Comminution Circuit Design and Simulation for

the Development of a Novel High Pressure

Grinding Roll Circuit

by

Persio Pellegrini Rosario

A thesis submitted in partial fulfillment of the requirements for the

degree of

Doctor of Philosophy
in

The Faculty of Graduate Studies

(Mining Engineering)

The University of British Columbia

(Vancouver)

November 2010

Persio Pellegrini Rosario, 2010

ABSTRACT

The application of High Pressure Grinding Roll (HPGR) in comminution circuits is well

established in processing cement, diamonds and iron ore. Recently, the application of

HPGR has been extended to high-tonnage precious and base metals operations with

hard ore. This is due to the HPGR: being more energy-efficient than grinding mills, not

requiring steel grinding media, and providing higher throughputs than cone crushers.

Although HPGR circuits are being used in high-tonnage precious and base metals, there

is limited quantitative knowledge to indicate the true benefits or drawbacks of HPGR

compared to Semi-autogenous mill (SAG). This lack of knowledge restricts the ability of

designers to determine the optimal circuit. To address this lack of knowledge the

research in this thesis:

Reviews the basics of the HPGR machine, its benefits and shortcomings.

Details the development of the SAG circuits and explains how the new

generation of crushing circuits, with HPGR for tertiary crushing, are starting to

replace SAG circuits in hard-rock mining.

Presents a structured methodology for comparison of the energy requirements

for HPGR versus SAG complete circuits. The process is based on industrial best

practices and advanced modelling tools, and is demonstrated through the

evaluation of two hypothetical mining projects (based on real ore data).

Investigates the feasibility of a novel AG-Crusher-HPGR circuit using rock

samples from a large copper-gold mining project. The approach was to develop

and evaluate the circuit design for high-tonnage operations with mixed hardness

ii
 ores containing clay. Previously, HPGR was considered only suitable for very

hard ores and the technology was rejected for other cases. A unique pilot-plant

test program was developed as a basis for experimental simulation. As a result

the suitability of the circuit was demonstrated.

The development of this novel circuit along with the findings of this research have the

potential to improve future mining operations dealing with similar orebodies that, in fact,

are major sources of base metals worldwide. The potential for significant savings in

energy and steel media have been demonstrated. This may also lead to the selection of

more sustainable circuits for a broader range of orebodies.

iii
PREFACE

Prof. Robert Hall is my PhD program supervisor and co-authored two manuscripts

(Chapters 3 and 4). Prof. Hall provided feedback on manuscript preparations and

contributed to the identification and design of my research program.

Prof. Bern Klein is my PhD program co-supervisor and co-authored the third manuscript

(Chapter 4). Prof. Klein provided input on the design of the testwork program applied in

this research as well as participated in the identification and design of the research

program.

Mr. Mike Grundy was a co-author of two manuscripts (Chapters 2 and 4). Being a senior

metallurgist with vast experience in AG/SAG mill application, he assisted me in the

clarification of parts of the manuscript, especially the ones covering the history and

recent applications of SAG circuit. In addition, he provided feedback during the

development of the novel HPGR circuit, verifying a number of my assumptions and

assisting me in specific engineering details for the operation of a circuit.

Mr. Ken Boyd was a co-author of the first manuscript (Chapter 2). Being a senior

mechanical engineer specialized in material handling systems he contributed with

information regarding the application of pebble crushers in recent SAG mill circuits.

The contributions of all the people above mentioned was important and very much

appreciated. However, the vast majority of the research and writing was conducted or

developed and directed solely by the author, i.e. more than 95% of the work. This

included the following:

Development of the research objectives, methodology and testing programs.

iv
 Performance of all simulation analysis.

Performance of all test work with support from laboratory personnel for some

manual labour and specialized tasks.

Review of the current state of the art as presented in the thesis.

Rewriting and integration of the papers into the current form in the thesis with

revisions based feedback from my supervisory committee.

v
TABLE OF CONTENTS

Abstract ................................................................................................................................. ii
Preface................................................................................................................................. iv
Table of Contents ................................................................................................................. vi
List of Tables........................................................................................................................ ix
List of Figures........................................................................................................................ x
Acknowledgements .............................................................................................................. xi
1 Introduction ................................................................................................................ 1
1.1 Comminution .................................................................................................. 1
1.2 Modern Metal Mining ...................................................................................... 3
1.3 HPGR in Hard Rock Mining ............................................................................ 5
1.4 Thesis Objectives ........................................................................................... 7
1.5 Thesis Outline ................................................................................................ 9
2 Comminution Circuits - Literature review ................................................................. 10
2.1 Introduction .................................................................................................. 10
2.2 Recent History of Comminution .................................................................... 11
2.3 SAG Mill Background ................................................................................... 13
2.3.1 AG/SAG Mill Machines ..................................................................... 13
2.3.2 SAG Operational Parameters ........................................................... 14
2.3.3 SAG Mill Original Circuit ................................................................... 15
2.3.4 Pebble Crushing for AG/SAG Circuits............................................... 17
2.3.5 SAG Feed Preparation ..................................................................... 21
2.3.6 Steel Wear........................................................................................ 22
2.4 HPGR Background....................................................................................... 25
2.4.1 HPGR Machine................................................................................. 25
2.4.2 HPGR Terminology and Operational Parameters ............................. 28
2.4.3 HPGR Original Circuits ..................................................................... 31
2.4.4 HPGR Precious/Base Metal Recent Circuits..................................... 35
2.4.5 Energy Savings ................................................................................ 37
2.4.6 Metallurgical Extraction Advantages ................................................. 40
2.4.7 HPGR Feed and Product Specifics ................................................... 41
2.4.8 Limitations and Disadvantages ......................................................... 42
2.5 Other Developments .................................................................................... 43
2.5.1 Increasing Machine Sizes ................................................................. 43
2.5.2 Stirred Mills ....................................................................................... 43
2.5.3 Fully Autogenous Grinding................................................................ 44
2.6 Summary of Current State ............................................................................ 46
3 Guidelines for Energy Requirement Comparisons between HPGR and SAG
Mill Circuits in High-Tonnage Hard Rock Mining ...................................................... 47
3.1 Introduction .................................................................................................. 47
3.2 Modelling and Simulation Background ......................................................... 49
3.3 Case Studies................................................................................................ 51
3.4 Design Criteria Development........................................................................ 52
3.5 Flowsheet Development ............................................................................... 54

vi
 3.6 Developed Models ....................................................................................... 56
3.7 Equipment Sizing ......................................................................................... 61
3.8 Results and Discussions .............................................................................. 62
3.8.1 Pure Comminution Energy ................................................................ 62
3.8.2 Complete Circuit Comminution Energy ............................................. 63
3.8.3 Steel Usage ...................................................................................... 64
3.8.4 Ore Variability ................................................................................... 65
3.8.5 Heating and Ventilation..................................................................... 66
3.8.6 Availability and Maintainability .......................................................... 66
3.8.7 Additional HPGR Benefits................................................................. 67
3.8.8 HPGR Circuit Drawbacks.................................................................. 67
3.9 Summary...................................................................................................... 69
4 Testwork Program for the Evaluation of a Novel HPGR-Based Circuit to Treat
Mixed Hardness Ores Containing Clays .................................................................. 71
4.1 Introduction .................................................................................................. 71
4.2 Novel HPGR Circuit for Ores Containing Clayish Material ............................ 75
4.3 Testwork ...................................................................................................... 77
4.3.1 Sample ............................................................................................. 77
4.3.2 Testwork Design ............................................................................... 77
4.3.3 Test Equipment ................................................................................ 79
4.4 Results and Discussion ................................................................................ 81
4.4.1 Sample Properties ............................................................................ 81
4.4.2 Tumbling Test ................................................................................... 81
4.4.3 HPGR Feed PSD.............................................................................. 83
4.4.4 HPGR Feed Moisture Content .......................................................... 88
4.4.5 HPGR Tests ..................................................................................... 91
4.4.6 HPGR Product Cakes ..................................................................... 102
4.4.7 Bond Ball Mill Work Indices ............................................................ 106
4.5 Summary.................................................................................................... 108
5 Feasibility Assessment of the AG-Crusher-HPGR Circuit to Treat Clayish
and/or Mixed Hardness Ores ................................................................................. 109
5.1 Introduction ................................................................................................ 109
5.2 Modelling and Simulation ........................................................................... 110
5.3 Energy Requirements................................................................................. 115
5.3.1 Ball Mill Energy ............................................................................... 115
5.3.2 Pure Comminution Energy .............................................................. 117
5.3.3 Complete Circuit Comminution Energy ........................................... 119
5.4 Operating and Capital Costs ...................................................................... 122
5.4.1 Operating Cost ............................................................................... 122
5.4.2 Capital Cost .................................................................................... 122
5.5 Discussions ................................................................................................ 124
6 Conclusions ........................................................................................................... 126
6.1 Main Research Contributions ..................................................................... 126
6.2 Future Research Opportunities .................................................................. 128
References........................................................................................................................ 131
Appendix A: Inputs Used for the JKSimMet Models .................................................... 142
Appendix B: SMC and MinnovEX SPI Test Results ....................................................... 145

vii
Appendix C: Sample Preparation and Test Flowsheet................................................... 146
Appendix D: HPGR Feed Test Blend Linear Programing ............................................ 148
Appendix E: HPGR Tests Complete Data................................................................... 150
Appendix F: AG-Crusher-HPGR Plant Layout ............................................................... 158
Appendix G: SABC Plant Layout ................................................................................... 160
Appendix H: Power Consumption Comparison .............................................................. 162

viii
LIST OF TABLES

Table 3-1: Summary of Design Criteria ......................................................................... 53

Table 3-2: Pure Comminution Energy ........................................................................... 62
Table 3-3: Complete Circuit Comminution Energy......................................................... 64
Table 3-4: SAG Mill Steel Ball Consumption ................................................................. 65
Table 4-1: Pilot-Scale HPGR Specifications .................................................................. 79
Table 4-2: Summary of Parameters and Calculated Results for Moisture Content ........ 91
Table 4-3: HPGR Tests Quick Reference Legend ......................................................... 92
Table 4-4: Summary of the Main Parameters and Results for All HPGR Pilot Tests...... 93
Table 4-5 Average Dimensions of Cakes Produced by the HPGR Tests .................... 104
Table 5-1: Simulation Results Pure Comminution Energy Requirements ................. 118
Table 5-2: Energy Requirements for the Complete Circuits ......................................... 121
Table 5-3: Capital Cost Summary ............................................................................... 123

ix
LIST OF FIGURES

Figure 2-1: Three Stages of Crushing, Rod Mill, Ball Mill ................................................ 16
Figure 2-2: SAG-Ball Mill Circuit ..................................................................................... 17
Figure 2-3: SABC Circuit ................................................................................................ 18
Figure 2-4: Open-Circuit SABC ...................................................................................... 19
Figure 2-5: SABC with HPGR......................................................................................... 20
Figure 2-6: Pre-Crushing in an SABC Circuit.................................................................. 22
Figure 2-7: Schematic of a HPGR (Napier-Munn et al, 1996) ......................................... 26
Figure 2-8: Open Circuit HPGR Closed-Circuit Ball Mill (Aydogan et al, 2006) ............ 32
Figure 2-9: HPGR Applied for Pebble Re-Crush at Empire Iron (Kawatra et al, 2003) .... 33
Figure 2-10: Re-Crush Circuit at Argyle Diamond Mines (KHD, 2008).............................. 34
Figure 2-11: Boddington HPGR (Dunne et al 2007) ......................................................... 35
Figure 2-12: Cerro Verde (Vanderbeek 2006) .................................................................. 36
Figure 2-13: Pebble Extraction and Milling ....................................................................... 45
Figure 3-1: Simplified SABC and HPGR Flowsheets ...................................................... 55
Figure 3-2: JKSimMet Screen Snapshot of Case A SABC ......................................... 57
Figure 3-3: JKSimMet Screen Snapshot of Case A HPGR ........................................ 58
Figure 4-1: Cerro Verde Flowsheet (Vanderbeek 2006) ................................................. 72
Figure 4-2: Hardness Distribution of the Deposit Based on Jk A*b Parameters .............. 74
Figure 4-3: Proposed HPGR Flowsheet for Clayish Ore ................................................. 75
Figure 4-4: UBC Pilot HPGR .......................................................................................... 79
Figure 4-5: Particle Size Distribution for the Samples as Received ................................ 81
Figure 4-6: Tumbling Test Feed and Product Size Distributions ..................................... 83
Figure 4-7: Lab-Scale Circuit to Prepare the Feed to the Pilot HPGR (open-circuit) ....... 84
Figure 4-8: PSDs for Fresh and Crushed Laboratory Screen O/S Material ..................... 85
Figure 4-9: PSDs from the Preliminary Simulation.......................................................... 86
Figure 4-10: PSDs for the Optimum Blend, Original Products and Simulated Product ...... 87
Figure 4-11: Lab-Scale Circuits Used for the Tests .......................................................... 88
Figure 4-12: Specific Throughput as a Function of Pressing Force................................... 95
Figure 4-13: Influence in Energy Consumption due to Pressing Force ............................. 96
Figure 4-14: Pressure Sensitivity Tests Feed and Product PSDs .................................. 96
Figure 4-15: F80/P80 and F50/P50 Reduction Ratios ...................................................... 97
Figure 4-16: Feed and Product PSDs for Closed-Circuit Tests ......................................... 98
Figure 4-17: Feed and Product PSDs for Full Feed and Tumbled-Screened Open-
Circuit HPGR Tests .................................................................................... 100
Figure 4-18: HPGR Test #1 Product Cake Samples....................................................... 103
Figure 4-19: Screen Oversize PSDs from the Tests for Assessment of HPGR Product
Cake Competency...................................................................................... 105
Figure 4-20: Bond Ball Mill Index Results in Different Points of the Circuit ..................... 108
Figure 5-1: Feed PSD for Circuit Modelling and Simulations ........................................ 111
Figure 5-2: JKSimMet Screen Snapshot of the SABC Circuit Simulation .................... 113
Figure 5-3: JKSimMet Screen Snapshot of the Final AG-Crusher-HPGR Circuit
Simulation .................................................................................................. 114
Figure 5-4: Ball Mill Cyclone Feed PSD from AG-Crusher-HPGR and SABC Circuits .. 116
Figure 5-5: AG Mill Feed (Combined) and Product PSDs ............................................. 117
Figure 5-6: AG-Crusher-HPGR Circuit Simplified Flowsheet ........................................ 120

x
ACKNOWLEDGEMENTS

I would like to express my gratitude to AMEC Mining & Metals, Vancouver, B.C., for the

generous support during the research period for his Doctoral Thesis. I would like to

extend a special thank you to my current and former managers, Alexandra Kozak and

Joseph Milbourne (respectively), for allowing me the significant amount of time required

to complete my research. I also would like to sincerely thank my friend and co-worker

Mike Grundy for his invaluable support and advice.

I am deeply thankful to my thesis supervisors, Prof. Robert Hall and Prof. Bern Klein, for

their guidance and patience. Members of the supervisory committee for valuable

advice. I would also like to thank the B.C. Mining Research, Koppern and the University

of British Columbia for providing the excellent research facilities needed for my work.

Of course, without my wifes encouragement, support and patience, and the love shown

by her and my three sons, this thesis would not be completed.

I would also like to acknowledge the support of the (anonymous) mining company for

supplying the sample used in experimental simulation and for co-sponsoring the

investigation of the feasibility of the novel HPGR circuit.

xi
1 INTRODUCTION

1.1 Comminution

The dictionary definition (source: Merriam-Webster Dictionary) of the verb comminute is:

to reduce to minute particles. In the mining and mineral processing industry, the term

comminution mainly refers to crushing and grinding processes, although the size

reduction of rocks starts in the blasting phase of mining.

Comminution is an essential phase in mineral processing as it is required to liberate the

valuable minerals from the gangue. The breakage action is also described as the

creation of new mineral surface. Increasing mineral surface is essential for metallurgical

extraction processes such as leaching and flotation.

The energy requirement in comminution is a function of the reduction ratio, product size,

and the hardness characteristics of the material, i.e. its breakage resistance. The

relationship between required comminution energy, reduction ratio, product size, and

material properties has been the object of research for more than a century. The

theoretical and empirical formulas derived from this previous work and are summarized

by Jankovic et al. (2010).

Comminution in mining operations usually comprises the reduction of large rocks with

sizes around 1 meter or larger to minute particles of 25 microns or smaller. However,

most of the energy is used by the industry (89%) during the reduction from

approximately 20 mm to 100 microns (Powell, 2010).

Currently, the comminution process is energy intensive and highly energy inefficient. It

is estimated that comminution accounts for 65% to 80% of all energy usage in mining

1
operations and that only 1% to 2% of the applied energy is effectively translated in the

production of new surface area (Tromans and Meech, 2002). This expensive and

inefficient process also represents a significant fraction of the world electric power

consumption, e.g. in 1981, comminution processing accounted for approximately 2% of

the total U.S.A. electric power usage (Kawatra and Eisele, 2005).

The combination of the energy intensive and inefficient characteristics of comminution

implies that there is a great opportunity for significant energy and economic savings by

the improvement of this process (Kawatra and Eisele, 2005).

2
1.2 Modern Metal Mining

In the recent few decades, there has been a shift from the mining of high-grade, near-

surface, and relatively soft orebodies to low-grade, deeper and harder ores. The

depletion of high-grade ores and the increasing demand for metals have stimulated the

development of large-scale operations.

These large-scale mining operations extract the valuable minerals from massive

orebodies and are the main source for many base and precious metals. For instance,

the twenty largest copper mines around the globe were responsible for more than 60%

of all copper production from mines in 2008 (International Copper Study Group, 2009).

The advent of large tumbling mills has facilitated the development of these high-tonnage

deposits for the last three to four decades. These high capacity mills, specifically the

autogenous (AG) and semi-autogenous mills (SAG), have progressively replaced

crusher-based circuits due to their simpler flowsheets with fewer pieces of equipment. In

addition, these circuits do not utilize washing plants. Washing plants are usually

required ahead of a crushing circuit when dealing with orebodies that contain a high

level of weathered material (regions with high-clay content) or high moisture content (not

rare characteristics in these large orebodies).

Even though SAG-based comminution circuits are dominant in the industry, they do

present some challenges for the treatment of several types of large orebodies. If the

orebody contains significant hard ore, the SAG mill becomes extremely energy inefficient

as its capacity is highly reduced (Morley and Staples, 2010). High hardness variability

throughout the orebody produces significant SAG capacity variation and thus provides

an adverse overall throughput fluctuation (Burger et al, 2006). Similar fluctuations occur

3
when the SAG feed size distribution cannot be maintained relatively uniform through

time (Morrell and Valery, 2001).

Usually, the larger these low-grade deposits are, the larger the variance in rock

properties such as hardness levels. For instance, large porphyry copper ore deposits

(currently the largest source of copper ore) can present highly variable hardness and

some examples of such orebodies are Freeport-McMoRans Chino Mine in New Mexico

(Amelunxen et al, 2001) and Newmonts Batu Hijau operation in Indonesia (Burger et al,

2006).

4
1.3 HPGR in Hard Rock Mining

In recent years efforts to improve the comminution process have led to the integration of

the High Pressure Grinding Roll (HPGR) into non conventional applications. Until

recently this relatively new type of crusher was used in the cement, diamond and iron

industries. Over the last few years HPGR has expanded its application to base and

precious metals high-tonnage hard rock processing.

With the application of HPGR to new types of ores there has been debate as to their

suitability compared to the more traditional AG/SAG circuits (Morley and Staples, 2010).

One area that the HPGR manufacturers emphasize is the energy efficiency advantages

of the HPGR when compared to tumble milling. The HPGR manufacturers claim

substantial energy savings (up to 40% savings) when the HPGR circuit is compared to

conventional crushing and grinding circuits (KHD, 2002; von Seebach and Knobloch,

1987; Koppern, 2006).

There have also been indications by comminution consultants outside HPGR

manufacturing field that significant energy savings may be achieved on very hard ores

(Morley, 2006; Morrell, 2008) and research are research confirming this trend (Napier et

al, 1996; Shi et al, 2006). The recognition of these possible advantages added to the

recent developments in HPGR roll surface wear resistance trigged the adoption of

HPGR circuits for recent high-tonnage projects dealing with relatively homogenous, hard

to extremely hard rock orebodies and with limited clay content (Vanderbeek, 2006;

Seidel et al, 2006).

These recent applications have in common the application of the HPGR for tertiary

crushing. Their circuits are very similar to the 3-stage crushing circuits that were vastly

5
applied until the 1960s. Now, in high-tonnage mining these circuits are restricted to

extremely hard ores or in processes where throughput stability is of primary importance.

In other words, at the time of this research the HPGR was only replacing tertiary cone

crushers in a very limited niche for base/precious metal mining.

As a matter of fact, the research community seems to be now realizing this limitation.

Very recently, Prof. Powell presented a paper at the Comminution10 conference, and

confirmed that although the great potential of HPGR is starting to be recognized, a better

understanding of the technology and development of different HPGR flowsheets are

required to ensure that this technology is fully exploited (Powell, 2010).

There are still many unknowns with respect to the types of applications in which HPGR

can be used and as of yet there has been little work done to develop a comprehensive

approach to evaluate the overall efficacy of HPGR circuits versus other circuits. This

research focuses on high-tonnage, base/precious metal operations, the most recent

area where HPGR has been applied. As with any new technology, there is limited

knowledge about it and its true benefits. This research aims to improve the

understanding of the potential benefits and applications of HPGR circuits, and address

some current uncertainties, such as:

Whether a complete HPGR comminution circuit is still able to provide substantial

net energy savings when compared to a SAG-based circuit.

Will the HPGR bring the same benefits when applied on orebodies with mixed

hardness and/or orebodies with high clay content, which are characteristics of

various large copper ore deposits in the world?

6
1.4 Thesis Objectives

The current energy inefficient comminution circuits that are applied in base/precious

metal mining present considerable opportunity for significant energy and economic

savings. In addition, the application of HPGR has demonstrated energy benefits in

comparable applications such as in cement and diamond processing.

Recently, similar benefits are being claimed through the replacement of cone crushers

by the HPGR in conventional 3-stage crushing circuits for some specific hard rock metal

mining cases. However, to take fully advantage of their benefits and to broader their

applications improved understandings of their technology as well as the development of

different HPGR flowsheets are required.

This research focuses on high-tonnage, base/precious metal comminution circuits and

the primary objective of this work is to improve the understanding of the potential

benefits and applications of the HPGR in such circuits. In pursuit of the primary

objective the following secondary objectives are targeted:

Expand on current work to develop a structured methodology for the evaluation

of the energy requirements of the complete HPGR circuits by the application of

circuit design best practices and advanced modelling techniques.

Evaluate and demonstrate the applicability of the structured methodology for the

comparison of SAG and HPGR overall circuit energy requirements through case

studies.

Develop an innovative HPGR flowsheet to treat mixed hardness ores and/or

weathered ores with a high proportion of clays and moisture.

7
 Through a case study assess the suitability and the potential benefits of the novel

circuit for the comminution of hard, weathered ores containing clayish material.

Develop a rigorous approach for testing HPGR circuits by the application of a

unique pilot-plant test program as a basis for experimental simulation.

8
1.5 Thesis Outline

This Thesis is divided into five sections:

The first section (Chapter 2) covers the history of comminution circuits and basic

concepts related to grinding mills and the HPGR.

The second section (Chapter 3) presents the developed structured process for the

design and energy requirement evaluation of comminution circuits. In addition, two

trade-off case studies, based on real ore data, are detailed to demonstrate the

applicability of the procedures.

The third section (Chapter 4) introduces the novel HPGR flowsheet and details the

testwork program used for its evaluation. Also in this section, testwork results are

presented and discussed.

The fourth section (Chapter 5) described the design of a comminution circuit utilizing the

novel HPGR flowsheet for an existing copper-gold orebody. Analyses of the expected

outcomes are given as well as a comparison between the proposed circuit and the

conventional circuit that was previously proposed for the development of the same

orebody.

The final section (Chapter 6) covers the research main contributions and future research

opportunities.

9
2 COMMINUTION CIRCUITS1 - LITERATURE REVIEW

2.1 Introduction

Some facts are self-evident: commodities prices fluctuate; high-grade, large deposits

with easy-to-process ore are uncommon; and energy efficiency is a public matter.

However, mineral processors adapt as well as most to the changing economic

environment, especially in the field of comminution. Comminution is the largest energy

consumer in mineral processing, and, if the ore is hard, requires the largest capital and

operating cost. In modern low-grade mining operations, the scale of the use of energy

and other consumables is unprecedented (Charles and Gallagher, 1982; Abouzeid and

Fuerstenau, 2009).

Proper design of the comminution circuit is a critical task, especially for large-scale hard-

rock projects. Today, several options are analyzed when designing such a circuit.

