16 shows the distribution of the calculated safety factor (SF) at the first level of Melisa 3 (elevation 2,795 masl) for an initial and final extraction state. From this figure, it can be determined that initially the SFs are above 1.7. Upon extracting the mining fans, the SF values decrease to 1.3 at the front, while the bypass sector remains at 1.7.

Based on this analysis, it can be observed that, as extraction has not yet begun, the drifts are in a stable state. On the other hand, when stresses concentrate on the production front due to extraction, safety factors are in the order of 1.3. It can therefore be concluded that the design would be stable at all locations and times in the production plan. 

Ground Support in the SLC Zone 

Given the rock quality conditions and the induced stresses at Catalina Huanca, the necessary ground support is determined based on a hybrid methodology (Wattimena, 2003), which combines numerical modeling to determine induced stresses with empirical charts (Barton, 1988). Additionally, experience from caving mines in other parts of the world was incorporated.

The ground support system is based on the use of:

νShotcrete: Applied to the walls and roofs in “fresh rock,” with an additional layer placed over the steel mesh to provide extra protection and improve sector stability.

νBolts: Helical bolts are recommended, as they provide better permanent support.

νSteel mesh: In mining, this addresses deformations. Woven or chain-link mesh is primarily used, which helps better contain possible deformations, as the mesh acts collaboratively, distributing point loads to adjacent nodes.

νCable anchoring systems: This type of support is used in specific zones affected by faults.

Two types of ground support are recommended depending on the productive level zone, as shown in Table 5 and Figure 17.

Conclusions

1. Caving methods, particularly Sublevel Caving, are not commonly used in Peru due to ore deposit conditions. This is further compounded by specific constraints that make it difficult to connect the caving to the surface, given the presence of nearby communities. There is experience with SLC at Yauricocha in Peru. 

2. Catalina Huanca, owned by Trafigura, is therefore one of the pioneers in implementing the SLC method. This implementation follows a detailed study of the method’s applicability in the mineralized bodies of Mariela 4, Melisa 3, and Doña María Piso 6. This mining method is recommended for these orebodies given their geotechnical characteristics, geometry, stress conditions, costs, and recovery potential. To evaluate the suitability of the method under the conditions at Catalina Huanca, a Pilot Stope was designed and extracted at Mariela 4, which was successfully mined and resulted in a controlled propagation of the caving.

3. It is recommended that the methodology presented here be replicated in other projects using a theoretical design basis (numerical modeling with FlowSim SLC and RS3) and implementation in pilot stopes, which allowed maximizing project value while maintaining high standards of safety and physical stability integrity.

Acknowledgements 

We extend our sincere thanks to the staff at Catalina Huanca for providing the input data, as well as for their drive and enthusiasm in implementing the mining method, especially Jonhy Orihuela, Gorki Román, Hernán Pantaleón, Jesús Ascarza, Gustavo Cruzado, and Jorge Poma. Additionally, we would like to express our gratitude to the BCTEC Team, specifically Pablo Álvarez, Alejandro Espinosa, Brandon Flores, Rodrigo Lazo, and Marcelo Ávila, for their collaboration and support.

References

Barton, N. 1988. Rock mass classification and tunnel reinforcement selection using the Q-system. In Rock classification systems for engineering purposes. ASTM International.

Castro, R 2006, Study of the Mechanisms of Gravity Flow for Block Caving, PhD thesis, The University of Queensland, Brisbane.

Castro, R., Arancibia, L., Guzman, D., Henriquez, JP., 2018. Experiments and simulation of gravity flow in block caving through FlowSim. In Caving 2018: Proceedings of the Fourth International Symposium on Block and Sublevel Caving, pp. 312-322.

Carter, T; Castro, L, 2017. Guía para la evaluación de la estabilidad de los pilares corona (Guide for the Stability Assessment of Crown Pillars). Dirección General de Asuntos Ambientales Mineros, Republic of Peru, Ministry of Energy and Mines.

Carter, T. G., 2014. Guidelines for use of the scaled span method for surface pilar corona stability assessment. Ontario Ministry of Northern Development and Mines, Ontario, 1-34.

Flores, G., 2004. Rock mass response to the transition from open pit to underground cave mining. PhD Thesis. University of Queensland. 

Hoek E., Carranza-Torres C. & Corkum B. 2002. “Hoek-Brown failure criterion – 2002 Edition”. Proceedings of the 5th North American Rock Mechanics Symposium and 17th Tunneling Association of Canada Conference. Canada.

Kvapil, R., 1982. The mechanics and design of sublevel caving systems. Underground mining methods handbook, p. 880-897. 

Nicholas, 1981. Method selection – A numerical approach. Design and operation of caving and sublevel stoping mines, pp. 39-53. 

USACH, 2023. Informe técnico: Determinación del tensor de esfuerzos #1 y #2 mediante emisión acústica y análisis tasa de deformación (Technical Report: Determination of Stress Tensors #1 and #2 Using Acoustic Emission and Strain Rate Analysis).

Valencia, M. 2014. Desarrollo e implementación de FlowSim para su aplicación en minería de Block/Panel Caving (Development and Implementation of FlowSim for Application in Block/Panel Caving Mining). Bachelor Thesis, Universidad de Chile.

Wattimena, R., 2003. Designing undercut and production level drifts of block caving mines. PhD thesis, University of Queensland.