Some are based on long-established technologies, and others are based on more

recently developed technologies, or technologies that have been adapted from other

types of projects. Selecting the most appropriate circuit is of paramount importance, not

only in deciding the equipment, but also how it is configured. The design task can be

quite different in greenfields projects than in expansions, or in modifications of existing

circuits (Barratt and Sherman 2002).

1
A version of this chapter has been published. Rosario P.P., Boyd K. and Grundy M. (2009). Recent
Trends in the Design of Comminution Circuits for High Tonnage Hard Rock Mining. Recent Advances in
Mineral Processing Plant Design, eds. Malhotra D., Taylor P.R., Spiller E., and LeVier M., Society for
Mining, Metallurgy, and Exploration, Inc. (SME), pp. 347-355

10
2.2 Recent History of Comminution

From the 1920s to 1950s, most comminution circuits were designed with several stages

of crushing, followed by rod and ball mills. During the 1960s, the use of rod mills

declined, as larger diameter ball mills, accepting coarser feeds, became available. The

1960s also saw the advent of autogenous, and, later, semiautogenous mills, and by the

early 1970s, large-diameter autogenous grinding mills (AG), and semiautogenous

grinding mills (SAG), often together with ball mills, became the accepted norm. Although

the power consumption was generally higher, the simpler circuits with fewer components

and smaller footprints made the overall economics of SAG mills superior to three-stage

crushing in most cases (Bond 1985). These SAG circuits opened the door to the high-

tonnage, low-grade operations that have characterized the base metal industry for the

past 40 years. The application of these large tumbling mills increased in such a way that

from the early 1980s to the early 2000s most new or expansion mining projects have

selected some circuit configuration that includes either an AG or a SAG mill (Barratt and

Sherman 2002).

More recently, two factors have driven a change in this trend, especially in hard ore

operations. Firstly, the wish to reduce energy consumption intensified, driven not only

by economics, but also by public interest in climate change, greenhouse gas emissions

and carbon footprint. Secondly, high-pressure grinding rolls (HPGR) became more

attractive as their manufacturers developed roll-wear protection systems to better deal

with hard and abrasive ores. As HPGRs are more energy-efficient than conventional

grinding mills, and because large HPGRs can deliver higher unit throughput at higher

reduction ratios than tertiary cone crushers, some projects are now using HPGRs in

combination with secondary cone crushers instead of SAG mills.

11
Stirred milling technology was developed in the 1950s but has only been applied for

mineral processing during the last couple of decades. There are a few different models

of stirred mill machines on the market and they have been mostly used for regrind

applications. The stirred mill presents better energy efficiency than ball mills for fine

grinding and during the last few years there has been an increasing interest in applying

this technology to coarser grinding ranges (Valery and Jankovic 2002).

12
2.3 SAG Mill Background

2.3.1 AG/SAG Mill Machines

A SAG or AG mill, as with any other type of tumbling mill, is a metallic drum of cylindrical

or in most cases cylindro-conical shape which rotates on its horizontal axis. Raw

material and water are fed through an opening at one end of the mill and discharge

through the other end. The interior surface is lined with resistant material such as

rubber, steel or a combination of them to provide wear protection. In addition, lifters, i.e.

raised sections of the liners, are used to lift and direct the fall of the charge during

rotation.

AG and SAG mills are usually characterized by their large diameter dimension and their

aspect ratio (diameter to length relation) which, differently than the ball and rod mills, is a

high ratio in the order of 1.5 to 3 (Napier-Munn et al, 1996). Another difference is related

to the discharge design, AG and SAG mills are usually equipped with grated discharge

ends to hold back large pieces of rock and steel balls (in SAG mills) and to allow the flow

of the slurry containing the fines (usually a portion of the feed and obviously the ground

material).

These mills can be either shell or trunion bearing supported and most of them are

electric motor-gear driven with single or twin-double pinion arrangements. However, as

currently the limit of power transmission through a pinion is around 7,500 kW (Evans et

al, 2001), the large mills with 11 m diameter (36 ft) and higher and requiring 15,000 kW

and more, are equipped with gearless electric drives. Currently the largest mill in

operation has a diameter of 12.2 m (40 ft) and is equipped with a gearless drive with

22,000 kW. Based on mill vendors information, the largest mill that could be currently

engineered would be limited to 13.4 m (44 ft) diameter (Vanderbeek, 2004).

13
Three breakage mechanisms occur inside a SAG or AG mill, they are: abrasion, attrition

and impact (Napier-Munn et al, 1996). Impact breakage is achieved by the cataracting

of the load (steel media and slurry raw material plus water) due to the high speed

rotation; cataracting action meaning the free fall of the load above itself. Abrasion and

attrition are generated by the rolling movement of the load as the material lifts and slips

together. The balance of the energy applied in the comminution of the rocks is

dissipated in the form of heat, noise and the wear of the grinding balls and the mill liners

(Norman and Decker, 1985).

The control of raw material and water feed-rates, and mill speed (for mills equipped with

variable feed drives) is essential for smooth operation and minimum comminution of

media and liners. For example, if the property of the feed rapidly changes and softer

and finer than normal feed is present, the operator (or automated control system) may

need to decrease the feed-rate of the raw material and lower the speed of the mill to

avoid a decrease in the mill load level and thus an increase in the frequency of media-

media and media-liner impacts.

2.3.2 SAG Operational Parameters

Steel ball charges range from 0% (AG mill) up to 20% by volume, and a typical value for

SAG is 12%. The total charge (balls plus slurry) is usually between 20% and 35%, and

the slurry is usually between 65% to 75% solids. The most frequent ball size for large

mills is 127 mm diameter, but it can vary from around 90 mm to a maximum of 152 mm

(Sepulveda 2008).

The recent trend has been to operate at increasingly high ball loads, and at increasingly

low total loadingit has been observed that a lower total charge improves capacity.

Today some operations operate with ball charges up to 20%. Total mill volumetric

14
loading has decreased from around 35% in the early days to as low as 24% or below

(Sepulveda 2008).

High ball charges have only been made possible by the advent of the variable-speed

drive, one of the most significant advances in SAG milling. The variable-speed drive

was first installed on a SAG mill at Afton (1977) (Thomas 1989) and is now almost

universally used. An early example of the advantage of variable-speed drives was at

Lornex (now Highland Valley Copper), where a variable-speed mill installed in 1981 was

operated at up to 19% ball load, compared to 12% for fixed-speed mills in parallel

circuits. The operators could drive the new mill harder, confident that if the ore suddenly

became softer, they could slow the mill down to protect the shell.

2.3.3 SAG Mill Original Circuit

Before SAG milling entered the scene, large grinding plants consisted of many trains of

two or three stages of crushing, rod milling, ball milling, and the associated conveyors,

screens and surge bins (Figure 2-1). The SAG mill gained its leading status in large mill

operations because of its ability, in a single unit, to receive coarse primary crusher

product and deliver adequate ball mill feed at high operational availability (approximately

93%) (Figure 2-2). Development since the early days has centered on increasing the

amount of ball mill feed that a single unit produces.

15
Figure 2-1: Three Stages of Crushing, Rod Mill, Ball Mill

Process Water
Coarse Ore Flotation

Fine
Secondary Tertiary Ore Bin
Crushers Crushers
Rod Mills Ball Mills

Secondary Tertiary
Screens Screens

Since their appearance in the 1970s, SAG mills have increased in size and power, their

drive systems are more advanced, they are equipped with better control systems, and

their benefits and shortcomings are better understood. These developments resulted in

new circuit configurations and programs to improve the quality of feed. Many of the

more significant advances were made by operators determined to extract more from

what they were given.

Large diameter SAG mills have been selected for new hard rock projects and

expansions (Los Bronces Development Project, Phoenix Project, San Cristobal) which

indicates that, depending on the ore type and project specifics, a SAG circuit may still be

the preferred choice.

16
Figure 2-2: SAG-Ball Mill Circuit

Process Water Flotation

Coarse Ore SAG Mill

Discharge Ball Mills
Screen

SAG Mill

2.3.4 Pebble Crushing for AG/SAG Circuits

Competent rocks in the 12 mm to 75 mm range (critical size) present reduced breakage

rates in autogenous (AG) mills. A significant contribution of grinding media in a SAG mill

is to accelerate the breakage of critical size material to reduce its tendency to

accumulate in the mill. Another, nowadays less common, method of preventing the

build-up is the Autogenous Mill-Ball Mill-Crusher (ABC) circuit, where the critical size

material is extracted from the mill, crushed, and returned to the mill. These two

techniques were combined during the 1980s, when there were several successful

attempts by operating mines, to improve their SAG mill performance by using pebble

crushersthe Semiautogenous-Ball Mill-Crusher (SABC) circuit. Examples are Los

Bronces, Similkameen (Major and Wells 2001) and Chino (Vanderbeek 1989). Inclusion

of a pebble circuit has become almost standard in the design of grinding circuits (Figure

2-3). Even if it is not thought appropriate to install pebble crushers at the outset, it is

usually considered prudent to leave space should circumstances require pebble

crushing later in the operation.

17
Figure 2-3: SABC Circuit

Process Water
Flotation

Pebble
Pebble Bin
Crushers

Ball Mills
SAG Mill
Discharge
Coarse Ore Screen

SAG Mill

For hard and very hard ores (JK Axb values below 40 and Bond Work indices above

16 kWh/t), correct forecasting of the production of critical-size material, and of its

extraction rate through the mill grates, is still difficult. There have been reports of

operations that spent great effort to achieve the designed pebble extraction, and

therefore the design throughput, for quite some time after startupfor example Cadia

Hill (Hart et al 2001) and Sossego (Delboni et al 2006).

Until recently, AG mill and SAG mill circuits were invariably designed in closed circuit

with the screen and pebble crusher, with the screen oversize portion being crushed and

completely recycled to the mill feed. Recently, however, some SABC installations have

been operated in open circuit by having the screened crusher product report to the ball

mill circuit (Figure 2-4). The effect of opening the circuit is to pass more tonnage at

coarser size to the ball mill circuit. Consequently in most cases it has been used to

increase throughput of an existing operation which had extra ball mill capacity or could

tolerate a coarser grinding-circuit product size. There are also new installations (most at

the planning stage) where the largest available SAG mill could not reach desired

18
capacity with the pebble crusher in closed circuit. Thus an open-circuit SABC was

chosen. An example is the El Teniente Colon Concentrator with an 11.6m diameter

SAG and four parallel pebble crushers in SAG open circuit configuration (Spann and

Ottergren 2004).

Figure 2-4: Open-Circuit SABC

Process Water Flotation

Pebble Pebble
Crushers Bin

Crushed
Pebble Screen
Ball Mills
SAG Mill
Discharge
Coarse Ore Screen

SAG Mill

A design where the pebble crusher can either be used or bypassed provides the

operator with some external operating control of the SABC circuit. The ability to open or

close the circuit during operation provides additional flexibility. The authors recently

completed a study for a property where the run-of-mine ore had zones of greatly

fractured ore, and zones of very competent ore, both with high ball-mill work indices. It

was proposed to lay the plant out so that the operator could bypass the crushers and

operate the SAG mill in closed circuit when receiving fractured ore (to maximize SAG

mill power) and, when the ore was competent, use the pebble crushers and even open

the circuit, to pass more of the work to the ball mills.

19
Another recent development in pebble crushing is the addition of HPGRs to treat the

pebble crusher product (Figure 2-5). The pebbles are reduced to a much finer product

thereby decreasing ball mill power requirements. Depending on the original circuit,

opening the SABC circuit and adding an HPGR stage may achieve a significant capacity

increase, without increasing the ball mill duty requirement (Dixon et al, 2010). This

concept can also be applied to a circuit that will ramp-up after startup. For example the

Peasquito project has started up with a single SAG line in mid 2009, a second SAG line

was added in mid 2010, and later one HPGR will be added (Goldcorp 2009).

Figure 2-5: SABC with HPGR

Process Water

Flotation

Pebble HPGR
Bin Storage
Bin

Pebble
Crushers
SAG Mill
HPGR SAG Mill
Trommel
Ball Mills
Screen
Crushed
Pebble
HPGR
Screen
Pebble
Washing
Screen
Coarse Ore

Screens

Early designs for the screens closing the AG/SAG mill circuit were either trommel

screens with water jets returning pebbles through the discharge end of the mill, or

vibrating screens with a series of conveyors returning the oversize to the feed end of the

mill. Some (such as Lornex and Copperton), used a combination of trommel screens,

pumps and vibrating screens. Since pebble crushing circuits have become common, the

20
trommel screen/water jet has become less so. Some companies (e.g. Alumbrera and

Antamina) have later added recycle conveyors and pebble crushers to their trommel

screen system, and keep the water off when the pebble crushers are in use. More

recent large operations employ trommel screens to remove most of the slurry, followed

by vibrating screens to wash the pebbles before discharging them onto the recycle

conveyors.

Pebble Surge Capacity

Early SABC circuits incorporated crushers retrofit into SAG mill recycle conveyor

systems, and often had no surge capacity. Surge capacity is highly desirable, enabling

the crushers to be choke fed by controlling the feed rate. Thus pebble bins are now

included in circuits as a matter of course. The scale of many recycle operations is now

at the point where a pebble stockpile is more economic than a pebble bin.

2.3.5 SAG Feed Preparation

In the early years after the advent of the SAG mill, typical ball charges were in the range

of 3% to 7% of mill volume, and the general consensus was that large rocks in the feed

were always necessary to assist in breakage. Under present operating conditions of

high ball load and low total loads, the contribution of large rocks as grinding media is

insignificant. It is now realized that improving the blasting and primary crushing phases

to deliver a consistently fine feed to the mill are cost-effective contributions to the overall

comminution system. Several operations have demonstrated substantial improvement

in SAG production by feed preparation programs (mine-to-mill), improving production by

a factor of up to 15% (Lam et al 2001).

SAG throughput is very susceptible to changes in the hardness of the ore and this

should be assessed at early stages of design. In cases where the orebody presents a

21
high variability of friability, provision for blending may be an option to minimize high

fluctuation in production (Dance 2004). If a well defined plan to maintain reasonably

uniform ore hardness is not possible, the operation should be prepared to sustain

fluctuations in tonnage.

In some cases, pre-crushingscalping off and crushing coarse material in the SAG mill

feedhas been applied to manage the SAG feed size distribution (Figure 2-6).

Pre-crushing can be used where there is limited scope to optimize blasting and primary

crushing, as in block-caving underground mines. In addition, pre-crushing has been

applied to maintain designed production levels at mines where the ore hardness has

increased over time or to expand production (Sylvestre et al 2001).

Figure 2-6: Pre-Crushing in an SABC Circuit

Process Water

Coarse Ore
Coarse Ore
Screens
Pebble Pebble
Crushers Bin

Secondary
Crushers Crushed
Pebble Screen

SAG Mill
Discharge
Screen

SAG Mill To Ball Mills

2.3.6 Steel Wear

In comminution circuits, steel is used in the form of steel balls as media for the tumbling

mills, both for the SAG and the ball mills. Steel is also used in many other components

such as: mill liners, HPGR rolls, crusher liners, chute liners, bin liners, etc. The total

consumption of steel is usually a high operational cost (Charles and Gallagher, 1982). In

22
addition, the consumed steel requires energy for its mining, refinement, manufacturing

and transportation phases and represents a significant indirect (or embedded) energy

consumption even when compared to the amount of direct comminution energy

(Radziszewski, 2002; Pokrajcic and Morrison, 2008; Musa and Morrison, 2009).

Although the precise estimation of the steel ball consumption is not a straightforward

task, it is common during the design phase of the projects to estimate the wear rate

through a combination of ore abrasiveness test work, empirical models and historical

data.

The empirical model most commonly used is based on the work by Bond for small

diameter ball mills with some reduction in the magnitude of its constants, as suggested

by Norman and Decker. (Bond, 1964; Norman and Decker, 1985). This model utilizes

the Bond Abrasion index (Ai) as input to determine the wear in grams relative to the

specific power applied. The original equation formulated by Bond for wet ball mills, is as

follows:

Ball-wear in lb/kW-h = 0.35 (Ai - 0.015)1/3

There are two main deficiencies of this model:

the Ai is determined in a dry test and the differences in chemical characteristics

of the pulp in wet milling are not taken into consideration, and,

steel quality differences are not included in the model and metal quality has

significantly improved since the development of the model in 1963.

Halbe and Smolik (2002) state: Unpublished data indicates that for current high quality

metallurgical steel these calculated values [of ball wear] could be reduced [by] as much

23
as 50%. A good procedure is to conduct Ai tests to determine how the sample

evaluated compares with others [ores]. With Ai information it is possible to review

operating data from other plants with similar conditions and Ais, and make a reasonable

estimate of expected wear. Generally the lab performing the tests will have a data base

of this sort of information. Engineers at the test lab or consulting engineers with

extensive experience in grinding circuit [design] can be very useful here.

Radziszewski and his associates at McGill University are developing a comprehensive

mathematical model of steel media wear as a function of mill operating parameters as

well as a set of test procedures to simulate the effect of both, corrosion and abrasion

wear mechanisms (Radziszewski, 1997; Radziszewski et al, 2005). Unfortunately, as

per his last known publication on this matter in 2005, the model seems to be still in the

development phase.

24
2.4 HPGR Background

2.4.1 HPGR Machine

The origin of HPGR can be linked to coal briquetting equipment developed in the early

1900s (Morley, 2006). However, HPGR as a comminution machine was developed in

the early 1980s and is a product of fundamental and applied research on fracture

phenomena conducted by Professor Klaus Schonert (Bearman, 2006). The HPGR was

first introduced around 1985 to treat relatively soft material in the cement industry.

Comminution in a HPGR is achieved by the high pressure compression of a bed of

material which results in high interparticle stresses, i.e., the crushing principle could be

viewed as having rocks compressed in a piston press. The retention time for the

material in a HPGR is very short. The interparticle breakage mechanism enables a low

level of consumed energy and results in a high proportion of fines in the HPGR product

(Tavares, 2005; Gunter et al, 1996).

The HPGR machine has two counter-rotating rolls mounted in a sturdy frame as shown

in Figure 2-7. One roll rotates on a fixed axis and the other one, the floating or

moveable roll, is allowed to move linearly on rails and is positioned by the action of a

hydro-pneumatic system. The material is fed through a shaft feeder creating a forced

feeding action by gravity. The use of the rotating rolls enables a continuous pressing

process instead of a batch process that would be achieved by a limited throughput

piston press type of machine.

25
Figure 2-7: Schematic of a HPGR (Napier-Munn et al, 1996)

Nitrogen cylinder Feed

Oil cylinders Moveable roll Fixed roll

Product

HPGR should not be confused with conventional crushing rolls. Klymowsky et al (2006)

detailed the distinctive characteristics between them, as summarized next:

the HPGR is equipped with a hydro-pneumatic system to apply and maintain a

high pressure condition within the crushing region

they are operated at much lower speeds than crushing rolls (around 20 rpm,

approximately one third of the crushing rolls speed)

HPGR has a unique feed system to maintain constant choke feed conditions,

and, the surfaces of their rolls are made of highly wear resistant materials.

There are three manufacturers of the HPGR machines, all with headquarters in

Germany, they are:

ThyssenKrupp Polysius

Koeppern (or Kppern in German)

KHD Humboldt Wedag AG

26
There are a few differences in the design of the machines depending on the

manufacturer. Polysius machines usually have a high aspect ratio roll design, i.e., the

ratio between the diameter and the length of the roll. An example of a currently large-

size Polysius machine would be one with 2400 mm roll diameter and 1600 mm roll

length. The other two makers favour a low aspect roll design, and an example of a KHD

standard construction machine size would be one with 1700 mm roll diameter and

1400 mm roll length.

In order to minimize roll surface wear when treating abrasive materials, all

manufacturers are able to provide some kind of protection layer for the rolls. Tires with

tungsten carbide studs are used to create an autogenous layer on the roll surface, i.e.

material builds up on the surface area in between the studs to create an ore layer on the

roll. This technology is offered by KHD and Polysius. Koppern has developed

Hexadur, a hard surface layer consisting of ceramic hard phases embedded in a

hardenable steel matrix (Gardula et al, 2005).

The application of HPGR in comminution circuits has increased over the past two

decades and is well established in processing cement, diamonds and iron ore

(Broeckmann and Gardula, 2005). In the last few years HPGR plants to process

precious and base metals from hard ores have been designed and started up. The main

examples are:

SM Cerro Verde, copper, Peru

Boddington, gold, Australia

Mogalakwena North, platinum, South Africa

PT Freeport Indonesia, copper-gold, Indonesia

27
 Zapadnoe, gold, Irkutsk-Russia

Bendigo, gold, Australia

2.4.2 HPGR Terminology and Operational Parameters

A number of terms and operational parameters are particular to the HPGR and the most

relevant ones are listed as following:

HPGR product cake or flake

specific throughput, m-dot

operating gap, Xg

specific pressing force, FSP

specific energy consumption, ESP

HPGR product cake or flake

The HPGR product generally contains a blend of loose particles and agglomerated

cakes or flakes, in different proportions and sizes, depending upon ore characteristics

and machine operational parameters; such as feed PSD and moisture content, applied

pressure, and gap width. Cake strength or competency is usually low, and commonly

these brittle lumps can be easily broken by hand. (Gruendken et al, 2008). To the best

of the authors knowledge there are no standard procedures to evaluate cake

competency.

Specific throughput

The specific throughput, m-dot, is a factor that is regularly obtained from a laboratory- or

pilot-scale HPGR test and is calculated by dividing the value of the measured throughput

(t/h) by the testing machine roll diameter (m), roll width (m) and the peripheral roll speed

28
(m/s). The m-dot consequently is expressed in ts/hm3 units and indicates what

throughput would be achieved from a machine with 1 m x 1 m rolls operated at 1 m/s for

the tested material. If the testwork is properly conducted to closely simulate expected

industrial-scale conditions, such as: moisture content, operating pressure, and roll

surface properties, the m-dot can be assumed to be constant and directly used for

throughput estimation of different size machines (Bearman, 2006).

 
- 



where:

m-dot = specific throughput (ts/hm3)

M = throughput rate (tph)

D = roll diameter (m)

L = roll width (m)

= peripheral roll speed (m/s)

Operating gap

The operating gap is the minimum distance between the rolls. The HPGR gap,

differently than the close side setting (CSS) in the crushers, is not the determining factor

for size reduction but just an indication of the top size in the product. The inter-particle

comminution mechanism enables high levels of size reduction and fines production even

with apparent large gaps (Gruendken et al, 2008).

The gap is a function of the roll diameter and the friction between the feed material and

the roll surface. Larger diameters and/or higher friction factors provide larger gaps and

thus higher throughputs. The friction is affected by feed material proprieties (such as

29
moisture and particle size distribution) and roll surface properties, e.g. studded rolls with

their substantial autogenous layer provide higher friction than hard-faced smooth rolls

(Klymowsky et al, 2006).

The operating gap, together with the roll speed, can also be used for throughput

calculations by the use of the continuity equation, as follows:

        3.6

where:

M = throughput rate (tph)

Xg = operating gap (mm)

= peripheral roll speed (m/s)

L = roll width (m)

rc = density of the product cake (t/m3)

Specific pressing force

The specific pressing (or grinding) force corresponds to the total hydraulic force exerted

on the rolls divided by the roll surface area, i.e. it is the total force divided by the roll

diameter and width and is expressed in N/mm2 (Klymowsky et al, 2006).


 
1000 
 

where:

Fsp = specific pressing force (N/mm2)

F = applied pressing force (kN)

30
D = roll diameter (m)

L = roll width (m)

The specific pressing force is a useful parameter for machine scale-up and its value is

usually in the range of 1 to 9 N/mm2 (Bearman, 2006).

Specific energy consumption

The specific energy consumption (or specific energy input) corresponds to the machine

power input (kW) divided by throughput rate (tph), and thus is expressed in kWh/t. For

machine scale-up and performance comparisons the net specific energy consumption is

more appropriate whereas the calculation is performed with the total net power input,

i.e., the idle power draw is discounted.

Usually the value of the specific energy consumption ranges from 1 to 3 kWh/t. The

specific energy is directly proportional to the specific pressing force and, similarly to

other comminution machines, harder ores require higher values of specific energy when

compared to softer ores to achieve similar size reductions (Bearman, 2006).

2.4.3 HPGR Original Circuits

Cement

HPGR was first introduced in 1985. Since then it has found usage in the cement

industry. Cement production usually involves three phases. In the first phase raw

materials, such as limestone, are ground. In the second phase, the ground components

are mixed and undergo a chemical reaction in a rotary kiln at high temperatures

producing the cement clinker. The third phase is a final grinding phase to reduce the

clinker nodules to 100% passing 90 microns size. Both the pre-treatment of raw

31
materials and the final clinker grinding phases are performed dry. Ball mills, usually two-

compartment mills divided by a diaphragm and using different steel ball sizes in each

compartment, are generally used for clinker grinding (Jankovic et al, 2004).

The first applications of HPGR in cement were in manufacturing plant retrofits. This was

done by the addition of the machine upstream of the clinker grinding mill. With time,

different circuit configurations have been applied and the HPGR has been able to

provide 10% to 50% energy savings in cement grinding (Patzelt, 1992). Figure 2-8

shows one type of circuit that is applied for the precrushing of clinker.

Figure 2-8: Open Circuit HPGR Closed-Circuit Ball Mill (Aydogan et al, 2006)

Iron Ore

In the iron industry, the HPGR was first introduced in 1995 and its application has been

growing since then. For iron ore the HPGR has been applied either as a standalone

stage or ahead of ball mills to improve the efficiency of the grinding required in pellet

feed production. In addition, the machine has been installed in AG-mill circuits (primary

32
grinding of the ore) to re-crush pebble crusher product (KHD, 2008). The HPGR product

is either returned to the AG mill feed or directed to the downstream processes.

An example of the application of the HPGR for re-crushing and being in closed circuit

with the AG mill is shown in Figure 2-9. This type of circuit enhances the mill capacity by

releasing the AG from pebbles which have a limited breakage rate and may build up the

charge volume inside the mill unless the feed rate is reduced. It is noteworthy that most

of the comminution energy is still applied by the conventional AG mills and that the

crusher and HPGR act mostly as auxiliary equipment.

Figure 2-9: HPGR Applied for Pebble Re-Crush at Empire Iron (Kawatra et al, 2003)

Maybe the initial success of the application of HPGR in iron ore AG pebble re-crush

motivated Krupp-Polysius to patent several circuit configurations with HPGR for pebble

crushing. The patent was issued in 1999 and covers the application of the HPGR as a

33
standalone unit for AG/SAG circuit pebble crushing (Knecht, 1999) but the author is not

aware of any industrial application to date.

Diamonds

In diamond ore processing, the main drive for the utilization of the HPGR is the selective

grinding capability that enables the crushing of kimberlite-diamond ore while preserving

the relative large diamond gems. The circuit design correlates to the objective of

extracting the large gems. Thus, common circuits are built with quite complex

classification systems, such as combination of multi-deck screens and density media

separators (DMS), and scrubbers. Figure 2-10 shows a portion of the circuit (re-crush

stage) applied for Argyle Diamond Mines in Australia. (KHD, 2008)

Figure 2-10: Re-Crush Circuit at Argyle Diamond Mines (KHD, 2008)

34
2.4.4 HPGR Precious/Base Metal Recent Circuits

In the last few years HPGR-based plants processing hard ore in high-tonnage precious

and base metals operations have started production. The two main examples in high-

tonnage operations are SM Cerro Verde (start-up in 2006) and Boddington (start-up in

2009).

It is claimed that of the many possible flowsheets that have been proposed for HPGR,

those using HPGR as tertiary crushers, in closed circuit with fine screens, are expected

to provide maximum energy efficiency in hard-rock applications (Morley 2006a). In

addition, the secondary crushing product is screened before feeding the HPGR to avoid

oversized material damaging the rolls. This configuration is illustrated by the Boddington

and Cerro Verde flowsheets (Figure 2-11 and Figure 2-12).

Figure 2-11: Boddington HPGR (Dunne et al 2007)

Process Water Gravity Separation

Flotation
Secondary
Crushers
(5) Ball Mills
Coarse (4)
Fine Ore
Screens
Stockpile

Primary
Crushers Fine
(2) Screens
Coarse Ore (8)
Stockpile HPGR (4) (Wet)
Flash
Flotation

35
Figure 2-12: Cerro Verde (Vanderbeek 2006)

Process Water
Flotation

Fine Ore Ball Mill

Surge Feed
Screens Bin Surge Bin

Ball Mill
Ball Mills
HPGR Feed
(4)
(4) Screens
( Wet)
Coarse
Ore Secondary
Primary Crushers
Crushing Surge
Bin (4)

In a variation of these flowsheets (with the HPGR in closed circuit with screens), the

HPGR can be equipped with a dividing chute; the product from the centre of the rolls is

directed to the ball mill, and the material produced at the edges of the rolls, which is

coarser, is returned to the HPGR feed. This form of HPGR product recirculation has

already been applied in iron ore projects and has been recently developed for some

base metals projects (Gruendken et al 2008).

The Boddington project has a design capacity of 35Mtpa (approximately 96,000 t/d) and

processes two very hard gold ores with average Bond ball mill work indices (BWi) of

15.1 and 16.6 kWh/t, Bond rod mill work indices (RWi) of 22.8 and 24.2 kWh/t, and JK

Axb values of 27.9 and 25.5. The circuit is comprised of: five 746 kW cone crushers,

four 2.4 m diameter(D) x 1.65 m length(L) 5.5 MW HPGRs, and four 7.9 m D x 11.9 m L

(26 x 39 ft) 15.6 MW ball mills (Dunne et al 2007). The projected roll surface wear life

was estimated at 4,250 hours. A 2006 trade-off study showed that a preliminary SABC

circuit would have 7% lower capital costs than the HPGR circuit, and that the HPGR

circuit provided 12% savings in comminution operational costs. Furthermore, the study

concluded that the lower operational costs of the HPGR circuit offset its higher capital

costs (Seidel et al 2006).

36
Cerro Verde has a design capacity of 108,000 t/d of hard copper-molybdenum ore

(average BWi 15.3 kWh/t). The circuit is comprised of: four 746 kW cone crushers; four

2.4 m D x 1.65 m L 5.0 MW HPGRs; and four 7.3 m D x 10.7 m L (24 x 35 ft) 12 MW ball

mills. The projected roll surface wear life is 6,000 hours. In 2006, just prior to startup, it

was reported that although estimated capital costs were higher for the HPGR circuit than

an equivalent SAG circuit, the estimated total comminution operational costs were 1.33

US$/t and 1.70 US$/t for the HPGR and SAG circuits respectively. The main

contributors for this difference are the costs of power and grinding media. The estimated

total comminution circuit specific energy for the SAG circuit was 20.1 kWh/t and for the

HPGR circuit 15.9 kWh/t. In addition, risk analysis results and internal rate of return

factors were responsible for the decision to build an HPGR circuit instead of SAG circuit

(Vanderbeek 2006).

AG circuits are notoriously sensitive to changes not only in ore hardness, as previously

noted, but also in feed size. SAG mill circuits are more stable, and SABC circuits eeven

more stable. However, SAG-based circuits are still very sensitive to feed variations

(Vanderbeek 2006; Morrell and Valery 2001). Anglo Platinum, at the Mogalakwena

North concentrator, selected crushing technology in large part because it gave stability in

feed rate and product size. HPGR was selected in particular, because of its economic

advantages over tertiary and quaternary crushing (Rule 2006).

2.4.5 Energy Savings

As described in section 2.4.1, breakage in the HPGR is associated with high interparticle

stresses in the machines compression zone and occurs relatively fast. This breakage

mechanism enables a low level of consumed energy and creates a high proportion of

fines in the HPGR product, thus providing a high level of comminution energy efficiency

37
(Tavares, 2005), (Gunter et al, 1996). For tumbling mills, such as the SAG mill, the

comminution energy efficiency is lower than the HPGR. This is due to their breakage

system that behaves as an unconfined system (loose bed), i.e., a great portion of the

applied energy is lost through several non-breakage dynamics that are inherent from the

machine design and interparticle interaction effects (Fuerstenau and Abouzeid, 1998).

The HPGR manufacturers emphasize the energy efficiency advantages of the HPGR

when compared to tumbling mills. In one of KHDs brochures it is claimed that: For

most ores, the specific energy consumption lies at around 0.8 3.0 kWh/t. Especially

when coupled with subsequent downstream processes or high efficiency classifiers,

overall grinding energy reductions as high as 40% have been observed. (KHD, 2002).

Polysius declare that the HPGR allows for lower operating and maintenance costs due

to energy savings of up to 20% and reduction of wear to less than 1% for dry and less

than 0.1% for wet milling. (von Seebach and Knobloch, 1987). Koppern reports that

operating experiences and collected data indicate substantial advantages in energy

savings, material throughput capacities and product quality of HPGR technology versus

traditional crushing and grinding equipment. (Koppern, 2006).

Some researchers and consultants in the mineral industry also discuss the HPGR

energy benefits, but usually less emphatically. Napier et al (1996) commented that

savings between 15% to 50% had been reported but cautioned that many of these

reports were based on small-scale machines. Morley asserts that the HPGR is the

most energy-efficient comminution technology available (Morley, 2006). Morrell also

believes in overall energy benefits for the circuit but highlights that the benefits may be

achieved on very hard ores and in circuits where a large portion of the work shifts from

milling to crushing and HPGR (Morrell, 2008). In 2006, Shi et al observed 8% to 29%

38
savings in total energy in a HPGR study treating platinum ores. The comparison were

made between two lab-scale circuits, one comprised of a jaw crusher, conventional

crushing rolls and ball mill, and the other comprised of a HPGR and ball mill. This study

also indicated that the benefits are more pronounced in harder ores and with coarser ball

mill targeted products.

Although no full-scale operational results have been released yet, estimates of energy

consumption calculated during the design phases of the Boddington and Cerro Verde

projects are available. For the Boddington project, a 2006 trade-off study showed that a

preliminary HPGR circuit would provide approximately 5% savings in comminution

power (Seidel et al, 2006). For Cerro Verde, in 2006 it was estimated that the total

comminution circuit specific energy for the SAG circuit was 20.1 kWh/t and for the HPGR

circuit 15.9 kWh/t, approximately a 21% savings (Vanderbeek, 2006).

Energy savings in comminution at downstream grinding phases, usually ball milling, are

expected through the reduction of the Bond ball mill work index (BWi) of the HPGR

product. This reduction in the hardness of the ore, particle weakening, is due to the

production of microfractures in the high pressure process. In addition to this particle

weakening phenomenon, the HPGR produces a high proportion of fines that further

decrease energy requirements in the subsequent mill (Tavares, 2005; Patzelt et al,

1995).

Tests performed by Polysius on siliceous gold ores resulted in 5 to 20% BWi reduction

(Patzelt et al, 1995). Tests performed in a lab-scale HPGR at Anglo Research on

several different ores resulted in BWi reductions between 3% to 7% (van Drunick and

Smit, 2006). Differences in the evaluation of this reduction factor are common. The

author has observed significant differences between testwork performed on the same

39
ore through different HPGR vendors. This may be related to different test procedures.

Some procedures may combine the results of particle weakening with fine product and

others may report the particle weakening factor alone.

2.4.6 Metallurgical Extraction Advantages

The enhanced extraction in processes, such as flotation and leaching following HPGR, is

credited to the generation of micro-cracked rocks during high pressure process, and by

the theory that these micro-cracks are formed predominantly at grain boundaries; which

consequently increases mineral liberation and reagent penetration rate (von Michaelis,

2005; Morley, 2006). There are a number of studies linking the application of the HPGR

to real benefits in gold leaching, especially at coarse fractions as applied in heap

leaching (Klingmann, 2005; Baum et al, 1997; Dunne et al, 1996; Gardula and Sheriff,

2005). Polysius conducted laboratory tests on copper oxide ores suitable for heap

leaching and reported encouraging results (Baum et al, 1996) and von Michaels (2005)

described the kinetics of copper heap leaching and concluded that the better

permeability of the HPGR product may in fact bring benefits in copper leaching.

There have also been reports of significant improvements in gravity recovery of ores

containing coarse gold (Johansen et al, 2005; Dunne et al, 1996). In addition, filtering

and thickening benefits can be expected if a reduction of slimes production is achieved

with a HPGR circuit (von Michaels, 2005).

In the case of flotation, research that demonstrates the flotation benefits on HPGR

products seems to be more limited. In the late 1990s, a few papers written by Mr W.

Baum from Pittsburg Mineral & Environmental Technology and coauthored by two

professionals from Polysius, claimed flotation benefits. One of these papers reports that

rougher flotation on copper sulphide ore improved between 3% to 5% and final overall

40
recovery was up to 7%, but no details of the experimental approach is given (Baum et al,

1997).

Flotation studies conduct with copper ore in a lab-scale HPGR and ball mill were not

very conclusive (van Drunick and Smit, 2006). On platinum ores, a similar study with a

lab-scale HPGR and a Ball mill found some benefits at coarse size fractions flotation

feeds but not at finer fractions (Shi et al, 2006).

Another study indicates doubts about the potential HPGR benefits in flotation, as in most

cases the HPGR product is subjected to ball milling ahead of the flotation phase. In this

recent research, HPGR products were subjected to the JKMRC mineralogical analysis

(MLA) and the images confirmed the production of micro-cracks. However, it was also

observed that these cracks were destroyed when subjected to ball milling. In addition,

no evidence was found of any significant alteration in the characteristics of the liberation

distribution of the valuable minerals within the size distributions studied. This however

should be confirmed with flotation tests for example. (Daniel, 2008)

2.4.7 HPGR Feed and Product Specifics

Like other crushers, the HPGR operation will present challenges when fed with a high

proportion of very fine material (clayish material), ores with elastic properties, or

significantly soft ores (Morley, 2006b). These substances tend to cushion the crushing

action and make the process inefficient. High moisture in the HPGR feed is also

problematic as it may cause slippage of the material on the roll surface, accelerating

wear. Tramp metal can severely damage the roll surface and means for its removal

from the HPGR feed are necessary (Klymowsky et al., 2006).

41
Both Boddington and Cerro Verde ores produce a moderately friable HPGR cake

product that is de-agglomerated in wet screening by the action of water sprays and

vibration. However, if an ore has the tendency to create competent product cakes it may

be necessary to have a more powerful process to break it down prior to sending it to

downstream processes. Scrubbers are standard in the diamond industry, not only to

wash out the clay prior to the HPGR feed, but also after each HPGR stage. Flowsheets

incorporating low-speed tooth roll sizers (MMD sizers) used as de-agglomerators

downstream of the HPGR have been developed (Valery and Jankovic, 2002).

2.4.8 Limitations and Disadvantages

The main disadvantages of a crushing circuit, with the HPGR as its tertiary phase, are

the increased dust generation that requires dust suppression/collection systems, the

complexity of the circuit especially on material handling systems with a large number of

conveyor belts and stockpiles/bins, and possible higher capital cost (Morley, 2006,

Vanderbeek, 2006).

At the start of this research, the industry consensus was that HPGR are not

recommended to treat high weathered ores, very soft ores or a feed that contains high

level of moisture (Morley, 2006). In addition, capital costs for HPGR circuits are higher

than the costs for a similar SABC circuit, as was observed during the Cerro Verde

design phases (Vanderbeek, 2006).

The simplicity and the high availability of SAG circuits seem to be hard to achieve with

HPGR circuits. The HPGR machine can provide high availability alone, however the

complete circuit, which has a cone crusher and the required screens and conveyors can

only provide a high circuit availability when redundancy is added to the circuit.

42
2.5 Other Developments

2.5.1 Increasing Machine Sizes

Over the past four decades, tumbling mill sizes increased considerably; the largest mills

were 6,750 kW in the early 1970s, and now there are mills operating at 23,000 kW. The

increase appears to have stalled somewhat, not because of a lack of need for larger

units, or that the mills have reached a manufacturing limit, but seemingly because of a

lack of confidence in the industry that motor manufacturers can produce reliable drives in

larger sizes. On the other hand, crusher sizes are increasing; recently crusher

manufacturers launched larger machines than were previously available. For example,

Metso has a new heavy duty cone crusher with 932 kW. Both Polysius and KHD

produce HPGR up to 6.6 MW. The increasing sizes of crushing units mean fewer units

and a simpler, less costly, plant; thus reducing the chief advantage of the SAG mill.

2.5.2 Stirred Mills

The stirred mill presents better energy efficiency than ball mills for fine grinding. One of

the reasons being that it can operate with smaller media and provide a better match

between the particles and the media. It has been demonstrated that media sizing is a

key factor for ball mill comminution efficiency (McIvor 1997). The Metso Vertimill has

been used in tertiary grinding for many years, for example, at Red Dog and Chino

(Vanderbeek 1997) and has now been successfully used in secondary milling (Valery

and Jankovic 2002; Jankovic and Valery 2004). A large scale secondary grinding circuit

is now being built and the manufacturer has recently launched a higher capacity

machine equipped with a 2,240 kW motor, claiming that this mill can handle 6mm top

size in the feed (Metso 2009).

43
The horizontal stirred mill IsaMill, manufactured by Xstrata Technology is also now

available in a larger scale unit equipped with a 3,000 kW motor (Isamill 2009). This type

of mill has been tested in the platinum industry as a primary mill receiving the product of

two HPGRs in series. In addition to the high energy efficiency, this experimental

flowsheet is also aiming at metallurgical efficiencies by having a comminution circuit free

of metal media. This is achieved by the HPGR rolls being built to hold a layer of ore and

by the IsaMill utilizing ceramic grinding media (Ayers et al 2008).

2.5.3 Fully Autogenous Grinding

The same economic and social factors that have driven the inclusion of HPGR into the

realm once almost exclusive to SAG mills, have driven the reversion of some operations

to two-stage fully autogenous milling (Koivistoinen and Levanaho 2006). It may be that

the current high cost of steel will persist or strengthen, in which case the replacement of

ball milling with pebble mills may become a trend. Provided that the ore is reasonably

competent, pebbles may be obtained from the feed to either ABC or HPGR circuits.

Figure 2-13 shows one of the circuits in the now defunct Bethlehem Copper mill. This

use of pebble mills was commonly practiced in the Canadian iron ore and uranium

operations.

44
Figure 2-13: Pebble Extraction and Milling
Primary Crushed Ore

Primary -125mm+75mm Pebbles

Screens
Pebble Flotation
Bins

Secondary Tertiary
Crushers Crushers

Secondary Tertiary
Screens Screens

Pebble
Process Water Mills

45
2.6 Summary of Current State

The high capacities and simple circuits of AG and SAG mills opened the door to the

high-tonnage, low-grade operations that predominate in todays mining industry. From

the early 1980s to the early 2000s, the use of SAG mills prevailed in high-tonnage hard

rock operations throughout the world. The SAG technology has improved over time and

currently their main features and limitations are well understood.

More recently, crushing circuits using high pressure grinding rolls (HPGR) as their

tertiary stage have been also applied to the treatment of hard and very hard ores in

base/precious metal mining applications. The HPGR technology presents promising

benefits especially for energy savings and the elimination of grinding steel media. The

use of steel media represents indirect energy consumption and in some cases is

detrimental for downstream processes. Currently there are other developments in

parallel to the HPGR application to free the comminution circuits from steel media aiming

at completely autogenous grinding.

The HPGR technology has been considered one of the best innovations to tackle the

inefficiencies of the current comminution circuits. However, a better understanding of

this technology along with the development of different circuits is required to fully exploit

their benefits in metal mining. Areas of particular need for additional understanding

include the evaluation of energy requirements of complete HPGR circuits and the

development of different HPGR flowsheets for non conventional ores such as ores with

high clay content.

46
3 GUIDELINES FOR ENERGY REQUIREMENT COMPARISONS
BETWEEN HPGR AND SAG MILL CIRCUITS IN HIGH-TONNAGE
HARD ROCK MINING2

3.1 Introduction

Climate change, energy efficiency, greenhouse gas emissions and carbon footprint have

gained a high priority status in the public interest (Sullivan and Oliva, 2007). Electricity

costs have been rapidly increasing in response to record fuel costs for power generation.

Moreover, several countries, such as the Southern African countries, are now facing a

shortage of electric power. Chile, the world leader in copper production, may face

electricity rationing in the near future (Walsh, 2008). The mining industry has realized

the high importance of energy-efficiency and is currently placing this issue at high priority

level (Bearman, 2006).

Comminution processes account for the largest portion of energy consumption in mining

operations, and it has been demonstrated that significant energy savings can be realized

by their optimization (Valery and Jankovic, 2002). Fundamental and applied research on

fracture phenomena demonstrates that interparticle comminution at high pressure

consumes substantially less energy than conventional grinding processes, and served

as the origin for the development of the HPGR (Schonert, 1988 and 1991).

An important additional benefit in replacing the SAG mill with HPGR circuits in hard rock

mining is the reduction is grinding media. The reduction in steel ball consumption not

only has direct economic impacts, it also provides savings in indirect energy, i.e. the

2
A version of this chapter has been published. Rosario, P.P., and Hall, R.A. (2010). A Structured
Approach to the Evaluation of the Energy Requirements of HPGR and SAG Mill Circuits in Hard Ore
Applications. The Journal of the Southern African Institute of Mining & Metallurgy, vol. 110, pp. 117-123

47
energy that would be used in the fabrication and transportation of the balls (Pokrajcic

and Morrison, 2008).

Although the development of recent HPGR-based plants to process precious and base

metals from hard ores indicates a greater acceptance of their application, there is still

uncertainty as to the extension of its benefits. Among them is the net circuit energy

savings when comparing complete SAG circuits to complete HPGR circuits, since extra

equipment is required when using the HPGR.

With the aim to improve the understanding of the direct and indirect energy requirement

differences between complete comminution circuits using SAG mill and HPGR

technologies, a methodology was developed based on comminution circuit design best

practices. This structured methodology involves modern modelling and simulation tools

among other more traditional design tasks, as summarized in the following:

48
3.2 Modelling and Simulation Background

One of the principal application areas of process simulation is in the design and

optimization of comminution circuits. Process modelling and simulation are well

recognized and important tools in mineral process engineering (Herbst et al, 2002)

Modelling and simulation packages provide a reliable way to evaluate different

flowsheets with different types of equipment in a reasonable time. This facilitates the

determination of an optimum choice of circuit (Urrejola et al, 2008; Morrell et al, 2001).

There are three main simulation packages for the AG/SAG milling that are commonly

used to assist engineers in the development of a new comminution circuit or its

optimization; they are JKSimMet, SGS CEET, and Millpower 2000 (Barratt and Doll,

2008).

The best known and most readily available simulation package is the JKSimMet, this

package includes several comminution models that have been developed at the Julius

Kruttschnitt Mineral Research Centre (JKMRC) in Brisbane, Australia. The AG/SAG mill

model utilizes impact and abrasion breakage parameters derived from specific test

procedures. The model is capable of simulating product size distributions and energy

requirements for AG/SAG mills (Morrell, 1992). In addition, the package contains

several additional models for classification, such as cycloning, and other comminution

processes, such as cone crushing.

The JKSimMet package has been used for design and optimization of several high-

tonnage hard rock operations, such as: Cadia Hill (Hart et al, 2001), Sossego (Delboni et

al, 2006), and Candelaria (Muoz et al, 2008). It is a powerful tool for analysing the

effect of ore characteristics and machine operating parameters on crushing and grinding

49
circuits. It is generally well accepted in the industry after 20 years of successful use.

Although this simulator has been used predominantly for SAG mill circuits it has been

increasingly applied for HPGR circuits. In addition, JKSimMet to the best of the

authors knowledge is the only commercially available package that incorporates HPGR

modelling capabilities.

The HPGR model embedded in the package was developed at the JKMRC at the

University of Queensland, Australia (Morrell et al, 1997). The model has been verified

against data collected from industrial units operating at different diamond mines (Daniel

and Morrell, 2004), and the results indicate that the model is robust enough for the

evaluation of new and optimization of existing comminution circuits. As a recent

example, this model was applied for the flowsheet design by Anglo Platinum, at the

Mogalakwena North concentrator project (Rule et al, 2008).

The ore parameters for the HPGR model in the JKSimMet come from the JK drop-

weight and compressed-bed breakage tests in a piston press3. The remaining model

parameters are derived from laboratory or pilot-scale HPGR tests, and they include:

dimensions and operating conditions of the testing machine, experimental size

distributions of the feed and the product, bulk density of the feed and the product flake,

experimental power draw, and throughput rate. There are indications that the accuracy

of the model, and therefore the scaled-up simulation results, improves if the tests are

conducted in a pilot-scale testing machine (Rule et al, 2008). In addition, all HPGR

manufacturers conduct pilot-scale testing to recommend machine sizes for design

purposes (Klymowsky et al, 2006).

3 JKTech are not currently conducting these piston press tests as this information can be derived from the drop-weight
tests accompanied by lab/pilot scale HPGR test results.

50
3.3 Case Studies

To better describe and demonstrate the applicability of the proposed comparison

methodology, two case studies were used in this research.

The hypothetical cases used for this study are two mines each with the option of SABC

(SAG mill-ball mill-crusher) or 3-stage crushing with HPGR comminution circuits for the

processing of precious or base metals hard ores.

These hypothetical projects are assumed to be located in very distinct regions of the

world. One project would be located in a very remote area subjected to harsh winters,

with plant heating requirements, and relying on electricity from the power grid. The other

project is assumed to be located in a semi-arid region subjected to very mild winters,

with no heating requirements, and connected to the power grid. Both cases are typical

of new projects for precious and base metals mining operations.

The physical and grindability parameters of the ore were based on real ore data from

two different sites from previous work. HPGR modeling parameters came from pilot test

results conducted with the same ores (Appendix A).

51
3.4 Design Criteria Development

The battery limits for this comminution circuit evaluation are the primary crusher product

as the feed to the circuit and the Ball mill product (i.e. ball mill cyclone overflow).

The primary crusher product size distribution can be estimated based on survey data

from similar operations, blasting fragmentation and primary crusher modelling (requires

comprehensive testing), or based on a correlation between the abrasion breakage

parameter ta and the crusher P80 as recommended by JKTech (Bailey et al, 2009).

For the proposed case studies, based on survey data from hard rock operations and

taking a simplified and rather conservative approach a primary crusher product with a

125 mm P80 was assumed for both cases. And the complete particle size distribution

utilized as the feed for all circuits is shown in detail in Appendix A.

For each mine, the values for daily production and final grinding were estimated based

on the authors personal experience and historical information.

As previously stated, test results from real ore were used for the physical and grindability

parameters such as S.G. and Bond Ball Mill Work indices. HPGR testing results which

are key components of the design criteria and major inputs for model construction were

based on the same ores (Appendix A lists the complete dataset).

The HPGR lab/pilot tests indicated that product cakes are moderately competent and

can be de-agglomerated by the application of wet screening.

It was assumed that the two orebodies that are used for this study contain a low

percentage of clays and are formed of non stick rocks, i.e., ores that are amenable to

crushing and high pressure grinding.

52
The daily production, final grinding values, physical and grindability parameters of the

ore, and the HPGR modeling parameters were grouped to generate the core process

design criteria for the two mines as shown in Table 3-1.

Table 3-1: Summary of Design Criteria

Case A Case B
Design Criteria (summary) SABC HPGR SABC HPGR
Feed rate (t/d) 55,000 55,000 40,000 40,000
Solids SG 2.60 2.60 2.73 2.73
Circuit F80 (mm) 125 125 125 125
Bond Abrasion Index (g) 0.32 0.32 0.22 0.22
Crusher (Impact) Work Index (kWh/t) 12.5 12.5 14.4 14.4
JK Parameter A x b 50.4 50.4 28.7 28.7
JK Parameter ta 0.46 0.46 0.38 0.38
Ball Mill Work Index (kWh/t) 14.1 13.2 16.5 16.0
Ball Mill Work Index Reduction N/A 6.3% N/A 3.0%
HPGR Net Spec. Energy Req. (kWh/t) N/A 1.68 N/A 1.85
HPGR Specific Throughput rate (ts/hm3) N/A 261 N/A 237
Final Product P80 (m) 140 140 180 180
Climate Mean Temp. Winter/Sum. (C) -25/+14 -25/+14 +17/+22 +17/+22

Other ore parameters that were used in the design process were the complete crusher

table parameters from the JK drop-weight tests and HPGR test results including the feed

and product size distributions. This information is shown in Appendix A.

53
3.5 Flowsheet Development

Currently, after many years of SAG application, it has been recognized and

demonstrated that the inclusion of pebble crushing in hard ore SAG circuits results in

substantial energy savings (Vanderbeek, 2004). Since for this research the JK A*b

parameters and their positions in JKTechs database indicate both ores are hard, SABC

circuits will be use for the comparisons. For the HPGR plant design there are several

possible flowsheet arrangements. However, it has been demonstrated that having the

HPGR in closed circuit with fine screens for the tertiary crushing stage provides

maximum energy efficiency. For hard ore the secondary crushing product should be

screened before feeding the HPGR to avoid over size material damaging the rolls

(Morley, 2006).

Figure 3-1 shows the simplified flowsheets for the two types of circuits used in this

research. The SABC circuit is very typical and the HPGR circuit has great similarities to

the one applied at Cerro Verde (Vanderbeek, 2006). The only modification is that the

order of the material being fed at the fine ore conveyor is inverted to minimize problems

in the transport of the moderately wet material.

54
Figure 3-1: Simplified SABC and HPGR Flowsheets

HPGR Circuit SABC Circuit

Coarse
Coarse
Ore (C.O.)
Ore (C.O.)
Pebble
C.O. Bin SAG Crusher

O/S Sec. Trommel O/S

Screen Crusher Screen

U/S

O/F to
Flotatio
HPGR O/S Cyclones n
wet
U/F

F.O.
Screen Ball Mill
O/F to
Flotatio
Cyclones n

U/F
Ball Mill

55
3.6 Developed Models

Complete comminution circuit models were developed using JKSimMet simulator.

Figure 3-2 is a screen snapshot of one SABC model developed with JKSimMet (A-

SABC case) and Figure 3-3 shows a HPGR case (A-HPGR case). As shown in these

figures, the simulation produces a comprehensive set of information, among them the

stream values of solids, the power draw per equipment, and size distribution details.

56
Figure 3-2: JKSimMet Screen Snapshot of Case A SABC

57
Figure 3-3: JKSimMet Screen Snapshot of Case A HPGR

58
For this study, the SAG mill simulations utilized breakage constant parameters based on

a comparable size mill that had been surveyed in the past (Appendix A), and the ore

characteristics from grindability testwork as listed in Table 3-1. Usually, several tests are

conducted from samples collected in different parts of the orebody and an analysis of the

hardness variability is conducted. Thus, a number of simulations are conducted to refine

the sizing of the mill and usually one test result is selected for final design and the

nominal material balance is derived from there.

This final design test result usually represents the medium hardness or a certain

percentile of the hardness, depending on the variability throughout the orebody, the

mining schedule or other particular characteristics of the project. For this work, as only

one set of tests per case was utilized, it is assumed that this was the test result that

would be selected for the actual final design (i.e., the 75th percentile of hardness value

from a vast database of grindability tests showing modest variability).

Crusher modelling was developed based on Andersens model (Andersen, 1988) and

the necessary constant parameters were determined using the JKSimMet Model-Fit

capability, i.e., inputting information from test work or plant survey, (such as feed and

product size distributions, lab machine dimensions and operating parameters), to the

model and running a special simulation that back-calculates the constants several times

until the best fit is found (JKSimMet, 2004). For the model-fit of the crushers the

information required consisted of results from drop-weight tests and feed and product

size distributions. Since no crusher pilot test was available for this work, the size

distributions were assumed based on simulations performed with Metso Bruno crushing

and screening simulation software (Kaja, 2002). The ore properties are from the real ore

testing results and are listed in Table 3-1.

59
Ball Mill energy requirements were based on Bonds third theory of comminution and the

adjustment of the Ball Mill feed rate and F80 by the application of the phantom cyclone

method (Napier-Munn et al, 1996).

Screens and hydrocyclones were modelled with the JKSimMet standard efficiency

curve model (Kavetsky, 1979). The required parameters for this model were based on

manufacturers recommendation for the cyclones and screens.

Development of the HPGR models was based on the pilot scale testwork results and the

Model-Fit process. Model fitting was considered adequate for both cases by analysing

the correlation between the experimentally measured product size distribution from the

pilot tests and the simulation predictions based on the model-fitted parameters. This

methodology was described by Daniel and Morrell in 2004.

60
3.7 Equipment Sizing

The selection of the comminution equipment to be used in the circuit involved several

simulation iterations and a number of corresponding ball mill energy requirement

calculations. During this process, different values for SAG mill machine/operational

parameters, such as ball load, total grate open area and speed, and HPGR parameters,

such as roll dimensions and speed, were tried. In addition, for this systematic procedure

it was taken into consideration equipment vendor information, and information gathered

from past projects.

The four final refined JKSimMet models were used for the specification of the main

pieces of equipment for the comminution plants. With the information from these

models, preliminary general arrangement drawings were developed. These layouts,

along with resulting material balances, were used for the material handling system

design. The design of the conveyors, surge bins, feeders and dust

collection/suppression systems, was performed with the assistance of AMECs Material

Handling department to assure a prefeasibility level of accuracy.

61
3.8 Results and Discussions

3.8.1 Pure Comminution Energy

The simulation work produced an estimation of the instantaneous energy draw of the

mills, crushers and HPGRs. These data were corrected based on the circuit availability

to estimate the average power draw per hour as reported in Table 3-2. The availability

factors were assumed based on historical information from operations and current

industrial practice.

Table 3-2: Pure Comminution Energy

Unit Power Total Specific
Inst. Simu. Avai. Power Energy
Description Qt. (kW) (kW) Factor (kW) (kWh/t)
SABC Case A
SAG Mill 11.6m D x 7m L ( 38 x 23 ft ) 1 19,985 18,007 0.92 16,566 7.23
Pebble Crusher MP 800 1 597 307.9 0.92 283 0.12
Ball Mill 6.7m D x 11.1m L ( 22 x 36.5 ft ) 2 10,000 9,193 0.92 16,915 7.38
Totals 33,765 14.73
HPGR Case A
Sec. Crusher MP 1000 2 746 672.9 0.89 1,198 0.52
Tert. HPGR 2.2 D x 1.7 W m 2 5,000 4304 0.89 7,661 3.34
Ball Mill 6.7m D x 11.1m L ( 22 x 36.5 ft ) 2 10,000 8,928 0.92 16,428 7.17
Totals 25,286 11.03
Energy Savings HPGR Circuit Case A 25.1%

SABC Case B
SAG Mill 11.6m D x 5.5m L ( 38 x 18 ft ) 1 15,000 13,766 0.92 12,665 7.60
Pebble Crusher MP 1000 1 746 736.2 0.92 677 0.41
Ball Mill 7.6m D x 11.6m L ( 25 x 38 ft ) 1 14,500 13,323 0.92 12,257 7.35
Totals 25,599 15.36
HPGR Case B
Sec. Crusher MP 800 2 597 558.6 0.89 994 0.60
Tert. HPGR 1.9 D x 1.55 W m 2 3,700 3056 0.89 5,440 3.26
Ball Mill 7.6m D x 11.0m L ( 25 x 36 ft ) 1 13,500 12,442 0.92 11,447 6.87
Totals 17,881 10.73
Energy Savings HPGR Circuit Case B 30.2%

62
As shown in Table 3-2, the HPGR circuit provides savings in comminution energy in the

order of 25% and 30% for cases A and B respectively. From these results the potential

for energy savings provided by the application of HPGRs in these types of ores becomes

apparent. This fact has already been reported by a number of authors (Morrell et al.,

1996; Vanderbeek, 2006) some of them working for HPGR suppliers (Gunter et al.,

1996; Broeckmann and Gardula, 2005; Klymowsky et al., 2006) and was in some extent

expected at the beginning of this study.

3.8.2 Complete Circuit Comminution Energy

Following the same methodology described above, the hourly average power draw

values were calculated for all significant systems and pieces of equipment included in

the design of the four concentration plants, as shown in Table 3-3.

As presented in the table, the HPGR energy savings for the complete circuit decreased

substantially from 25% to 7.7% in case A, and from 30% to 18% in case B. Although the

magnitude of the savings decreased, the amounts of reduction are still significant since it

is related to the highest energy consumption portion of any mining operation.

63
Table 3-3: Complete Circuit Comminution Energy
Case A Case B
Total Spec. Total Spec.
Power Energy Power Energy
Description (kW) (kWh/t) (kW) (kWh/t)
SABC Circuit
Comminution Equip 33,765 14.73 25,599 15.36
Conveyors & feeders 610 0.27 575 0.34
Pumps 1,649 0.72 1,199 0.72
Dust & Heating Systems & Extras 2,346 1.02 28 0.02
Total 38,370 16.74 27,402 16.44
HPGR Circuit
Comminution Equip 25,286 11.03 17,881 10.73
Conveyors & feeders 3,660 1.60 2,388 1.43
Screens 300 0.13 195 0.12
Pumps 1,731 0.76 1,259 0.76
Dust & Heating Systems & Extras 4,454 1.94 634 0.38
Total 35,432 15.46 22,357 13.41
Energy Savings HPGR Circuit 7.7% 18.4%

3.8.3 Steel Usage

As detailed in section 2.3.6, steel is used in different forms in the comminution circuits.

The SAG and the Ball mills grinding media, their liners, crusher liners, HPGR rolls, and

the liners for bins and chutes are consumed over time and are an important component

in operational costs. In section 2.3.6 there is also information about the methods for

testing and estimating steel consumption.

For this study, it was assumed that the HPGR circuits and the SAG mill circuits would

have similar steel consumption for liners (including HPGR roll surfaces and SAG mill

liners, grates and pulp lifter bars) and ball mill media. In other words, it is assumed that

the main difference in steel consumption lies in the SAG mill balls. Therefore, the HPGR

circuit benefit would be the value corresponding to the wear of SAG steel balls of the

corresponding SABC circuit.

64
In order to estimate this potential savings in steel consumption, the SAG mill ball wear

rates were calculated based on Bond abrasion index results, Bonds wet ball mill

equation, and a correction factor of 65% (based on the authors experience) and the

results are shown in Table 3-4.

Table 3-4: SAG Mill Steel Ball Consumption

Case A Case B
SAG Balls SAG Balls SAG Balls SAG Balls
Description (kg/t) (t/h) (kg/t) (t/h)
Consumption of SAG steel balls 0.504 1.16 0.465 0.78

The estimate savings in SAG steel ball consumption for both cases are significant.

Assuming a cost of Can$ 1,000 per ton of SAG steel ball, the operating cost savings in

steel grinding media for the HPGR-circuit would be 0.50 Can$/t and 0.47 Can$/t, for

cases A and B respectively. Moreover these savings represent approximately Can$

28,000 and Can$ 19,000 per day for cases A and B respectively.

3.8.4 Ore Variability

Since the objectives of this research were only the description and demonstration of the

applicability of the proposed comparison methodology, the ore variability of the studied

deposits was not taken into account. However, it should be noted that ore properties will

vary through project life and that this work provides two points in a possible large

spectrum. Therefore the results here presented should not be seen as two precise

estimations for two very specific cases but should be interpreted as an indication of the

potential energy savings for projects dealing with average ore hardness between these

two points.

Additionally, the extension of the ore variability through the orebody needs to be

considered when comparing the SABC to the HPGR circuit. Research, testing and

65
modeling has indicated that the application of SABC brings higher risk of production

variability than HPGR circuits when dealing with highly variable ores (Humphries et al.,

2006; Vanderbeek, 2006).

3.8.5 Heating and Ventilation

For case A, the energy required for heating systems in a project located in an area with

harsh winters was taken into consideration. The estimated energy requirement had a

considerable effect on the total energy consumption. The total energy value for the

SABC circuit increased from 36 MW to 38.4 approximately, and for the HPGR circuit

from 31.8 to 35.4 when the heating requirements were included in the analyses. The

HPGR circuit requires significantly more energy than the SABC circuit because it has a

separate building for the secondary crushers and there are numerous conveyor belt

galleries that have to be maintained at a minimum of +5C. The 7.7% reduction in total

energy requirements represents a considerable reduction and if heating requirements

where not present in this case the savings would raise to 11.7%.

This result indicates that heating energy requirements have an effect in the comparison

between HPGR and SAG circuits if the project is located in a cold climate area.

However, for this specific case, it was assumed that the power would not be generated

at site, which may not be the case for a number of operations at very cold and remote

areas. In such cases, the fact that the energy can be easily recovered from fuel power

generators may give the HPGR case a greater advantage.

3.8.6 Availability and Maintainability

Since only a small number of precious and base metal HPGR plants are being operated

at present, the assumptions for circuit availability should be further explored to increase

the confidence in the findings of this research. However, the major problem of short life

66
of rolls reported at early trials with HPGR treating hard abrasive ores resulted in the

development of superior design for these components. Substantial improvements in this

area are claimed by all three HPGR manufacturers. A good example is the application

of tungsten studded roll tyres to provide an autogenous layer on the roll surface

minimizing its wear rate.

Ore abrasiveness is an important factor that has not been included in this study. This

parameter can not only influence the wear rate of HPGR rolls, but also the consumption

of mill liners and steel balls. Future analyses should be conducted in this area.

3.8.7 Additional HPGR Benefits

Additional benefits in downstream processes are not addressed in this work but it should

be recognized that benefits such as improved flotation performance through preferential

liberation and reduced levels of ore oxidation through the reduction of steel usage as

grinding media have the potential to decrease overall energy consumption. It can be

expected that improved flotation equals to fewer/smaller flotation mechanical cells and

therefore less power. These benefits are being assessed by other researchers

(Humphries et al., 2006) and more information is expected to emerge in the future.

3.8.8 HPGR Circuit Drawbacks

The HPGR circuit selected for the comparisons is basically the former 3-stage crushing

circuit and thus may suffer from the same deficiencies or issues.

As described in the section 2.4.7, the HPGR will present challenges when dealing with

high moisture and/or a high proportion of fine material. For this research it was assumed

hard rock homogenous orebodies with limited clay content similar to what is reported for

Cerro Verde and Boddington (Vanderbeek, 2006; Seidel et al, 2006). These specific

67
conditions may not be always encountered in every application and in many cases the

application of this type of circuit may become detrimental due to frequent operational

problems especially in the crusher and HPGR systems. Such problems will impact the

overall circuit maintainability and availability negating any economic benefits.

The higher number of pieces of equipment and the additional requirements for surge

bins, conveyors, transfer points, dust suppression/collection systems, and feeders

certainly turn the HPGR circuit into a more complex circuit when compared to the SAG

circuit. Complexity increases the risk for design and construction errors that may delay

the achievement of design targets and/or disturb operations. The complexity and

possible deficiencies in the final circuit may increase labour requirements in both

operational and maintenance departments.

Currently, a crushing-HPGR circuit as used in the comparisons generally represents a

higher capital cost investment than the corresponding SABC circuit. Although the two

case studies demonstrated the potential for significant savings in operational costs to

offset the capital cost disadvantage, that may be not the case for other deposits. In

addition, junior or mid-tier mining companies may face challenges in raising capital for

project development which would then favour the SAG mill option.

The HPGR-based circuits studied here present a considerable overall circuit footprint.

The crusher and HPGR embedded close circuits, the bins and feeding requirements for

these machines, along with the application of conventional conveyor belts make the

overall plant layout larger than the SAG-based one. This large footprint and the

corresponding additional building requirements not only effect capital and operational

costs but may be a limiting factor for projects with restricted room for plant allocation

what is the case in some mountainous regions.

68
3.9 Summary

In this chapter the development and application of a methodology for evaluating the total

energy requirements of HPGR and SAG comminution plants were presented. This

systematic methodology is based on current industry best practices and involves

flowsheet development, modelling and simulation, equipment sizing and selection, plant

layout development, and power and steel media consumption calculations.

Based on the proposed methodology, comparisons between the total energy usage for

HPGR and SABC circuits for two metal deposits have been conducted. The analysis

included detailed design of four complete comminution circuits.

The work produced an estimation of the real energy savings that can be achieved in the

treatment of similar precious and base metal hard ores. It has demonstrated that these

savings are within the range of 7.7% and 18.4%. In addition, it has shown that a

significant reduction in steel consumption could be achieved based on the elimination of

the SAG mill steel media. As well, this lower steel usage reduces the energy

requirements and or carbon footprint of external suppliers.

The limitations of this work relate to assumptions regarding clay contents, heating and

ventilation requirements and ore hardness variability. These were discussed along with

potential benefits and drawbacks of the selected HPGR circuit.

The work demonstrated the potential benefits of HPGR in energy and carbon footprint

reduction and may serve as an incentive for similar trade-off studies at early stages of

project design evaluations.

In addition, the demonstration of these potential benefits may also serve as incentive for

the development of alternative HPGR circuits for a broader application of this technology

69
by the mining industry. And in the next chapters the development and evaluation of an

innovative HPGR flowsheet to treat mixed hardness ores and/or weathered ores with a

high proportion of clays and moisture are covered.

70
4 TESTWORK PROGRAM FOR THE EVALUATION OF A NOVEL
HPGR-BASED CIRCUIT TO TREAT MIXED HARDNESS ORES
CONTAINING CLAYS 4

4.1 Introduction

Recently HPGR-based plants processing hard ore in high-tonnage precious and base

metals operations have also entered operation. One of the major drivers for using

HPGR in hard-rock operations is that HPGR are found to have better energy efficiency

and savings in grinding media than the circuits based on autogenous mills (AG) or

semiautogenous mills that have been prevalent over the past four decades (Rosario et

al, 2009). A number of base metal operations treating unusually hard ore have

determined that the operating cost savings of the HPGR outweigh the capital cost

disadvantage. The capital cost is generally higher than in a SAG-based circuit (Seidel et

al, 2006). Examples are SM Cerro Verde (Vanderbeek, 2006), which was started up in

2007, and Boddington (Dunne et al, 2007), commissioned in 2009.

Another significant driver is the relative stability in throughput rate and product size

achieved by crusher-based circuits. AG circuits are notoriously sensitive to changes not

only in ore hardness but also in feed size; SAG mill circuits are more stable, and SABC

circuits more stable still, but SAG-based circuits are still very sensitive to feed variations

(Morrell and Valery, 2001). Anglo Platinum, at the Mogalakwena North concentrator,

selected crushing technology in large part because it gave stability in feed rate and

product size, and HPGR in particular because of its economic advantages over tertiary

and quaternary crushing (Rule, 2006).

4 A version of this chapter has been accepted for publication. Rosario, P.P., and Hall, R.A., Grundy M., Klein
B. (2010). A Preliminary Investigation into the Feasibility of a Novel HPGR-based Circuit for Hard,
Weathered Ores Containing Clayish Material. Minerals Engineering, Special Issue: Comminution 10

71
Finally, HPGR are known to produce more micro-fractures in the crushed ore, improving

the power efficiency of a subsequent ball-milling operation or the metal recovery in a

subsequent heap leach (von Michaelis, 2005).

Of the many possible flowsheets that have been proposed for HPGR, those using HPGR

as tertiary crushers, in closed-circuit with fine screens, are expected to provide

maximum energy efficiency. The fine screens are also applied to ensure an acceptable

top-size for the downstream process such as ball milling. A safety coarse-screen

preceding the HPGR prevents oversized material from damaging the rolls (Morley,

2006a). This flowsheet configuration was selected for both the Cerro Verde (Figure 4-1)

and Boddington projects.

Figure 4-1: Cerro Verde Flowsheet (Vanderbeek 2006)

Process Water
Flotation

Fine Ore Ball Mill

Surge Feed
Screens Bin Surge Bin

Ball Mill
Ball Mills
HPGR Feed (4)
Primary (4) Screens
Crushing ( Wet)
Coarse
Ore Secondary
Surge Crushers
Bin (4)

However, although significant benefits are derived from their present applications, HPGR

still present some shortcomings which currently limit broader application. Like other

crushers, the HPGR is challenged by ore with a high proportion of very fine material

(clayish material), by wet feed, or by feed likely to contain tramp metal. Fine material (or

material with elastic properties) cushions the crushing action, reducing process

efficiency. Wet feed may cause slippage of the material on the roll surface, accelerating

72
wear. And tramp metal can damage the roll surface, creating serious maintenance

issues (Morley, 2006b).

Some base metal ore bodies contain a mixture of very hard rocks, softer material, and

clays, in proportions that can vary greatly throughout the deposit. The author was

involved in such a base metal project through participation in a prefeasibility assessment

conducted by AMEC. The project is a low-grade copper-gold porphyry deposit, requiring

a high-tonnage operation to take advantage of economies of scale. The current plan

calls for an 180,000 t/d processing plant.

Early testwork and economic pre-assessments had already recognized that a significant

portion of this orebody contains high proportions of sericite (clays) and that this portion

of the orebody would be processed in the early years of the project. Based on this fact,

all the work conducted prior and during the prefeasibility assessment was based on a

comminution circuit consisting of SAG mills, ball mills and pebble crushing (SABC).

Conventional three-stage crushing circuits, with or without HPGR for tertiary crushing,

were ruled out because of the high clay content of the orebody. Reinforcing this

decision was the position of SAG-based comminution as the standard for porphyry

deposits (Barratt and Sherman, 2002). As a result, no HPGR testwork had been

conducted.

A comprehensive body of grindability testwork has been conducted on the ore at this

property over the past few years and it indicates that the two mineral zones in this

property contain a mixture of very hard rocks, softer material, and clays in proportions

that will vary throughout the life of the mine. Figure 4-2 represents the hardness

distribution of 371 samples that were tested, by the mining companys contracted

laboratory, for impact breakage, either by JK full drop-weight tests or by SMC tests.

73
Figure 4-2: Hardness Distribution of the Deposit Based on Jk A*b Parameters

Soft Very Soft Very Hard

12% 1% 8%
Hard
16%
Mod. Soft
16%

Mod. Hard
15%

Medium
32%

Figure 4-2 illustrates the high variability of hardness and implies that an SABC circuit will

result in substantial fluctuations in tonnage and product size unless a very effective ore

blending program is in place. However, the considerable amount of material in the

moderately soft to very soft classes along with the presence of high clays suggest that a

conventional crushing or crushing-HPGR circuit will present operational challenges.

This combination of facts prompted the author to develop a research project to

investigate the applicability of an innovative HPGR-based flowsheet.

74
4.2 Novel HPGR Circuit for Ores Containing Clayish Material

The novel HPGR flowsheet was developed to take advantage of the potential operating

cost savings in processing hard rocks, stabilize production, and mitigate issues caused

by the soft and clayish materials. The proposed flowsheet is shown in Figure 4-3 and

evaluating its capabilities is the object of this research.

Figure 4-3: Proposed HPGR Flowsheet for Clayish Ore

Crusher Feed Bin

Trommel
HPGR
Screen
Cone Crusher
Washing
Screen
Autogenous
Diverter
Mill/Scrubber

To Ball Mills
Coarse Ore

In this circuit, the primary-crushed ore is slurried in a low-power autogenous mill, where

weaknesses in the fresh rock are immediately exploited and hard material is scrubbed.

The autogenous mill has a non-conventional design with an overflow type of discharge.

The mill product is screened in two stages. The first stage is a trommel screen, which

removes the bulk of the fine product. The trommel screen oversize is then washed on a

vibrating screen and now free from fines it is conveyed to a cone crushing stage followed

by an HPGR. An automatic bypass arrangement protects the HPGR when the cone

crushers release coarse material to relieve jams. The HPGR product is recycled to the

autogenous mill, where it is scrubbed, slurried and screened with the fresh feed.

75
The potential advantages of this circuit over conventional SAG-based or crusher-based

circuits are:

the production rate is less sensitive to variations in ore hardness than a SAG-

based circuit

the autogenous mill deals with any clays and obviates possible issues with high-

capacity, wet, fine screening

competent HPGR cakes are de-agglomerated in the autogenous mill before

screening

dust generation is less than in conventional crushing circuits

no SAG steel media consumption

no potential for steel ball scats to report to the crusher (as in SAG-based

circuits).

To evaluate this circuit, it was simulated using pilot plant data obtained from tests run at

the University of British Columbia (UBC). Pilot testing was conducted rather than less

costly bench-scale tests as there are indications that the accuracy of simulations

improves if the tests are conducted in a pilot-scale testing machine (Rule et al, 2008).

A sample of ore from the deposit previously discussed was sent to the mineral

processing laboratory at the University of British Columbia, Vancouver (UBC), and a test

program was developed to represent a full-scale circuit. These tests (including HPGR

pilot-scale) provided inputs to a modelling program in which the circuit was compared to

conventional SABC technology. In parallel, a single HPGR test was conducted on the

same ore for a preliminary evaluation of HPGR performance in a conventional tertiary-

crushing application.

76
4.3 Testwork

4.3.1 Sample

An ore sample for the test program was obtained from the mining company. The sample

was a blend of material remaining from an extensive comminution characterization

testwork program conducted on several HQ and PQ drill-cores from the main mineral

zone of the property. These samples had been coarsely (-32 mm) crushed using a jaw

crusher, blended, homogenized and stored by the laboratory for an eventual HPGR

testing program. The total weight of the material available was approximately 9,300 kg.

This composite sample was expected to provide a good representation of the

mineralized zone. The characterization drilling program had integrated all the main

mineralized zones of the complete orebody, including the regions containing high

proportions of sericite (clays). The mining company extracted two sub-samples from the

composite, and had their contracted laboratory perform one SAG Mill Comminution

(SMC) test and one MinnovEX SPI test (these results are shown in Appendix B). A

further sub-sample weighing approximately 3,300 kg was shipped to UBC for the

research described herein. (Appendix C details the sample preparation).

4.3.2 Testwork Design

The testwork was designed to represent the full scale process as much as possible. As

part of this, effort was taken to generate a feed for the HPGR that approximated the feed

that it would receive in the actual processing plant. The test flowsheet is shown in

Appendix C, and is summarized as follows:

1. Conduct a preliminary circuit simulation, using JKSimMet, to obtain a target

screen analysis of the cone crusher product.

77
 2. Tumble and screen the sample to scrub and remove the clays and very soft

material, to simulate the effect of the autogenous mill in the full circuit.

3. Crush a portion of the tumbled sample in the laboratory crusher.

4. Run the synthesised sample through the HPGR, at three different pressing

forces.

5. Repeat the procedure twice; at each repetition, before the tumbling stage, mix

the product from the previous stage with a calculated quantity of the fresh ore

sample, to simulate the recycle in the plant (approximately 85%). (Note: four

tests were planned, but after three tests the results (HPGR product screen

analysis) converged, and the last cycle was not required.)

Following the HPGR tests, a series of tests was done to determine the effect, in terms of

screening efficiency, of passing the HPGR product through the autogenous

mill/scrubber. First, product cake from the closed-cycle tests was scrubbed in the

tumbling mill, then washed on a screen with high-pressure water for one minute, and the

screen efficiency determined. Second, a standard HPGR test on the unscrubbed

sample, full feed, was conducted, following which the product, still unscrubbed, was

washed on a screen with high-pressure water for one minute, and the screen efficiency

determined. Third, a standard HPGR test on the full feed sample was conducted, then

the product was scrubbed in the tumbling mill, followed by washing on a screen with

high-pressure water for one minute, and the screen efficiency determined.

78
4.3.3 Test Equipment

High-Pressure Grinding Rolls

The HPGR at the University of British Columbia was manufactured specifically for pilot

plant work. The test unit, shown in Figure 4-4, is fitted with 0.75 m diameter rollers; test

machines of this size require minimal scale-up factors. Table 4-1 summarizes the

HPGR specifications.

Figure 4-4: UBC Pilot HPGR

Table 4-1: Pilot-Scale HPGR Specifications

Item Scale
Roller diameter 750 mm
Roller width 220 mm
Press frame Hinged design
Wear surface Hexadur WTII
Main motor power 200 kW
Maximum specific pressing force 8.5 N/mm2
Variable speed drive up to 40 rpm (1.55 m/s)

79
The unit was designed to provide test data for sizing and selection of commercial units

and to improve knowledge of the HPGR process and applications. The press is fitted

with Koepperns patented Hexadur wear lining with profiled surface (hexagonal tiles

with different heights) and is capable of crushing at specific pressing forces (FSP) of up to

8.5 N/mm2.

Tumbling Mill

The tumbling mill used to scrub and de-agglomerate the ore is a 60mm diameter x 1m

long smooth-lined vessel. It rotates at 22 rpm, providing mixing and tumbling action in

the longitudinal direction.

Screen

A vibrating screen, Sweco Vibro-Energy Separator model ZS40 with 40" diameter

screen panel with steel-wire mesh was used as the process screen for the pilot-scale

work.

Other equipment

Additional equipment used for the tests were laboratory-scale gyratory and cone

crushers, standard Bond ball mill, and dry and wet screen shaking apparatuses.

80
4.4 Results and Discussion

4.4.1 Sample Properties

The samples were blended and homogenized at the laboratory contracted by the mining

company before shipment; this previous homogenization was checked by preparing

samples from two drums randomly selected from among the twelve received, and

determining their particle size distributions (PSDs) and Bond ball mill work indices. The

Bond ball mill work indices (BWIs) were quite similar, 16.4 and 16.6 kWh/t. The particle

size distributions also proved to be quite similar with 80% passing size (P80) of 21.8 and

21.3 mm. Figure 4-5 shows the complete particle size distribution curves.

Figure 4-5: Particle Size Distribution for the Samples as Received

Particle Size Distributions - Original Samples

100%

90%

80%
Sample A
Cumulative Passing [%]

70%
Sample B
60%

50%

40%

30%

20%

10%

0%
0.01 0.1 1 10 100
Particle Size [m m ]

4.4.2 Tumbling Test

Tumbling is the first step of the AG-HPGR circuit and therefore the first test conducted in

the program. The laboratory tumbling mill provides mild autogenous grinding action and

81
is appropriate for the pilot tests. The AG-Scrubber in a full scale circuit is primarily used

to scrub the clays and de-agglomerate cakes produced by the HPGR. As well, some of

the coarse material coming from the primary crusher (large rocks in the range of 120 to

280 mm) is expected to be broken by the action of a large diameter tumbling mill.

However, this can not be properly simulated by pilot testing. Thus, as in the case of

AG/SAG milling, modelling and simulation are used to forecast commercial scale results.

The tumbling tests were conducted wet, with the addition of water to reach 70% solids

by weight. Batch tests were conducted for 7 minutes with loads of approximately 70 kg.

The graph in Figure 4-6 gives an indication of the grinding action as both the feed and

the product particle size distributions are plotted for one of several tests conducted.

As can be observed on the graph, very little size reduction occurs on material larger than

10 mm but there is a significant increase in the fine portion. The -1 mm fraction changed

from approximately 10% to 14.5% and the very fine minus 45 microns fraction increased

from approximately 2% to 7%.

82
Figure 4-6: Tumbling Test Feed and Product Size Distributions

Pre and Post Tumbling PSD Analysis

100

90
Feed PSD
80
Tumbling Product PSD
Cumulative Passing [%]

70

60

50

40

30

20

10

0
0.01 0.1 1 10 100
Particle Size [m m ]

Even though the full-scale circuit is designed with the application of a non-conventional

AG mill with a deliberately low applied power, the mild breakage action achieved by this

tumbling test is probably lower than what would be seen in the full-scale mill. This

difference is acceptable for two reasons. First, the assessment of the full-scale AG mill

performance will be conducted through modelling and simulation (details are given in

chapter 5) based on material hardness characteristics (assessed by other tests) and a

very distinct feed size distribution. Secondly, the main purpose of the laboratory

tumbling test is to simply scrub off the clays as a first phase in the preparation of an

adequate feed for the HPGR test.

4.4.3 HPGR Feed PSD

In the full-scale circuit the product of the tumbling mill is wet screened and the oversize

portion reports to the cone crushers. At full scale the mesh to be used in the trommel

and wet screens is expected to be in the range of 9 to 15 mm, similar to AG/SAG

83
conventional operations. For this research full scale simulations assumed 12.7 mm

screen apertures. For the pilot tests the screen selected had a 6.35 mm aperture based

on the fact that the sample is finer than what will be seen in the full scale plant and

because the pilot HPGR operational gap would be around half that of the size of a full-

scale unit. Figure 4-7 shows a schematic of the lab-scale circuit used to prepare the

feed for the pilot HPGR (open-circuit).

Figure 4-7: Lab-Scale Circuit to Prepare the Feed to the Pilot HPGR (open-circuit)

Tumbling
Mill Wet Vibratory
Screen

+6.35 mm
Pilot
HPGR
-6.35 mm Laboratory
Crusher

In the full-scale circuit, the oversize material from the screen reports to the cone

crushers and their product in turn feeds the HPGR. For the laboratory testwork, there

was a concern regarding the crushing action provided by the small scale crushers, as

the available laboratory crushers would produce too fine a product which would not

represent what would happen at a full-scale operation. However, if no crushing was

applied, the HPGR feed could also be misleading, as no fines would be present and the

size distribution would be significantly truncated at larger than 6.35 mm material. Figure

4-8 shows the particle size distributions for the screen oversize material with and without

laboratory crushing.

84
Figure 4-8: PSDs for Fresh and Crushed Laboratory Screen O/S Material

Particle Size Distributions

100%

90%

80% 100% crushed material

Non-crushed (100% screen O/S)
Cumulative Passing [%]

70%

60%

50%

40%

30%

20%

10%

0%
0.01 0.1 1 10 100
Particle Size [m m ]

To get a better understanding of the PSD in a full-scale operation, a preliminary

simulation was conducted in JKSimMet using standard and benchmarked parameters

previously gathered for different projects (Rosario and Hall, 2010). Figure 4-9 shows the

PSDs for the screen oversize and the cone crusher product from the simulation.

85
Figure 4-9: PSDs from the Preliminary Simulation

Particle Size Distributions - Preliminary Circuit Simulation

100%

90% JK Cone Prod

80% JK - Screen O/S

Cumulative Passing [%]

70%

60%

50%

40%

30%

20%

10%

0%
0.01 0.1 1 10 100 1000
Particle Size [m m ]

The preliminary simulation predicted that some fines will be generated from the cone

crusher and provided an indication of the PSD curve shape that may be expected.

Having this information and the laboratory testing results it was decided to investigate

what would be the optimum blend between the screen oversize and laboratory crushed

material to obtain a PSD with the approximate shape of the full scale circuit and with

minimum size reduction. With the use of linear programming it was calculated that

approximately 20% of the screen oversize material should be crushed for an optimum

blend (Appendix D provides details on the linear programming calculation). Figure 4-10

shows the final blend PSD as well as the original products and the full-scale simulated

cone crusher product. Figure 4-11 shows a schematic of the refined lab-scale circuit for

the tests.

86
 Figure 4-10: PSDs for the Optimum Blend, Original Products and Simulated Product

Particle Size Distributions

100%
Pre-simulated HPGR feed
90%
100% crushed material
80%
Non-crushed (100% screen O/S)
Cumulative Passing [%]

70% Optimum blend (~1/5 crushed)

60%

50%

40%

30%

20%

10%

0%
0.01 0.1 1 10 100
Particle Size [mm ]

87
Figure 4-11: Lab-Scale Circuits Used for the Tests
FULL FEED CLOSED CIRCUIT (CYCLES 2 & 3)

Open Circuit (Cycle 1)

Tumbling Wet Vibratory
Sample Mill Screen

Pilot ~14% Laboratory Pilot

HPGR Gyratory HPGR
-6mm
(~30%) Crusher
~70%
Split

~56%

4.4.4 HPGR Feed Moisture Content

As describe in section 2.4.2, the moisture content in the HPGR feed is an important

operational parameter. Moisture content has an effect on the friction between the feed

material and the roll surface and thus in the operating gap dimension. Consequently the

moisture content affects the machine throughput as well as the energy consumption.

Furthermore, feed moisture has an effect on the characteristics of the cakes produced.

In conventional crusher-HPGR circuits, such as Cerro Verde and Boddington, the

HPGR feed moisture content is driven directly by the fresh feed moisture and the

moisture content of the recirculation stream (wet fine screen oversize). As the fresh feed

moisture usually varies based on regions or depths being mining as well as local

weather, it is difficult to properly estimate the moisture for the HPGR feed. For the test

of this type of circuit it is usually assumed a feed moisture range and a number of tests

are conducted to evaluate its effects.

88
In the proposed AG-Crusher-HPGR circuit, the fresh feed moisture has no effect as all

material reporting to the HPGR is first slurried in the AG mill and then screened through

a combination of trommel and wet vibrating screens for high efficiency.

Due to the importance of the HPGR feed moisture content in the machine performance,

an investigation was conducted to determine the expected moisture content at full scale

operation and thus the most appropriate moisture value to be used in the HPGR tests.

The approach taken for this research was to estimate the full-scale HPGR feed moisture

based on two main parameters. First, the tested material surface moisture, i.e. the

amount of water that remains retained on the rock surface subsequent to tumbling and

screening. Second, the full-scale vibrating screen oversize PSD based on the

preliminary simulation as described in section 4.4.3.

The procedure utilized for this research involved testing phases and calculations

described as follows:

1. tumbling a sub-sample through the laboratory mill;

2. screening the tumbled material on the laboratory vibrating screen (9.25 mm

aperture panel) with abundant water spray;

3. determining the amount of water retained on the screened oversize material by

weighing it soon after the screening, drying it, and weighing it again;

4. determining the PSD of the screen oversize material;

5. calculating the total surface area of this material by:

a. assuming that the rocks, or particles, have perfect cubical shape,

89
 b. assuming that all rocks remained in each PSD screen are cubes with an

edge-length value equals to the average of the screen apertures they

passed and were retained. For example, the material that passed the 32

mm screen and was retained on the 25 mm screen was assumed to be

formed by 38.5 mm edge-length rock cubes,

c. converting the PSD weights in volumes using the S.G. value of 2.70

(value from the tests conducted by the contracted laboratory),

d. calculating the number of rocks per size fraction through the division of

the PSD volume by the corresponding rock unit-volume (the assumed

edge-length raised to the 3rd power),

e. having the number of rocks and their dimensions, the total surface area of

the material is calculated by the sum of the rock surface areas for the

entire size distribution;

6. dividing the amount of water retained (step 3) by the calculated total surface area

(step 5), the surface moisture for the tested material is determined;

7. calculating the total surface area of the AG-Crusher-HPGR circuit screen

oversize material (same moisture content as the HPGR feed) by repeating step 5

with the full-scale vibrating screen oversize PSD (based on the preliminary

simulation);

8. assuming that the tested surface moisture would be the same at full-scale

operation; and,

90
 9. determine the full-scale HPGR feed moisture content based on the calculated

total surface area for the full-scale material and the tested surface moisture.

Table 4-2 lists the main observed values for the test parameters along with the assumed

and calculated values for the full-scale operation. Based on the 3.5% moisture content

observed on the tests, it was calculated that the moisture content for full-scale operation

would be approximately 1.7%.

Table 4-2: Summary of Parameters and Calculated Results for Moisture Content
Description Test Full-Scale
Screen aperture (mm) 9.35 12.7
Oversize P80 (mm) 22.5 98.2
Oversize P50 (mm) 16.8 45.2
Total surface area (dm2) 164.4** 78.3**
Solids weight (g) 8718 8718*
Water weight (g) 315 150**
Liquid content (%) 3.49 1.69**
Surface moisture (g/dm2) 1.92** 1.92*
Note: *assumed, **calculated

4.4.5 HPGR Tests

In total, seven pilot-HPGR tests were conducted during the period of this research.

Table 4-3 is a quick reference legend for the tests and detailed test results and

operational parameters are listed in Table 4-4. The schematic of the lab-scale circuits

for the tests is shown in Figure 4-11 (page 88). Moreover, the test results and details

regarding feed preparation and operational parameters are given in the sections that

follow.

91
Table 4-3: HPGR Tests Quick Reference Legend
Test Tag Circuit FSP** Feed
#01 Cycle 1 Open 4.0 Tumbled & screened (T-S)
#02 Cycle 1 Open 3.0 Tumbled & screened (T-S)
#03 Cycle 1 Open 3.5 Tumbled & screened (T-S)
#04* Full Feed Open 3.5 Original (as received)
#05 Cycle 2 Closed 3.5 T-S plus Test #03 product
#06 Cycle 3 Closed 3.4 T-S plus Test #05 product
#07 Full Feed Open 3.4 Original (as received)
Note: *Test disregarded due to operational problems, **Specific Pressing Force (N/mm2)

92
Table 4-4: Summary of the Main Parameters and Results for All HPGR Pilot Tests

Test Number: #02 #03 #01 #05 #06 #07

Item Symbol Unit Pressure Tests Cycle 1 Cycle 2 Cycle 3 Full Feed
Speed m/s 0.75 0.75 0.75 0.75 0.75 0.75
n rpm 19.1 19.1 19.10 19.1 19.1 19.1
Specific Pressing Force FSP N/mm2 3.0 3.5 4.0 3.5 3.4 3.4
Average Actual Speed wAV m/s 0.76 0.75 0.75 0.73 0.75 0.74
Actual Roller gap (average) XgAV mm 18.9 18.5 17.6 17.5 17.8 20.6
Actual Hydraulic Pressure (average) PAV Bar 61.3 71.6 81.4 71.1 69.5 69.9
Actual Pressing Force (average) FAV kN 493 576 655 572 559 562
Total Specific Energy Consumption ESP kWh/t 1.89 2.14 2.35 2.23 2.09 2.05
Net Specific Energy Consumption ESP net kWh/t 1.57 1.79 1.99 1.82 1.68 1.68
Specific Throughput Constant m-dot ts/hm3 217.6 215.3 210.8 216.2 209.4 246.9
Average Cake Density rC t/m3 2.30 2.30 2.33 2.29 2.32 2.30
Cake Thickness Average Xc mm 20.5 20.4 20.0 21.8 19.8 24.0
Feed Moisture % 1.6% 1.6% 1.6% 1.5% 1.2% 1.7%
Particle Size Distribution
90% Center & 10% Edge: 80% Passing Size P80 mm 6.5 6.2 5.4 5.4 5.7 5.9
90% Center & 10% Edge: 50% Passing Size P50 mm 2.6 2.6 2.3 2.3 2.4 2.5
Reduction Ratio F80/P80 (Centre & Edge) 3.2 3.4 3.9 3.7 3.2 3.6
Reduction Ratio F50/P50 (Centre & Edge) 5.5 5.5 6.3 5.0 4.4 5.2
Percentage Passing 6.35 mm (Centre & Edge) % 79% 81% 85% 85% 83% 81%

93
Cycle 1 Open-Circuit Pressure Sensitivity Tests

After four drums of material (approximately 1,200 kg) were tumbled, screened, partially

crushed, blended, moisture corrected, and homogenized, they were split into three

samples for HPGR testing. Three tests were performed at different specific grinding

forces of 3.0, 3.5 and 4.0 N/mm; the rotational speed was set at 19.1 rpm (circular

velocity of 0.75 m/s) for all tests. These tests resulted in specific throughput constants

(m-dot) of approx. 218, 215, and 211 ts/hm3, respectively (Figure 4-12). The net specific

energy consumption values were 1.6, 1.8, and 2.0 kWh/t, respectively (Figure 4-13).

The product P80s were 6.5, 6.2, and 5.4 mm for an adjusted product containing 90% of

centre material and 10% edge material5. For a better visualization of the level of size

reduction, the feed and product PSDs are provided in Figure 4-14, and F80/P80 and

F50/P50 ratios are given in Figure 4-15. In addition, a summary of the main parameters

and results for all HPGR pilot tests is shown in Table 4-4. The complete data for all tests

is shown in Appendix E.

These pressure-sensitivity tests provided an indication of how the product can get finer

at higher pressure levels, which may be advantageous in full-scale operations. On the

other hand, the tests also indicated the usual negative trends with increased pressure,

such as the decrease in specific throughput and the increase in energy consumption.

These test results did not indicate any anomaly in the application of HPGR for this type

of material with this partially truncated feed (very limited amount of fines). Reasonable

product fineness and moderate specific throughput were achieved at all pressure levels.

At this preliminary assessment phase, with a single composite sample and without

5 The pilot HPGR is fitted with rolls with a large aspect ratio, i.e. the ratio between the diameter
and the length of the roll is 3.4. An excessive amount of edge material is generated during the
tests and thus the product dividing chute is set to get approx. 70%center-30%edge split. In full-
scale machines the normal split is approx. 90%-10% and it is common procedure to report test-
product PSDs with a calculated final product at this split to plot a single PSD curve

94
proper HPGR roll wear tests conducted, it would be premature to recommend an

optimum specific grinding force for full-scale operations. Based on these initial results,

the subsequent closed circuit and full feed tests were conducted at 3.5 N/mm2.

Figure 4-12: Specific Throughput as a Function of Pressing Force

Assessment of Specific Pressing Force: Specific Throughput

300

280
Specific Throughput Mdot [ts/hm]

260
Full Feed #7
240
Cycle 1 Cycle 2
220
Cycle 3
200

180

160
2.0 2.5 3.0 3.5 4.0 4.5 5.0

Specific Pressing Force [N/m m ]

Cycle 1 Cycle 2 Cycle 3 Full Feed #7

95
Figure 4-13: Influence in Energy Consumption due to Pressing Force

Assessment of Spec. Press. Force: Net Specific Energy Consumption

2.4
Net Specific Energy Consumption [kwh/t]

2.1

#1 - Low Press
1.9
#5 - cycle 2
# 3 Medium Press
#6 - cycle 3 Test #7 - full feed
1.6
# 2 -High Press

1.4

1.1
2.5 3.0 3.5 4.0 4.5
Specific Pressing Force [N/mm]

Figure 4-14: Pressure Sensitivity Tests Feed and Product PSDs

Particle Size Distributions -Test #1 to #3

100%

90% HPGR Feed #1 to#3

HPGR Prod #1 (Fsp=4.0) 90%C-10%E
80%
HPGR Prod #2 (Fsp=3.0) 90%C-10%E
Cumulative Passing [%]

70% HPGR Prod #3 (Fsp=3.5) 90%C-10%E

60%

50%

40%

30%

20%

10%

0%
0.01 0.1 1 10 100
Particle Size [m m ]

96
Figure 4-15: F80/P80 and F50/P50 Reduction Ratios

Assessment of Spec. Press. Force: Reduction Ratio P80 & P50

8

6
Reduction Ratio

Test 07
(Full feed)
4

2
2.0 2.5 3.0 3.5 4.0 4.5 5.0
F80 / P80 F50 / P50 Specific Pressing Force [N/m m ]

Cycles 2 and 3 Closed-Circuit Tests

To investigate the effects of closed-circuit operation on the HPGRs performance, the

test procedure was repeated twice; at each repetition, before the tumbling stage, the

product from the previous stage was mixed with a calculated amount of fresh sample, to

simulate the recirculation expected in the plant (Figure 4-11 on page 88, note the

dashed line). The plan was to repeat the closed-circuit tests for three or more times to

achieve a good convergence, but after two iterations the results showed an acceptable

level of convergence and the third cycle was not required. The mdot showed very little

variation in the first repetition, Cycle 2, and a slight decrease in Cycle 3 (Table 4-4). The

convergence in feed and product size distributions can be observed in Figure 4-16.

From Cycle 1 to 2, the HPGR feed becomes moderately finer, especially below 20 mm,

and the product is just slightly finer. From Cycle 2 to 3, the HPGR feed becomes slightly

finer (small difference in the plus 6 mm portion) and the product is practically the same.

97
 Figure 4-16: Feed and Product PSDs for Closed-Circuit Tests

Particle Size Distributions - Cycles 1 to 3 (close-circuit tests)

100%
Feed test #3 (1st cycle)
90%
HPGR Prod #3 (Fsp=3.5) 90%C-10%E
80% Feed test #5 (2nd cycle)
HPGR Prod #5 90%Cen 10%Edg
Cumulative Passing [%]

70%
Feed test #6 (3rd cycle)
60% HPGR Prod #6 90%Cen 10%Edg
50%

40%

30%

20%

10%

0%
0.01 0.1 1 10 100
Particle Size [m m]

98
Full Feed Open-Circuit Test

To assess the differences on HPGR performance between the proposed circuit (based

on the developed testwork program) and a conventional HPGR circuit, an additional

HPGR test was performed (Test 7 full feed). This test was the type of test that is

commonly used for conventional HPGR circuit assessment, i.e., the entire (full) feed is

processed through the pilot-HPGR without any tumbling or scalping of fines or clays (a

sketch of the test procedure is shown in Figure 4-11).

The first observation during the test execution was the difference in the cakes produced.

Visually, the product cakes from the full-feed test were large, approximately double the

size of the cakes produced in the previous tests (refer to Table 4-5 on page 104).

Subsequently, when handling this product through the rotary splitter, the chute ahead of

the vibrating feeder plugged a number of times, which gives an indication of the kind of

problems that may occur in full scale production. Problems related to material-handling

in crushing plants dealing with clayish or high-fines material are expected and the same

is expected for HPGR application (Morley, 2006a).

The full-feed test (Test 7) product PSD revealed a significant correlation to the product

obtained from Test 3 (tumbled and screened) as shown in Figure 4-17 (along with feed

PSDs). The comparison is done with Test 3 as it is also an open-circuit test performed

at the FSP of 3.5 N/mm but with the prepared feed. Interestingly, although Test 7 feed

presents considerably finer feed than Test 3 (31.4% and 9.4% -6.35 mm, respectively),

the HPGR product size distribution is almost the same.

The author suspects that the observed higher performance, i.e. increased size reduction,

for the Test 3 (tumbled and screened) occurred because its feed had most of the softer

and very fine material scalped off and thus was comprised of almost entirely hard

99
material. Similar observations have been recently reported by researchers that indicate

a higher HPGR performance when treating a homogenous feed of hard material if

compared to a blended feed containing the same hard material plus some other softer

component(s) (Abouzeid and Fuerstenau, 2009; Benzer et al, 2010).

Figure 4-17: Feed and Product PSDs for Full Feed and Tumbled-Screened Open-Circuit
HPGR Tests

PSDs HPGR Open Circ. Tests: Standard & Tumbled-Screened

100%

90% Tumbled-Screened Feed

80% Tumbled-Screened Product

Standard Feed
Cumulative Passing [%]

70%
Standard Product
60%

50%

40%

30%

20%

10%

0%
0.01 0.1 1 10 100
Particle Size [m m ]
Measured 68-32% Center-Edge - Calculated 90-10% C-E

As shown in Table 4-4, the Net Specific Energy Consumption (ESP net) result for Test 7

was 1.68kWh/t which presents an approximate 6% reduction when compared to Test 3

ESP net (1.79 kWh/t). In addition, there is a gain of approximately 15% in the value for

Specific Throughput Constant (m-dot) from 215.3 ts/hm3 in Test 3 to 246.9 ts/hm3 in

Test 7. These results appear to concur with the findings of other researchers (van der

Meer and Gruendken, 2009; Morley, 2006a) which advise against the use of HPGR for

the treatment of truncated feeds.

100
However, the full-feed case has 15% higher specific throughput, but the tumbled-and-

screened case only includes 70% of the ore, because 30% (fine material) bypasses the

circuit. Therefore the total plant throughput of the tumbled-and-screened case is actually

24% higher. The calculation follows:

 
-  MD  L   (1) ( Bearman, 2006)

where:

M = throughput rate (tph)

D = roll diameter (m)

L = roll width (m)

= roll speed (m/s)

Assuming similar machine sizes and at similar speed:

 !"# $ %&'( ) 
 7
 3

 !"# $ %&'( )  1.15 (2)

and,
 !"# )  %&'( ) 0.7 (3)

and combining equations 2 and 3,

 !"# )  %&'( )  1.24

101
Similarly, the 6% specific power reduction refers only to the feed to the HPGR; the

specific power applied to the total fresh feed is actually less, even when taking into

account the minimal extra power required for screening and partial crushing.

The difference between these findings and those of the earlier research is that the feed

to the HPGR is only partially truncated, because a portion of the feed to the HPGR was

crushed using a laboratory crusher (as previously described in the HPGR Feed section).

In a full scale operation, the entire feed to HPGR will be crushed upstream.

This analysis indicates that the removal of a portion of fines from the feed proved to be

beneficial. Therefore, in this case a partially truncated feed can be effectively processed

by the HPGR.

4.4.6 HPGR Product Cakes

The HPGR product generally contains a blend of loose particles and agglomerated

cakes, in different proportions (0% to 80%) and cake sizes, depending upon ore and

machine characteristics. These characteristics include: feed PSD and moisture content;

applied pressure; and gap width. Cake strength or competency is usually low, and

commonly these brittle lumps can be easily broken by hand.

The agglomerated nature of the HPGR product influences the selection and

performance of downstream processes, and even the performance of the HPGR itself

when part of the product is circulated back to the machine. In hard ore applications, the

current perception is that no, or only mild, de-agglomeration action is required for

product fine (usually 6 to 9 mm) wet screening in closed-circuit HPGR (van der Meer

and Gruendken, 2009). This is the case for both the Boddington and the Cerro Verde

projects.

102
During the development of the AG-HPGR circuit, it was anticipated that, even with the

tumbling and screening phases ahead of the HPGR, its product could still contain

competent cakes due to the clayish nature of the ore, which prompted the concept of

having the HPGR product re-circulated through the AG mill.

Efforts were made to assess the properties of the cakes produced during the testwork.

First, it was acknowledged that the cakes produced for all HPGR tests could be broken

with similar pressure. Second, an attempt was made to gauge the compressive strength

with the use of the Point Load Test and Brazilian Test apparatus available at the rock

mechanics laboratory at UBC, but with no success (these types of tests and apparatus

are designed for at least one order of magnitude greater strengths). Finally, it was

decided to develop a test based on a wet-screening application, since this is usual in full

scale processes. In addition, standard cake density and dimensional measurements

were performed on samples from all HPGR test products.

Figure 4-18 shows cake samples produced in Test 1 and Table 4-5 lists the average

dimensions for the cakes produced by the tests. As shown in this table, the full feed test

(Test 7) produced significantly larger cakes in the HPGR centre product than the tests

conducted with prepared feed (tumbled and screened).

Figure 4-18: HPGR Test #1 Product Cake Samples

103
Table 4-5 Average Dimensions of Cakes Produced by the HPGR Tests

Length Ave. Width Ave. Thickness Ave. Volume

3
Tests (mm) (mm) (mm) (mm )
1 to 3 59 44 22 56,290
5&6 58 45 21 55,494
7 75 73 24 132,168

The wet-screening test was developed to assess the behaviour of the HPGR test

product when screening on the laboratory wet-screen and to give an indication of

possible screening efficiencies at full-scale circuits. The test protocol is summarized as

follows:

1. An HPGR centre-product sample, weighing approximately 10 kg, is collected

through the standard procedure of splitting the mass through a rotary splitter.

2. The laboratory screen oversize discharge is blanked off.

3. The sample is gently fed onto the vibrating screen panel and left there for one

minute under a controlled water spray (fixed flow rate and pressure).

4. At the end of one minute, the oversized material is collected, dried, weighed, and

its PSD determined.

5. The amount of undersized material in the oversize is determined and the original

PSD of the center-product is also determined (standard HPGR procedure on a

parallel sample) allowing the test screening efficiency to be calculated.

This procedure was used in a number of tests and some of the resulting oversize PSDs

are shown in Figure 4-19. Through these tests, it was possible to observe a significant

difference in screening efficiency between HPGR products from the test with prepared

104
feed and the one with the full feed, with the former producing a substantially higher

efficiency than the latter (90% vs. 77%). In order to assess the tumbling effect on the

HPGR product, tests were also performed with the HPGR product being tumbled prior to

the developed screening test. Figure 4-19 shows the substantial benefit that is achieved

with this procedure for both the prepared and full feed HPGR tests (94% and 96%

respectively).

Figure 4-19: Screen Oversize PSDs from the Tests for Assessment of HPGR Product
Cake Competency

Particle Size Distributions

100%

90% Prepared f eed HPGR prod - test screen o/s

Full feed HPGR prod - test screen o/s

80%
Prep. feed HPGR prod tumbled - test scr. o/s
Cumulative Passing [%]

70%
Full feed HPGR prod tumbled - test scr. o/s
60%

50%

40%

30%

20%

10%

0%
0.01 0.1 1 10 100
Particle Size [m m ]

A poor screening efficiency of the HPGR product may seriously degrade the

performance of a closed-circuit HPGR system. In such a case, there will be an

unproductive return of fine material which otherwise would report directly to the

downstream process.

The fine material compromises the HPGR performance in two ways. First, the fresh

feed throughput is decreased due to excessive return. Second, the returning fine

105
material carries back additional water to the HPGR increasing the total feed moisture

content which will further degrade the machine performance. High levels of moisture

represent a reduction in the friction between the feed material and the roll surface which

decreases the specific throughput and increases the energy consumption. In addition,

excessive moisture may cause slippage of the material on the rolls surface, accelerating

roll surface wear. (Klymowsky et al., 2006).

The tests confirmed the benefit of a scrubbing phase in the HPGR circuit. The scrubbing

increases screening efficiency and reduces moisture in the recycle stream, even

following a wet processing step.

4.4.7 Bond Ball Mill Work Indices

Energy savings in comminution in downstream grinding phases (usually ball milling), are

expected through the reduction of the Bond ball mill work index (BWi) of the HPGR

product. This reduction in the hardness of the ore is due to the production of

microfractures in the high-pressure process. In addition to this particle-weakening

phenomenon, the HPGR produces a high proportion of fines that further decrease

energy requirements in the subsequent mill (Tavares, 2005; Patzelt et al, 1995).

Tests performed by Patzelt et al (1995) on siliceous gold ores resulted in 5% to 20%

BWi reduction. Tested performed in a lab-scale HPGR by van Drunick and Smit (2006)

on several different ores resulted in BWi reductions between 3% and 7%. Differences in

the evaluation of this reduction factor are not rare; the authors have observed significant

differences between testwork performed on the same ore through different HPGR

vendors and that may be related to different test procedures. Some procedures may

combine the results of particle weakening with fine product and others may report the

particle weakening factor alone.

106
The HPGR effect in reducing the Bond ball mill work index was assessed in this

research and the methodology was based on the standard Bond ball mill test procedure.

On one occasion, following the routine Bond test on the feed and product of one HPGR

test (Test 6 with 19.3 and 17.4 kWh/t, respectively), the test on the product was repeated

with a PSD artificially adjusted to that of the HPGR feed. This was done to investigate if

the high amount of fines in such a product was affecting the work-index result. The new

WIBM result for the product was virtually the same (17.5 kWh/t). It is suspected that

because of the nature of the Bond test, (several cycles and makeup of the feed based

on the production of fines in the product until test convergence), it is only marginally

affected by the fines, and may turn out to be the proper test to investigate the particle

weakening effect alone.

Figure 4-20 summarizes the WIBM results in different points of the circuit. Both the full

and prepared feed HPGR products showed a sizeable reduction in the index (12.7% and

9.3% respectively). A higher WIBM for the screen oversize portion than for the feed was

observed (19.3 and 16.6 kWh/t, respectively) which may indicate that the softer material

does constitute the finer fractions of the sample, as one would expect. Due to practical

limitations, the WIBM of the whole screen undersize product (minus 6 mm portion and

approximately 30% of the feed) was not determined. This material was subjected to fine

wet screening for PSD analysis and only the plus 45 microns portion was used for WIBM

testing. Consequently, this WIBM result of 16.9 kWh/t is higher than what would be

expected for the whole screen underflow portion.

107
Figure 4-20: Bond Ball Mill Index Results in Different Points of the Circuit
FULL FEED CLOSED CIRCUIT (CYCLES 2 & 3)

WIBM 16.6 kWh/t

WIBM 16.6 kWh/t WIBM 19.3 kWh/t

Sample Sample
(2% -45 m) (2% -45 m)

Pilot Tumbling
Wet Vibratory
HPGR Mill
Screen
~14% Pilot
HPGR
-6mm
(~30%)
~70% Split
Laboratory
+ 45 m - 45 m ~56% Gyratory
Crusher
Laboratory
~23% Screen ~7%

WIBM 14.5 kWh/t WIBM 17.5 kWh/t

(12.7% Reduction) (9.3% Reduction)
WIBM 16.9 kWh/t

4.5 Summary

This chapter has demonstrated the complete procedure for pilot testing a proposed

HPGR circuit for the treatment of mixed hardness ors containing clays. The testing

program resulted in important observations indicating the suitability and potential

benefits of this novel flowsheet. In addition, a comprehensive testing results dataset

was produced, which includes the pilot HPGR operational parameters and the feed and

product characteristics. These results enabled the design, simulation and evaluation of

the proposed novel AG-Crusher-HPGR circuit which are detailed in the next chapter.

108
5 FEASIBILITY ASSESSMENT OF THE AG-CRUSHER-HPGR CIRCUIT
TO TREAT CLAYISH AND/OR MIXED HARDNESS ORES

5.1 Introduction

As explained in Chapter 4, the author has been involved in the assessment of a SAG-

based comminution circuit originally proposed for a high-tonnage copper-gold porphyry

deposit. This deposit is heterogeneous in rock hardness and present regions with high

clay content. In addition, a sample from this deposit was made available for this

research and has been used for tests at UBC for the development of an innovative

HPGR-based circuit for such ores.

Based on the test results from Chapter 4 and using the methodology proposed in

Chapter 3, the original SABC circuit and the proposed novel-HPGR circuit are evaluated

and their benefits and disadvantages discussed.

109
5.2 Modelling and Simulation

A JKSimMet model was developed for the proposed AG-Crusher-HPGR circuit. The

main inputs were:

data derived from results obtained through the SMC test (A=62.9, b=0.55, and

ta=0.33 and specific comminution energy)

Bond ball-mill work index (16.6 kWh/t)

appearance function from a previously performed JK full drop-weight test on

material from the same project with similar A and b parameters

specific gravity tests (S.G.=2.70)

HPGR pilot machine physical parameters(diameter, length, rpm of rolls)

HPGR pilot machine operational parameters (e.g. speed, power)

The results from the HPGR Test 7, Closed-Circuit Cycle 3 (m-dot, PSD of feed

and product). Some HPGR model parameters were calibrated using the model

fitting capability embedded in JKSimMet (Daniel and Morrell, 2004).

Additional details regarding JKSimMet model construction are described in Chapter 3.

The primary crusher product PSD was estimated from survey data from different

operations with similar hardness and also on the correlation with the ta parameter

recommended by JKTech (Bailey et al, 2009). Figure 5-1 shows the developed PSD of

the feed with a F80 of approximately 125 mm.

110
Figure 5-1: Feed PSD for Circuit Modelling and Simulations

Feed Size Distribution used for the AG-Crusher-HPGR and SABC Circuits
100

90 SABC & ACH Feed

80

70

60
Cumulative passing (%)

50

40

30

20

10

0
0.01 0.1 1 10 100 1000
Particle Size (mm)

During the prefeasibility assessment, the original project equipment (SABC circuit) was

selected to provide a life-of-mine average throughput in the order of 180,000 t/d.

Consequently, the AG-Crusher-HPGR circuit was designed and its model developed to

provide a throughput of approximately 180,000 t/d for the sample provided by the mining

company.

The SABC circuit was simulated using the previously developed JKSimMet model, the

same feed PSD, and the grindability characteristics of the sample utilized in the research

(based on the SMC test and Bond ball-mill work index). This simulation estimated that

the SABC circuit would deliver 139,000 t/d when fed with this ore. The screen snapshot

of the model given in Figure 5-2 shows the mass balance, the P80 of the streams, the

simulated power for the various pieces of equipment, and some machine operational

parameters.

111
Several simulation iterations were performed for the refinement of the AG-Crusher-

HPGR circuit. During this process, different values for full-scale machine/operational

parameters, such as HPGR roll dimensions and speed, were tried. In addition, this fine-

tuning process took into consideration equipment vendor information, and the provision

for parallel systems to increase the availability of the circuit. The screen snapshot of the

model given in Figure 5-3 shows the mass balance, the P80 of the streams, the

simulated power for the various pieces of equipment, and some machine operational

parameters.

112

Figure 5-2: JKSimMet Screen Snapshot of the SABC Circuit Simulation

113

Figure 5-3: JKSimMet Screen Snapshot of the Final AG-Crusher-HPGR Circuit Simulation

114
5.3 Energy Requirements

5.3.1 Ball Mill Energy

The Ball Mill energy requirement for both circuits were based on Bonds third theory of

comminution, the application of the phantom cyclone method (Chapter 3; Napier-Munn

et al, 1996), and the value of 16.6 kWh/t for the WIBM as per the test results conducted

on the sample.

Although a reduction on the WIBM was observed for the HPGR product, no discount was

applied for the Ball Mill power requirement for the HPGR circuit. This decision may have

added some conservatism to the design but seems appropriated due to the difficulty in

precisely estimate the WIBM for the effective Ball Mill feed. As discussed in section 4.4.7,

although a reduction in the HPGR product WIBM was observed, the HPGR feed is

expected to have a higher WIBM than the fresh feed as per the screen oversize test

results. In addition, the WIBM test result for the undersize portion was deemed

inconclusive.

For the required final product P80 of 200 microns, the calculation for the Ball Mill specific

energy resulted in 7.78 and 8.21 KWh/t for the AG-HPGR-Crusher and the SABC

circuits respectively. This difference is mainly due to the higher amount of fines in the

Ball Mill feed for the HPGR circuit as shown in Figure 5-4.

115
Figure 5-4: Ball Mill Cyclone Feed PSD from AG-Crusher-HPGR and SABC Circuits

Simulated Size Distributions: AG-Crusher-HPGR and SABC Circuits

100

90 AG-Cr-HPGR - Balll Mill Cyclone Feed

SABC - Ball Mill Cyclone Feed

80

70

60
Cumulative passing (%)

50

40

30

20

10

0
0.01 0.1 1 10 100
Particle Size (mm)

There are a number of reasons for the coarser Ball Mill feed in the SABC circuit. This

circuit was design to cope with the hard and extremely hard ores that are part of the

orebody and to minimize a decrease in the overall circuit capacity. Consequently the

SAG mill is designed to operate at significant high ball load and speed which implies a

coarser SAG product. In addition, the SAG mill discharge screen is designed with a

fairly large aperture of 15.9 mm thus further coarsening the material that reports to the

Ball Mill.

On the other hand, the HPGR circuit is designed with a smaller aperture for the AG

discharge screen (12.7 mm) and the AG mill, though at low applied power, may produce

extra fines. This mill acts on both the fresh feed and the HPGR product and provides

moderate abrasion breakage to these materials as can be observed in Figure 5-5. The

graph shows the simulated AG combined feed and product and indicates that moderate

116
breakage is achieved especially in the fine and extra coarse portions of the feed. The

simulated results are in agreement with the usual reduced breakage in the size range

between 12 mm to 75 mm that is a characteristic of AG milling (Napier-Munn et al,

1996).

Figure 5-5: AG Mill Feed (Combined) and Product PSDs

Simulated Size Distributions: AG-Crusher-HPGR and SABC Circuits

100

90 AG Combined Feed

AG Product
80

70

60
Cumulative passing (%)

50

40

30

20

10

0
0.01 0.1 1 10 100 1000
Particle Size (mm)

5.3.2 Pure Comminution Energy

Based on the simulations results and the ball mill energy requirement calculations, the

total energy applied by the comminution equipment was assessed. Table 5-1 shows the

description of the selected machine sizes and energy simulation results for the

comminution equipment for both circuits.

117
Table 5-1: Simulation Results Pure Comminution Energy Requirements

Average Specific
Description Unit Power
Consumption Energy
Qt. Inst. (kW) Sim. (kW) Total (kW) kWh/t
AG-Crusher-HPGR Option (2 lines)
4,000 t/h per line (simulated instantaneous)
176,640 t/d @ 92% Availability
Comminution equipment
AG Mills - 10.4 m D x 6.1m EGL (34 x 20 ft) 2 8,000 7,411 13,636 1.85
Sec. Crushers XL-1100 (8 - 6 oper. 2 stdby) 6 820 807 4,455 0.61
Tert. HPGR - 2.4m D x 1.7m W 4 5,000 3,665 13,487 1.83
Ball Mill - 7.9m D x 12.8m L ( 26 x 42 ft ) 4 18,000 15,563 57,272 7.78
Totals 112,920 88,850 12.07
SABC Option (2 lines)
3,080 t/h per line (simulated instantaneous)
138,970 t/d @ 94% Availability
Comminution equipment
SAG Mills - 12.2 m D x 6.7m EGL (40 x 22 ft) 2 25,000 23,279 43,764 7.56
Pebble Crusher XL-1100 4 820 361.4 1,359 0.24
Ball Mill - 7.9m D x 12.8m L ( 26 x 42 ft ) 4 18,000 12,636 47,511 8.21
Totals 125,279 92,634 16.00
Difference in Specific Energy 24.50%

Even though the ore provided by the mining company for the research was harder than

previously estimated for the life-of-mine average, the two circuits were simulated using

the same ore parameters. Therefore any energy requirement comparison should be

focused on specific energy values (last column of the table), as the installed equipment

may be larger than necessary for the AG-Crusher-HPGR circuit.

As shown in the table, the proposed HPGR circuit provides savings in pure comminution

energy in the order of 24.5%. However, as discussed in Chapter 3, most HPGR circuits

add a level of complexity, as more auxiliary equipment is usually required. Although, the

proposed circuit targets simpler operation than conventional HPGR circuits, the

118
determination of the real difference in the energy savings for the complete circuit

involves a higher level of circuit detail.

5.3.3 Complete Circuit Comminution Energy

For the complete circuit energy requirement comparison, auxiliary equipment was sized

and added to the HPGR flowsheet. This refined flowsheet was arranged in parallel to

the development of a preliminary plant layout. Figure 5-6 is the resulted AG-Crusher-

HPGR circuit flowsheet which includes the main additional equipment. The developed

plant layout is shown in Appendix F. And in Appendix G, the original SABC circuit plant

layout is reproduced.

119
 Figure 5-6: AG-Crusher-HPGR Circuit Simplified Flowsheet

Process Water Magnet

Crusher Bin Feed Conveyor

Crusher Bin Feeder

Crusher Feed Bin HPGR

HPGR Bin Feed
Feed Bin
Conveyor

Belt Feeder

To Ball Mills
Belt Feeder HPGR
Scrubber
Feeders

HPGR
Cone Crushers

Diverters
Magnet

Coarse Ore
Metal Detector

Trommel
Screen

HPGR Disharge Conveyor

Washing
Tramp Material Bypass Conveyor Screen
Magnet
Autogenous
Mill/Scrubber
Screen
Tramp Material
Discharge
Pile
Conveyor

To Ball Mills

120
Once the circuit was detailed, the estimation of the energy usage of the complete circuit

was also performed and the complete set of results is shown in Appendix H. Table 5-2

summarized the findings and shows that the complete circuit energy savings are in the

order of 22.7%.

Table 5-2: Energy Requirements for the Complete Circuits

Connected Sim. Average Specific
Description
Power Consumption Energy
kW kW kWh/t
AG-Crusher-HPGR Option (2 lines)
4,000 t/h per line (simulated instantaneous)
176,640 t/d @ 92% Availability
Comminution equipment 112,920 88,850 12.07
Conveyors and feeders 5,108 3,468 0.47
Pumps 12,080 8,562 1.16
Screens 119 82 0.01
Comm. Eq. Lube and Cooling systems 1,537 1,092 0.15
Dust and metal collection requirements 651 466 0.06
Heat & Ventilation Systems 325 299 0.04
Totals 132,740 102,819 13.97
SABC Option (2 lines)
3,080 t/h per line (simulated instantaneous)
138,970 t/d @ 94% Availability
Comminution equipment 125,279 92,634 16.00
Conveyors and feeders 2,554 1,809 0.31
Pumps 11,968 8,664 1.50
Screens 60 42 0.01
Comm. Eq. Lube and Cooling systems 1,476 1,089 0.19
Dust and metal collection requirements 204 148 0.03
Heat & Ventilation Systems 274 258 0.04
Totals 141,815 104,645 18.07
Difference in Specific Energy 22.71%

121
5.4 Operating and Capital Costs

5.4.1 Operating Cost

For the comparison of operating costs, it was assumed that the HPGR circuits and the

SAG mill circuits would have similar steel consumption for liners (including HPGR roll

surfaces and SAG mill liners, grates and pulp lifter bars) and ball mill media. In other

words, it is assumed that the main difference in steel consumption lies in the SAG mill

balls. For the SAG steel media consumption the same methodology as described in

Chapter 3 was used and resulted in 480 grams of steel per processed ton.

Assuming a cost of Can$ 1,000 per ton of SAG steel ball and assuming 180,000 t/d, the

operating cost savings in steel grinding media for the HPGR-circuit would be Can$

86,400 per day.

Assuming the unit power cost of Can$ 0.08 kWh, and having the specific energy

difference of 4.1 kWh/t, it is estimated that the energy savings with the HPGR circuit are

Can$ 59,000 per day. Thus the total estimated savings including steel grinding media is

Can$ 145,400 per day.

5.4.2 Capital Cost

For the purposes of this research, a rigorous materials take-off was not undertaken. To

arrive at an order-of-magnitude capital cost for the AG-Crusher-HPGR circuit, the total

area cost was factored from the cost of the installed equipment, using the ratio of the

area cost to installed equipment cost for the SABC circuit. The SABC circuit costs were

generated during the original pre-feasibility study and are summarized in Table 5-3. This

table also shows the summary of the calculated costs for AG-Crusher-HPGR circuit.

122
Table 5-3: Capital Cost Summary
Item Cost (USD) Ratio
SABC Circuit
SAG Area Total Cost from Pre-feasibility Study 225,943,408
Pebble Crushing Area Total Cost from Pre-feasibility Study 54,484,696
Total SABC Area Cost 280,428,104
Installed Major Process Equipment (SABC) 123,835,917
Factor, SABC Area Cost: Process Equipment Cost 2.26
AG-Crusher-HPGR Circuit
Installed Major Process Equipment 152,640,899
Factor (Area Cost: Process Equipment Cost) 2.26
Total AG-Crusher-HPGR Area Direct Cost 345,657,373

Difference (AG-Crusher-HPGR vs. SABC) 65,229,269

Indirects Cost @ 43% 28,048,586
Contingency at 20% of Direct plus Indirect 18,655,571
Total Cost Difference AG-Crusher- HPGR > SABC 111,933,426 23%

Based on the difference in the operating costs, the payout period for the additional

capital cost for the HPGR-circuit is estimated in approximately 2.1 years.

The simulation results and the cost analyses demonstrate that the novel circuit presents

the potential for significant energy savings and reduced grinding media consumption.

123
5.5 Discussions

Copper porphyry deposits are a major source of global copper production and, in several

cases, present quite heterogeneous orebodies (in terms of rock hardness) which present

challenges to SAG mill operation. Based on the promising results from this application

of the novel circuit for heterogeneous, hard ore with high clay content, it is possible that

this circuit, or one derived from it, may become an important alternative for SAG-based

circuits for processing of copper porphyry deposits.

An example of a property where this circuit might have had an application is the Batu

Hijau operation in Indonesia, with an orebody with high variability of hardness (JK Axb

results ranging from 23 to 107) and the consequent highly variable throughput rates for

SABC circuitsdaily variation ranging from 4,500 to 7,300 tph total for two parallel

circuits(Burger et al, 2006). This may disturb the flotation stage and compromise

metal recovery.

This work has demonstrated that the change in energy savings using the novel circuit

when additional equipment is added is much lower (from 24.5% to 22.7%) than those

calculated in Chapter 3 (from 25.1% to 7.7% in case A, and from 30.2% to 18.4% in

case B). It is believed this is due to some unique characteristics of the proposed circuit,

such as the circuit simplicity and the scalping of the softer components of the ore prior to

the crushers-HPGR portion of the circuit.

The circuit emphasises simplicity; compared to current HPGR-based circuit designs, the

amount and complexity of ancillary equipment is reduced by the elimination of the

nested closed circuits. In addition, the use of high-angle conveyors (sandwich

conveyor) reduces overall circuit footprint and thus building requirements.

124
As described in Chapter 2, crushers and HPGR are considerably more energy-efficient

than AG or SAG mills. For this circuit, the gain in efficiency is augmented by the fact

that the crusher-HPGR portion of the circuit deals with the harder component of the feed

only. Recent research in the application of HPGR for mixtures indicates that machine

performance might be enhanced if treating the hard component of the mixture separately

(Abouzeid and Fuerstenau, 2009; Benzer et al, 2010). Moreover, this circuit fully

exploits the AG mills strong point, i.e. its ability to scalp the softer components of the

fresh feed and the clays.

In the proposed AG-Crusher-HPGR circuit design, a single AG mill per line is used. This

mill has some non-conventional features. The AG is designed with unusual limited

power to perform the scrubbing action of conventional scrubbers and to provide limited

breakage on the soft portions of the ore. This will be achieved with a higher applied

power than conventional scrubbers provided by the higher diameter-length aspect ratio

and the proportionally high internal load inherent to the AG. However, one potential risk

of this design is the required high volumetric flow of slurry (flux) through the mill. To deal

with this, an overflow-discharge type of mill is proposed. Another option would be a

discharge mechanism with modern pulp lift design such as the kind proposed by

Latchireddi in 2009, in combination with larger than usual grate openings (slots)

positioned at the far circular periphery region of the mill.

As detailed in the introduction, the sample utilized in this research was a carefully

prepared composite for proper correlation with the average properties of the orebody.

However, as a single sample was used for the evaluations, the results are quite specific

for the corresponding grindability parameters of this sample.

125
6 CONCLUSIONS

6.1 Main Research Contributions

This thesis has contributed to the understanding of HPGR technology, especially for its

application in comminution circuits for high-tonnage, base/precious metal mining. This

research led to some specific contributions, as follows:

The development of a structured methodology for the evaluation of energy

requirements of complete HPGR and SAG mill circuits. This work involved using

advanced modelling tools and the demonstration of the methodology through

case studies.

The development of an innovative HPGR flowsheet with an AG mill/scrubber and

parallel trains of crusher-HPGR to treat mixed-hardness ores and/or ores with a

high proportion of clays and moisture.

The development of a systematic testing program for the design and evaluation

of HPGR circuits with the application of a unique pilot-plant test program as a

basis for experimental simulation. The method includes the pre-assessment of

HPGR feed properties in full-scale operations as well as corresponding

procedures to prepare the sample, and may serve as incentive to the

improvement of HPGR circuit design protocol.

The demonstration of the suitability and the potential benefits of the novel circuit

for the comminution of hard, weathered ores containing clayish material. This

126
 was achieved through the use of the testing protocol, and of the methodology for

energy evaluation, in the assessment of a full-scale mining project.

This research has unveiled the potential for innovative HPGR-based circuits, and may

serve as incentive for the future use of HPGR in other applications currently labelled as

ideal for different technologies.

127
6.2 Future Research Opportunities

The AG-Crusher-HPGR pilot-plant test program results, as detailed in section 4.4.5,

demonstrated the advantage of scalping the fine particles ahead of the crusher and the

HPGR. It was demonstrated that required HPGR capacity, size and power decreases

and the HPGR product is similar to the one achieved with full feed. It is speculated that

the fine/soft material cushions the breakage action (provide by the high pressure) on the

hard portion of the feed. In cement applications, recent research involving plant surveys

and laboratory piston-die tests provided results in agreement with the observations from

this research (Benzer et al, 2010). In addition, research based on limestone and quartz

mixtures, feeding a laboratory-scale HPGR, suggests that the breakage level of the hard

component is related to the mixture composition and that the presence of hard

components in a mixture enhances HPGR energy efficiency (Abouzeid and Fuerstenau,

2009). This research did not investigate this phenomenon further but follow-up research

in this area will add significant value in the understanding of the HPGR for base/precious

metal application.

More should also be done to create a suitable protocol for analysis of the weakening

effect provided by the HPGR. This research demonstrated a different level of reduction

when treating the scalped feed and the full feed, and indicated the difference in the Bond

Work indices for different size fractions in the sample received from the mining company.

For the energy calculations and comparisons to the SAG-circuit presented in Chapter 5,

a conservative approach was taken by not using any hardness reduction factor for the

Ball Mill power requirements. This seems to be the current consensus for design and

trade-off studies of comminution circuits involving HPGR (Vanderbeek, 2006; Morley and

Staples, 2009). However, the benefit seems to be quite significant, and in some cases

may play a critical role in trade-off studies and/or circuit evaluations, thus the sooner a

128
better understanding of this phenomenon is achieved, and proper testing protocols

accepted by the industry; the better. Tavares (2005) and Daniel (2008), developed

invaluable work in this area but more is needed.

The author recognizes that the low-power AG mill proposed for the AG-Crusher-HPGR

may present some challenges regarding the required high volumetric flow of slurry (flux)

through the mill. To mitigate the problem two alternative but unconventional designs are

proposed. Both the overflow-discharge option and the application of an enhanced grate

discharge system could be the object of future investigation. Probably by the application

of DEM modelling, such as the work being conducted by Cleary and his associates at

CSIRO (Cleary et al, 2008), the effect of different discharge arrangements and the

simulated level of breakage provided by this AG mill could be verified.

Higher precision in equipment and plant availability factors through a more elaborated

assessment would be advantageous for this research. This could be achieved if

dynamic simulation was conducted, such as that done during the feasibility studies for

the Cerro Verde project (Vanderbeek, 2006). Unfortunately, the lack of reliable

maintenance data for HPGR operation (especially in base/precious metal mining)

prevents this from being done at present. The disclosure of current maintenance data

from HPGR operations, especially from Cerro Verde and Boddington, would be of great

value for the industry. If such data was available, the accuracy of dynamic simulations

would be increased and their outcomes would facilitate design and engineering of

conventional and/or novel HPGR plants.

For the two case studies presented in Chapter 3, single samples were used for the

evaluations, and therefore the results are quite specific for the corresponding grindability

parameters of those samples. It was advised that the findings from those case studies

129
be interpreted as an indication of potential HPGR-circuit benefits for projects dealing with

average ore hardness between those two points.

For the evaluation of the innovative HPGR circuit, the sample utilized in the research

was a carefully prepared composite which is expected to correlate well with average

properties of the original projects orebody. However, the grindability parameters

obtained during the test program indicated that the composite may be harder than the

average, and this should be investigated further. This fact does not compromise the

findings of this research but it is recommended that a number of samples, collected from

different parts of the orebody, be tested, and that the program is repeated. Of special

value would be the comparison between the two circuits when treating ores softer than

the average, and the prediction of throughput values based on the mine plan.

The assessment of the ore variability and its effect in the overall AG-Crusher-HPGR

circuit performance may also determine if this preliminary design is appropriate. For

instance, the crusher-HPGR portion of the circuit may receive substantially less material

when dealing with softer than tested ores and changes to the circuit may be required.

For the preliminary design, it was anticipated that the feed rate to this sub-system will

fluctuate and that there will be occasions requiring the shutdown of one or more units. In

addition, the design incorporates some features for improved operational flexibility such

as: crusher and HPGR feed bins, variable frequency drives for the feeders and

machines. Nevertheless, future analysis may still indicate that the application of more

units at lower capacities will be preferable, e.g. four 600 kW operating crushers instead

of three 820 kW units along with three smaller HPGR units.

130
REFERENCES

Abouzeid A.-Z.M. and Fuerstenau D.W. (2009). Grinding of Mixtures in High-Pressure

Grinding Rolls. Int. J. Miner. Process, vol. 93, pp. 59-65

Amelunxen P., Bennett C., Garretson P., and Mertig H. 2001. Use of Geostatistics to
Generate an Orebody Hardness Dataset and to Quantify the Relationship between
Sample Spacing and the Precision of the Throughput Predictions. SAG2001-SAG Mill
Circuit. Int. Conf. on Autogenous and Semiautogenous Grinding Technology 4:207-220

Andersen J.A. and Napier-Munn T.J. (1988). Power Prediction for Cone Crushers. Mill
Operators Conference, Cobar

Aydogan N.A., Ergun L., Benzer H. (2006) . High Pressure Grinding Rolls (HPGR)
Applications in the Cement Industry. Minerals Engineering, vol. 19, pp. 130-139

Ayres, C., Knopjes, L., and Rule, C.M. 2008. Coarser feed applications of MIG IsaMilling.
MEI Comminution08 Conference, Falmouth, Cornwall, UK, 17-20 June 2008

Bailey C., Lane G., Morrell S., and Staples P. (2009). What Can Go Wrong in Comminution
Circuit Design? Proceedings of the Tenth Mill Operators Conference, Adelaide, SA, pp.
143-149

Barratt D. and Doll A. (2008). Testwork Programs that Deliver Multiple Data Sets of
Comminution Parameters for use in Mine Planning and Project Engineering
Proceedings of the V International Mineral Processing Seminar Procemin 2008,
Santiago, Chile, pp. 81-96

Barratt D. and Sherman M. 2002. Factors which influence the selection of comminution
circuits. Mineral Processing Plant Design, Practice, and Control Proceedings, eds. Mular
A.L., Halbe D.N., and Barratt D.J., Society for Mining, Metallurgy, and Exploration, Inc.
(SME), 1: 539-565

Baum W., Patzelt N. and Knecht J. (1996). The Use of High Pressure Roll Grinding for
Optimization of Copper Leaching. Society for Mining, Metallurgy, and Exploration, Inc.
(SME) Annual Meeting 11-14 March, Phoenix, Arizona. Preprint 96-68

131
Baum W., Patzelt N. and Knecht J. (1997). Metallurgical Benefits of High Pressure Roll
Grinding for Gold and Copper Recovery. Comminution Practices, ed. Kawatra, K.S.,
Society for Mining, Metallurgy, and Exploration, Inc. (SME), Chapter 16: 111-116

Bearman R. (2006). High-Pressure Grinding Rolls- Characterising and Defining Process

Performance for Engineers. Advances in Comminution, eds. Kawatra K.S., Society for
Mining, Metallurgy, and Exploration (SME), pp. 3-14

Benzer H., Aydogan N.A., and Dundar H. (2010). Investigation of the Breakage of Hard and
Soft Components under High Compression: HPGR Application. MEI Comminution10
Conference, Cape Town, South Africa, 13-16 April 2010

Bond F.C. (1964). Lab Equipment and Tests Help Predict Metal Consumption in Crushing
and Grinding Units Engineering and Mining Journal, vol. 165, June 1964, pp. 169-176

Bond F.C. (1985). General Aspects of Comminution History. SME Mineral Processing
Handbook, eds. Weiss N.L., vol. 1, section 3A, chapter 1, pp. 1-4 and pp. 3A-1 to 3A-4

Broeckmann C. and Gardula A. (2005). Developments in High-Pressure Grinding

Technology for Base and Precious Metal Minerals Processing. Proceedings of the 37th
Annual Meeting of the Canadian Mineral Processors, 18-20 January 2005, Ottawa,
Canada

Burger, B., McCaffery K., McGaffin I., Jankovic A., Valery W., and La Rosa D. (2006) Batu
Hijau Model for Throughput Forecast, Mining and Milling Optimization, and Expansion
Studies, Advances in Comminution , eds. Kawatra, K.S., Society for Mining, Metallurgy,
and Exploration, Inc. (SME), pp. 461-479

Charles, W.D., Gallagher, A.E.J. (1982). Comminution Energy Usage and Material Wear,
Design and Installation of Comminution Circuits, eds. Mular, A.L., Jergensen, G.V.,
Society for Mining, Metallurgy, and Exploration, Inc. S.M.E A.I.M.E., New York, vol. 2,
pp. 248274.

Cleary P.W., Sinnott M.D., Morrison R.D. (2008) DEM Prediction of Particle Flows in
Grinding Processes, Internationla Journal for Numerical Methods in Fluids, vol. 58, issue
3, pp. 319-353

Dance, A. 2004. Remove the disturbance: stabilising SAG mill tonnage through proper feed
preparation. SME Annual Meeting February 2004, Denver, USA. Preprint 04-110:1-5

132
Daniel, M. (2008). Summary of Research Outcomes of the P9N HPGR Project,
Incorporating Mineralogical Analyses, Particle Weakening and Micro-cracked HPGR
Products. Proceedings of the V International Mineral Processing Seminar -Procemin
2008, Santiago, Chile, pp.151-163

Daniel M.J. and Morrell S. (2004). HPGR Model Verification and Scale-up. Minerals
Engineering, vol. 17, pp 1149-116

Delboni H., Rosa M.A.N., Bergerman M.G., and Nardi R.P. 2006. Optimisation of the
Sossego SAG mill. SAG2006-SAG Mill Circuit. Int. Conf. on Autogenous and
Semiautogenous Grinding Technology 1: 39-50

Dixon S., Olson B., and Wipt E. 2010. Squeezing the extra 30% of a typical SABC Circuit
for 4.8 kWh/tonne. SME Annual Meeting Feb. 28- Mar. 03, 2010 Phoenix , USA.
Preprint 10-009

Dunne R., Goulsbra A. and Dunlop I. (1996). High Pressure Grinding Rolls and the Effect
on Liberation: Comparative Test Results . Proceedings Randol Gold Forum 1996,
Olympic Valley, CA, pp. 49-54

Dunne, R., Hart, S., Parker, B., and Veillette, G. 2007. Boddington gold mine an example
of sustaining gold production for 30 years. Proceedings of World Gold 2007 Conference,
Cairns, Australia, 22-24 October 2007

Fuerstenau D.W. and Abouzeid A.-Z.M. (1998). The Performance of the High Pressure Roll
Mill: Effect of Feed Moisture. Fizykochemiczne Problemy Mineralurgii, vol. 32 pp. 227-
241

Gardula A., Broeckmann C. and Paz A. (2005). Hexadur HIP Cladded Wear Protection
System for High Pressure Comminution of Minerals. Proceedings of the Randol
Innovative Metallurgy Forum held in Perth, W.A., Australia, 21-24 August 2005, pp. 31-
39

Gardula A. and Sheriff H. (2005). High Pressure Comminution of Gold Bearing Minerals
Potential for Increased Gold Recovery. Presentation at the 6th Annual Crushing and
Grinding Conference, Perth 21-23 March 2005

Goldcorp Inc. 2009. Peasquito Project Technical Report 10 March 2009. Via the internet,
April 2009, http://www.goldcorp.com/_resources/project_pdfs/penasquito2009.pdf

133
Gunter H., Muller M. and Sonntag P. (1996). The Application of Roller Presses for High
Pressure Comminution. Symposium on Grinding Processes, 14-15 February 1996
Toulouse, France

Gruendken, A., Matthies, E., and van der Meer, F.P. 2008. Flowsheet considerations for
optimal use of high pressure grinding rolls. MEI Comminution08 Conference, Falmouth,
Cornwall, UK, 17-20 June 2008

Halbe, D. and Smolik, T.J. (2002) Process Operating Costs with the Application in Mine
Planning and Risk Analysis Mineral Processing Plant Design, Practice, and Control
Proceedings, vol. 1, eds. Mular A.L., Halbe D.N., and Barratt D.J., Society for Mining,
Metallurgy, and Exploration, Inc. (SME), pp. 326-345

Hart, S., Valery Jr., W., Clements, B., Reed, M., Song, M., and Dunne, R. 2001. Optimisation
of the Cadia Hill SAG mill circuit. SAG2001-SAG Mill Circuit. Int. Conf. on Autogenous
and Semiautogenous Grinding Technology 1:11-30

Herbst, J., Rajamani, R.K., Mular, A., and Flintoff, B. (2002) Mineral Processing Plant/Circuit
Simulators: An Overview Mineral Processing Plant Design, Practice, and Control
Proceedings, vol. 1, eds. Mular A.L., Halbe D.N., and Barratt D.J., Society for Mining,
Metallurgy, and Exploration, Inc. (SME), pp. 383-403

Humphries G., Rule C. and Wolmarans E. (2006). The Development of a Process

Flowsheet for the New Anglo Platinum, PPRust North Concentrator, Incorporation HPGR
Technology. International Platinum Conference Platinum Surges Ahead. The Southern
African Institute of Mining and Metallurgy

International Copper Study Group (2009). The World Copper Factbook. Via the internet, July
2010, http://www.icsg.org/index.php?option=com_content&task=view&id=81&Itemid=64

Isamill 2009. Isamill overview. Via the internet, April 2009:

http://www.isamill.com/index.cfm?action=dsp_content& contentID=24

Jankovic A., Dundar H., and Mehta R. (2010). Relationships between Comminution Energy
and Product Size for a Magnetite Ore. The Journal of the Southern African Institute of
Mining, vol. 110, pp. 141-146

Jankovic A., Valery W., Davis E. (2004). Cement Grinding Optimisation. Minerals
Engineering, vol. 17, pp. 1075-1081

134
Jankovic A. and Valery Jr. W. 2004. Design and operation of Vertimill for secondary
Grinding. Presented at the 36th International Conference on Mining and Metallurgy,
29 September-2 October 2004, Bor Lake, Serbia and Montenegro

JKSimMet (2004). JKSimMet Steady State Mineral Processing Simulator User Manual.
JKTech, Queensland, Australia

Johansen, G., English, D., Lane, G., and Hayward, T. (2005), Bendigo Utilises High
Pressure Grinding Rolls to Maximise Recovery. AusIMM Bulletin, March/April 2005

Kaja D.M. (2002). BRUNO: Metso Minerals Plant Simulator. Mineral Processing Plant
Design, Practice and Control, eds. Mular A., Baratt, D, and Halbe D., Society for Mining,
Metallurgy, and Exploration (SME), vol. 1, pp. 404-420

Kavetsky A. (1979). Hydrocyclone modelling and scaling. JKMRC report to AMIRA,

November 1979

Kawatra S.K., Eisele T.C., and Walqui H.J. (2003) Optimization of Comminution Circuit
Throughput and Product Size Distribution by Simulation and Control Quarterly Technical
Progress Report October-December 2002, Dept of Chemical Eng. Michigan
Technological University, DOE Award Number: DE-FC26-01NT41062

Kawatra S.K. and Eisele T.C. (2005) Optimization of Comminution Circuit Throughput and
Product Size Distribution by Simulation and Control Final Technical Report - July 2005,
Dept of Chemical Eng. Michigan Technological University, DOE Award Number: DE-
FC26-01NT41062

Knecht J, inventor; Krupp Polysius Ag, assignee; (27 April 1999). Method of comminuting
ore material. United States patent US 5,897,063

KHD (2002). Roller Press Crushing and Grinding for Ore and Minerals Processing
Brochure 2-301e 11/02 E

KHD (2008). High Pressure Grinding Roller Presses Brochure 2-301.08e

Klingmann, H.L. (2005). Soledad Mountain Project Flow Sheet Development and Benefits
of the HPGR. Proceedings of the Randol Innovative Metallurgy Forum, Perth 21-24
August 2005

135
Klymowsky R., Patzelt N., Knecht J. and Burchardt E. (2006). An Overview of HPGR
Technology. Proceedings of the SAG Conference held in Vancouver, B.C., 23-27
September 2006, vol. 4, pp. 11-26

Koivistoinen, P. and Levanaho, J. 2006. The role of critical-sized material in AG and SAG
grinding. SAG2006-SAG Mill Circuit. Int. Conf. on Autogenous and Semiautogenous
Grinding Technology 2: 246-257

Koppern (2006). High Pressure Grinding Rolls and Hexadur Rolls: The Economical
Solution for the Mining and Mineral Processing Industry Brochure from April 2006

Lam, M., Jankovic, A., Valery, Jr. W., and Kanchibotla, S. 2001. Maximizing SAG mill circuit
throughput at Porgera Gold Mine by optimising blast fragmentation. SAG2001-SAG Mill
Circuit. Int. Conf. on Autogenous and Semiautogenous Grinding Technology 1:271-288

Latchireddi S. (2009). Improving Autogenous/Semi-Autogenous Grinding Performance and

Energy Efficiency. Proceedings of the 10th Mill Operators' Conference, Adelaide, SA,
pp. 89-93

Major, K. and Wells P. 2001. SABC cone crusher retrofits: where does the crusher go?
SAG2001-SAG Mill Circuit. Int. Conf. on Autogenous and Semiautogenous Grinding
Technology 1:215-233

McIvor, R.E. 1997. The effect of media sizing on ball milling efficiency. Comminution
Practices, ed. Kawatra, K.S., Society for Mining, Metallurgy, and Exploration, Inc. (SME),
Chapter 35: 279-292

Metso 2009. Vertimill for fine grinding. Via the internet, April 2009:
http://www.metsominerals.com/inetMinerals
/mm_home.nsf/FR?ReadForm&ATL=/inetMinerals/mm_grin.nsf/WebWID/WTB-041122-
2256F-74BB9

Morley C. 2006a. Flowsheets for HPGR. SAG2006-SAG Mill Circuit. Int. Conf. on
Autogenous and Semiautogenous Grinding Technology 4: 172-189

Morley, C. 2006b. High-pressure grinding rolls- a technology review. Advances in

Comminution , ed. Kawatra, K.S., Society for Mining, Metallurgy, and Exploration, Inc.
(SME). 15-40

136
Morley C. and Staples P. (2010). SAG or HPGR? The Current Dilemma. Proceedings of
the 42nd Annual Meeting of the Canadian Mineral Processors, 19-21 January 2010,
Ottawa, Canada

Morrell S. (1992). The Simulation of Autogenous and Semi-autogenous Milling Circuits

Comminution: Theory and Practice, eds. Kawatra K.S., (AIME), chapter 27, pp. 369-380

Morrell, S. (2008). A Method for Predicting the Specific Energy Requirement of

Comminution Circuits and Assessing their Energy Utilisation Efficiency. Minerals
Engineering, vol. 21, pp. 224-233

Morrell S., Lim W.I.L., Tondo L.A. and David D. (1996). Modelling the High Pressure
Grinding Rolls. Mining Technology Conference, pp. 169-176

Morrell, S., Lim, W., Shi, F., and Tondo, L. (1997). Modelling of the HPGR Crusher.
Comminution Practices, ed. Kawatra, K.S., Society for Mining, Metallurgy, and
Exploration, Inc. (SME), Chapter 17, pp. 117-126

Morrell, S. and Valery Jr. W. 2001. Influence of feed size on AG/SAG mill performance.
SAG2001-SAG Mill Circuit. Int. Conf. on Autogenous and Semiautogenous Grinding
Technology 1:203-214

Morrell S., Valery W. Banini, G., and Latchireddi, S. (2001). Developments in AG/SAG Mill
Modelling. SAG2001-SAG Mill Circuit. Int. Conf. on Autogenous and Semiautogenous
Grinding Technology vol. 4, pp. 71-84

Muoz, A., lvarez, L., Colacioppo, J., and Valery, W. (2008). Process Integration and
Optimisation at Freeport-McMoRan Candelaria Mine, Copiap, Chile Proceedings of the
V International Mineral Processing Seminar Procemin 2008, Santiago, Chile, pp. 291-
304

Musa F. and Morrison R. (2009), A More Sustainable Approach to Assessing Comminution

Efficiency, Minerals Engineering, vol. 22, pp. 593-601

Napier-Munn, T.J., Morrell, S., Morrison, R.D., Kojovic, T. (1996). Mineral Comminution
Circuits Their Operation and Optimisation., Julius Kruttschnitt Mineral Research
Centre, Monograph, vol. 2, The University of Queensland, Brisbane, Australia, pp.149-
150; pp. 154-191; and pp. 211-212

137
Norman T.E. and Decker J.D. (1985). Tumbling Mill Characteristics. SME Mineral
Processing Handbook, eds. Weiss N.L., vol. 1, section 3C, chapter 1, pp. 3C-3 to 3C-26

Patzelt, N., Knecht, H., Baum, W., (1995). Case made for high-pressure roll-grinding in gold
plants. Mining Engineering 47 (6), pp. 524529

Pokrajcic Z. and Morrison R. (2008). Evaluation by Simulation of Grinding Circuit Options to

Improve Eco-Efficiency. MEI Comminution08 Conference, Falmouth, Cornwall, UK, 17-
20 June 2008

Powell M. (2010). Enlightened Circuit Design Is Essential to the Take-up of New

Equipment. Draft paper and presentation for MEI Comminution10 Conference, Cape
Town, South Africa, 13-16 April 2010

Radziszewski, P. (1997), Predictive Model for Ball Mill Wear, Canadian Metallurgical
Quarterly, vol. 36, no. 2, pp. 87-93

Radziszewski P. (2002), Exploring Total Media Wear. Minerals Engineering, vol. 15,
pp. 1073-1087

Radziszewski, P., Varadi, R., Chenje, T., Santella, L., Sciannamblo, A. (2005), Tumbling Mill
Steel Media Abrasion Wear Test Development, Minerals Engineering, vol. 18, no. 3, pp.
333-341

Rosario P.P., Boyd K. and Grundy M. (2009). Recent Trends in the Design of Comminution
Circuits for High Tonnage Hard Rock Mining. Recent Advances in Mineral Processing
Plant Design, eds. Malhotra D., Taylor P.R., Spiller E., and LeVier M., Society for Mining,
Metallurgy, and Exploration, Inc. (SME), pp. 347-355

Rosario, P.P. and Hall, R.A. (2010). A Structured Approach to the Evaluation of the Energy
Requirements of HPGR and SAG Mill Circuits in Hard Ore Applications. The Journal of
the Southern African Institute of Mining, vol. 110, pp. 117-123

Rule, C. (2006). The development of a process flowsheet for the new Anglo Platinum,
PPRust North concentrator, incorporating HPGR technology. SAG2006-SAG Mill
Circuit. Int. Conf. on Autogenous and Semiautogenous Grinding Technology, vol. 4, pp.
94-109

138
Rule, C.M., Smit, I., Cope, A.J., and Humphries, G.A. (2008) Commissioning of the Polycom
2.2/1.6 5.6 MW HPGR at Anglo Platinums new Mogalakwena North Concentrator. MEI
Comminution08 Conference, Falmouth, Cornwall, UK, 17-20 June 2008

Schonert K. 1988. A first survey of grinding with high-compression roller mills. International
Journal of Mineral Processing 22: 401-412

Schonert K. 1991. Advances in comminution fundamental, and impacts on technology. XVII

International Mineral Processing Congress, Dresden, 23-28 September 1991, 1: 1-21

Seidel, J., Logan, T.C., LeVier, K.M., and Veillette, G. (2006). Case Study Investigation of
HPGR Suitability for two Gold/Copper Projects. SAG2006-SAG Mill Circuit. Int. Conf. on
Autogenous and Semiautogenous Grinding Technology, vol. 4, pp. 140-153

Sepulveda, J.E. 2008. The theoretical rationale behind current trends in operating practice
of semi-autogenous grinding operations. Presentation at V International Mineral
Processing Seminar Procemin 2008, Santiago, Chile

Shi F., Lambert S. and Daniel M. (2006). A Study of the Effects of HPGR Treating Platinum
Ores. Proceedings of the SAG Conference held in Vancouver, B.C., 23-27 September
2006, vol. 3, pp. 154-171

Spann, C., Ottergren, C. 2004. Is tertiary crushing replacing SAG milling as the optimal
grinding technique? The great debate 5th Annual Crushing & Grinding. Jupiters,
Townsville March 2004

Sullivan J. and Oliva, M (2007). Greenhouse Gases an Effective Strategy for Managing
GHG Emissions. Chemical Engineering, August 2007, pp. 34-40

Sylvestre, Y., Abols, J., and Barratt, D. 2001. The benefits of pre-crushing at Inmet Troilus
Mine. SAG2001-SAG Mill Circuit. Int. Conf. on Autogenous and Semiautogenous
Grinding Technology 3: 43-62

Tavares, L.M. (2005). Particle Weakening in High-Pressure Roll Grinding. Minerals

Engineering, vol. 18, June, pp. 651-657

Thomas, P.D.F. 1989. Development of S.A.G. mill motor drives. Advances in Autogenous
and Semiautogenous Grinding Technology, Mular A.R. and Agar G.E., Eds, Department
of Mining and Mineral Process Engineering, University of British Columbia 2:621-638

139
Tromans D., Meech J.A. (2002). A Fundamental Analysis of Fracture Mechanics of Minerals
during Comminution. Research being conducted at The Centre for Environmental
Research in Minerals, Metals and Materials (CERM3), UBC. Via the internet, July 2010,
http://www.mining.ubc.ca/cerm3/energy%20efficiency.html

Urrejola, S., Rodriguez, C., and Bobo, T. (2008) Design, Simulation and Optimisation of
SAG mill Circuit using the JKSimMet Simulator. Proceedings of the V International
Mineral Processing Seminar -Procemin 2008, Santiago, Chile, pp. 379-388

Valery W. and Jankovic A. (2002) . The Future of Comminution. Proceedings of 34th IOC on
Mining and Metallurgy Conference, 30 September-3 Octiber 2002, Bor Lake, Yugoslavia

van der Meer, F.P. and Gruendken, A (2009). Flowsheet considerations for optimal use of
high pressure grinding rolls. Minerals Engineering, doi:10.1016/j.mineng.2009.09.012 (in
press)

Van Drunick, W. and Smit, I. (2006). Energy Effcient Comminution HPGR Experience at
Anglo Research. Proceedings of the SAG Conference held in Vancouver, B.C., 23-27
September 2006, vol. 4, pp. 124-139

Vanderbeek, J.L. 1989. SAG mill recycle crushing plant performance at Chino Mines
Company. Advances in Autogenous and Semiautogenous Grinding Technology, Mular
A.R. and Agar G.E., Eds, Department of Mining and Mineral Process Engineering,
University of British Columbia 1:379-404

Vanderbeek, J.L. 1997. Tertiary grinding circuit installation at Chino Mines Company.
Comminution Practices, ed. Kawatra, K.S., Society for Mining, Metallurgy, and
Exploration, Inc. (SME), Chapter 31: 241-248

Vanderbeek J. (2004). State of the SAG. Proceedings of Plant Operators Forum 2004,
Society for Mining, Metallurgy, and Exploration (SME), pp. 141-145

Vanderbeek, J.L. (2006). HPGR Implementation at Cerro Verde. SAG2006-SAG Mill

Circuit. Int. Conf. on Autogenous and Semiautogenous Grinding Technology, vol 4, pp.
45-61

von Michaelis H. (2005). Real and Potential Metallurgical Benefits of HPGR in Hard Rock
Ore Processing. Proceedings of the Randol Innovative Metallurgy Forum held in Perth,
W.A., Australia, 21-24 August 2005, pp. 31-39

140
von Seebach, M. and Knobloch, O.R. (1987). High-Pressure Grinding Rolls in Industrial
Application. Presentation at the SME Annual Meeting, Denver, Colorado 24-27
February 1987. SME Preprint number 87-99

Walsh H. (2008). Chile Power Crunch May Cut Copper Output. Spur Record (Update2).
Bloomberg.com: Energy, 9 April 2008, URL.
http://www.bloomberg.com/apps/news?pid=20601086&sid=aJlEuFVIVRbs&refer=ne
ws

141
APPENDIX A: INPUTS USED FOR THE JKSIMMET MODELS

HPGR Pilot Test Scaling Data

Case A Case B
Rolls Diameter (mm) 800.0 710.0
Rolls Length (mm) 250.0 210.0
Working Gap (mm) 22.00 22.00
Speed (m/s) 0.670 0.290
Specific Comm. Energy (kWh/t) 1.680 1.726
Flake Density (t/m3) 2.620 2.360
Feed Bulk Density (t/m3) 1.700 1.604

HPGR Modelling Parameters - Pilot Test PSDs

Case A Case B
Particle Feed Product Particle Feed Product
Size Cum. Pas. Cum. Pas. Size Cum. Pas. Cum. Pas.
[mm] [%] [%] [mm] [%] [%]
45 100.00% 100.00% 45 100.00% 100.00%
31.5 100.00% 100.00% 31.5 100.00% 100.00%
22.4 93.31% 99.56% 22.4 91.10% 100.00%
16 84.64% 96.03% 16 81.90% 98.40%
11.2 77.67% 93.49% 11.2 68.70% 94.40%
8 66.24% 88.14% 8 57.60% 87.20%
5.6 49.55% 81.01% 4 39.70% 70.80%
2.8 22.04% 62.15% 2 28.20% 57.00%
1 12.75% 43.36% 1 18.80% 45.30%
0.5 9.87% 34.48% 0.5 13.60% 37.70%
0.315 8.40% 29.66% 0.2 8.40% 29.70%
0.2 7.01% 25.01% 0.09 5.60% 21.90%
0.125 5.77% 20.80% 0.045 4.00% 16.40%

142
 Feed PSDs for the Circuits
(Cases A & B)

Feed Cum.
Particle Size Pass.
[mm] [%]
270.2 100.00%
254 99.14%
203.2 95.56%
152.4 86.64%
127 80.61%
101.6 73.31%
76.2 62.26%
50.8 46.45%
25.4 30.65%
19.05 26.15%
12.7 21.13%
9.51 18.54%
6.35 15.38%
4.76 13.37%
3.36 11.23%
2.38 9.30%
1.68 7.71%
1.19 6.54%
0.841 5.39%
0.595 4.28%
0.42 3.56%
0.297 3.08%
0.21 2.65%
0.149 2.19%
0.105 1.79%
0.074 1.57%
0.053 1.46%
0.037 1.33%

143
 Crusher/HPGR Parameters from JK Drop-weight Tests

Case A - Appearance Function

Value of
t10 t75 t50 t25 t4 t2
10 3.64 4.27 5.86 20.64 44.06
20 6.89 8.22 11.37 39.83 67.74
30 10.35 12.51 17.20 55.60 82.17
50 17.03 20.70 28.48 70.45 99.70
Case B - Appearance Function
Value of
t10 t75 t50 t25 t4 t2
10 2.50 3.20 5.00 24.40 60.50
20 5.10 6.50 10.10 46.30 90.50
30 8.00 10.10 15.60 64.80 100.00
50 14.05 16.95 26.13 81.82 100.00

Case A - Breakage ECS Data

Value of
t10 14.53 20.63 28.89 Initial part. Size (mm)
10 0.19 0.18 0.18 ECS (kWh/t)
20 0.41 0.40 0.39
30 0.70 0.68 0.66
50 1.72 1.67 1.63
Case B - Breakage ECS Data
Value of
t10 14.53 20.63 28.89 Initial part. Size (mm)
10 0.56 0.45 0.36 ECS (kWh/t)
20 1.18 0.96 0.76
30 1.89 1.54 1.25
50 4.13 3.43 2.82

SAG Breakage Rate Constant

Parameters (Cases A & B)
Knot Size (mm) Constant
1 0.250 1.160
2 4.000 -0.910
3 16.00 0.750
4 44.80 -0.030
5 128.0 0.330

144
APPENDIX B: SMC AND MINNOVEX SPI TEST RESULTS

JKTech - SMC Drop-weight Test Results

DWI Mia Mih Mic Relative

Sample Name A b Axb ta
(kWh/m3) (kWh/t) (kWh/t) (kWh/t) Density
W3 HPGR Feed 62.9 0.55 34.6 0.33 7.81 22.0 16.8 8.7 2.70

Specific Comminution Energy

particle size (mm)
14.5 14.5 14.5 28.9 28.9 28.9 57.8 57.8 57.8
t10
10 20 30 10 20 30 10 20 30
kWh/t kWh/t kWh/t kWh/t kWh/t kWh/t kWh/t kWh/t kWh/t
0.41 0.89 1.46 0.31 0.65 1.05 0.23 0.49 0.76

MinnovEX SPI Test Results

CEET
SPI Hardness
Sample Name Crusher
(minute) Percentile
Index (Ci)
W3 HPGR Feed 41.8 94.4 65

145
APPENDIX C: SAMPLE PREPARATION AND TEST FLOWSHEET

146
147
APPENDIX D: HPGR FEED TEST BLEND LINEAR PROGRAMING

The calculation procedure for the determination of the optimum blend between the

laboratory screen oversize material and the crusher product is described through the

applied calculation sequence and the corresponding table of results, as follows:

1. The full-scale simulated cone crusher product PSD was scaled down by a

constant 81% factor to reduce the top size from 71 mm to 50 mm and the P80

from 28 mm to 22 mm (assumed for the lab-scale procedure) and to preserve the

PSD shape.

2. Based on the PSDs of the crusher product and the screen oversize material,

which were determined through the laboratory tests, the size distribution for 100

kg of each product were estimated.

3. Having the ideal distribution determined as well as the distributions of the two

components, a linear objective function was set and executed (using the Excels

Solver add-in functionality) to determined the proportion of the parts (a and b)

that would generate a blend that best approximate the ideal distribution. The

objective function was set to find the blend that minimizes the sum of the

standard deviation values calculated for all size fractions of the ideal and the

possible blends, as following expressed:

Minimize

Subject to: 0 a 1
0 b 1
a + b =1

148
Table of Results HPGR Feed Test Blend Determination

1 2 3
Full-scale Screen Screen Standard
Ideal Crushed Crushed Targeted
Preliminary Scaled-down Particle Size Distribution O/S O/S Deviations
Distrib. Prod. Prod. Blend
Simulation PSD Prod. Prod. (ideal-target)
Constant a b a+b Objective
Factors
0.81 0.214 0.786 1.000 Minimum Sum
Simul. C. Pas. New C. Pas. Stand. C. Pas. Weight Screen Screen Screen
Sizes (simul.) Sizes (same) Sizes adjusted Perc. Weight Weight Weight
[mm] [%] [mm] [%] [mm] [%] [%] [kg] [kg] [kg] [kg] [kg] [kg] [kg]
70.71 100% 57.3 100% 50 100% 0.00% 0.00 0.00 0.00 0.00 0.00 0.00 0.00
50 99.9% 40.5 100% 35.36 98.4% 1.63% 1.65 0.00 0.00 0.00 0.00 0.00 1.17
35.36 96.3% 28.6 96.3% 32 97.3% 1.05% 1.07 0.00 2.28 0.00 1.79 1.79 0.51
32 89.4% 25.9 89.4% 25 87.1% 10.19% 10.31 0.00 10.65 0.00 8.36 8.36 1.38
25 74.8% 20.3 74.8% 19 71.6% 15.54% 15.73 0.09 32.35 0.02 25.42 25.44 6.86
19 56.6% 15.4 56.6% 16 58.8% 12.73% 12.89 1.60 11.66 0.34 9.16 9.50 2.39
16 45.7% 13.0 45.7% 12.5 43.7% 15.17% 15.36 8.56 9.99 1.83 7.85 9.68 4.01
12.5 33.0% 10.1 33.0% 8 23.8% 19.92% 20.16 39.91 14.78 8.55 11.61 20.16 0.00
8 17.1% 6.5 17.1% 5.6 15.0% 8.73% 8.84 20.75 11.45 4.45 9.00 13.45 3.26
5.6 12.5% 4.5 12.5% 4 11.4% 3.59% 3.63 9.69 3.25 2.08 2.55 4.63 0.71
4 10.0% 3.2 10.0% 2 7.6% 3.87% 3.92 8.34 1.45 1.79 1.14 2.93 0.70
2 6.8% 1.6 6.8% 1 5.4% 2.15% 2.18 4.06 0.86 0.87 0.67 1.54 0.45
1 5.0% 0.810 5.0% 0.5 4.0% 1.39% 1.40 2.18 0.47 0.47 0.37 0.83 0.40
0.5 3.7% 0.405 3.7% 0.25 3.0% 1.00% 1.01 1.46 0.27 0.31 0.21 0.53 0.34
0.25 2.8% 0.203 2.8% 0.125 2.3% 0.75% 0.76 1.11 0.17 0.24 0.14 0.37 0.27
0.125 2.1% 0.101 2.1% 0.045 1.5% 0.79% 0.80 1.06 0.17 0.23 0.13 0.36 0.32
0.045 1.4% 0.036 1.4% 0.02 1.2% 0.28% 0.28 1.19 0.21 0.25 0.16 0.42 0.09
Totals 100 100 100 21.43 78.57 100 22.87

149
APPENDIX E: HPGR TESTS COMPLETE DATA

Roller Diameter (D) [m] 0.750

HPGR
Roller Width (W) [m] 0.220

Test Number: #001 #002 #003 #005 #006 #007

Data Description
Symbol Unit Pressure Tests - Cycle 1 Cycle 2 Cycle 3 Full Feed
Process Set Points

[m/s] 0.75 0.75 0.75 0.75 0.75 0.75

Speed
n [rpm] 19.10 19.10 19.10 19.10 19.10 19.10
Static Gap X0 [mm] 9.0 9.0 9.0 9.0 9.0 9.0
Hydraulic Pressure P [bar] 82 62 72 72 72 72
Pressing Force F [kN] 659.5 494.6 577.0 577.0 577.0 577.0
Specific Pressing Force FSP [N/mm2] 4 3 3.5 3.5 3.5 3.5
Test Time t [s] 19.59 23.80 20.60 20.40 20.20 19.59
Average Actual Speed: AV [m/s] 0.75 0.76 0.75 0.73 0.75 0.74
Standard Deviation 0.10 0.22 0.08 0.11 0.09 0.07
Actual Roller gap (average) XgAV [mm] 17.61 18.95 18.51 17.49 17.81 20.62
Standard Deviation X 0.50 0.37 0.51 0.52 0.58 0.54
Actual Hydraulic Pressure (average) PAV [bar] 81.4 61.3 71.6 71.1 69.5 69.9
Process Data

Standard Deviation 0.6 1.5 1.7 0.7 2.4 1.0

Actual Pressing Force (average) FAV [kN] 654.7 492.7 575.7 571.7 558.9 562.3
Actual Specific Pressure (average) FSPAV [kN/m2] 4.0 3.0 3.50 3.5 3.4 3.42
Idle Power Draw Pi [kW] 9.39 8.73 9.70 9.64 10.61 10.96
Power Draw P [kW] 61.3 51.4 58.0 56.1 54.6 61.1
Total Specific Energy Consumption ESP [kWh/t] 2.35 1.89 2.14 2.23 2.09 2.05
Net Specific Energy Consumption ESP net [kWh/t] 1.99 1.57 1.79 1.82 1.68 1.68
Press throughput W [t/h] 26.02 27.19 27.04 25.50 26.12 29.77
Specific Throughput Constant m dot [ts/hm3] 210.78 217.63 215.25 216.24 209.37 246.93

150
 Test Number: #001 #002 #003 #005 #006 #007
Data Description
Symbol Unit Cycles Full Feed
Cycle No. 1 2 3
Average Cake Density rC [t/m3] 2.33 2.30 2.30 2.29 2.32 2.30
Cake Thickness Average XC [mm] 20.0 20.5 20.4 21.8 19.8 24.0
Feed Moisture [%] 1.63% 1.63% 1.63% 1.56% 1.23% 1.72%
Feed Bulk Density [t/m3] 1.68 1.68 1.68 1.68 1.68 1.87
Feed: 80% Passing Size F80 [mm] 21.00 21.00 21.00 20.08 18.41 21.26
Feed: 50% Passing Size F80 [mm] 14.44 14.44 14.44 11.59 10.72 12.83
Centre: 80% Passing Size P80 [mm] 5.16 6.02 5.66 4.59 5.43 5.52
Centre: 50% Passing Size P50 [mm] 2.13 2.47 2.46 2.17 2.27 2.34
Edge: 80% Passing Size P80 [mm] 8.96 10.09 9.75 8.83 8.71 10.15
Edge: 50% Passing Size P50 [mm] 4.17 4.64 4.75 4.26 4.23 4.03
Material Data

90% Center & 10% Edge: 80% Passing Size P80 [mm] 5.40 6.48 6.18 5.42 5.70 5.93
90% Center & 10% Edge: 50% Passing Size P50 [mm] 2.28 2.62 2.63 2.32 2.42 2.46
Reduction Ratio F80/P80 (Centre & Edge) 3.89 3.24 3.40 3.71 3.23 3.59
Reduction Ratio F50/P50 (Centre & Edge) 6.34 5.50 5.48 4.99 4.43 5.22
Percentage Passing 6.35 mm (Centre & Edge) [%] 84.5% 79.4% 80.7% 84.7% 82.8% 81.5%
Total Feed Material MF [kg] 297 296 297 266 254.3 224
Total Centre Product MC [kg] 96.4 120 102.9 98.5 99.9 110.9
Centre Product % of Centre & Edge Material MCE% [%] 68.1% 67.9% 67.8% 68.2% 68.1% 68.8%
Total Edge Product ME [kg] 45.2 56.8 48.8 46.0 46.7 50.2
Edge Product % of Centre & Edge Material MEF% [%] 31.9% 32.1% 32.2% 31.8% 31.9% 31.2%
Edge Product % of Centre Product MEC% [%] 46.9% 47.3% 47.4% 46.7% 46.7% 45.3%
Total Waste Product MW [kg] 152.1 116.8 145.3 118.7 103.8 56.7
Waste Product % of Total Feed MWF% [%] 51.2% 39.5% 48.9% 44.6% 40.8% 25.3%
Total Recovered Product MP [kg] 294 294 297 263 250 218
Mass Reconciliation (+ "gain; - "loss") MPF% [%] -1.1% -0.8% 0.0% -1.1% -1.5% -2.8%
Throughput (Continuity Formula - Based on Op Gap) [t / h] 24.39 26.31 25.36 23.10 24.63 27.95
Mdot - Coninuity Formula (Based on Op Gap) m-dot [ts/hm3] 197.17 208.91 204.67 192.29 198.37 228.01
Throughput (Continuity For.- Based on Cake Thick.) [t/h] 27.75 28.51 27.98 28.85 27.36 32.58
Mdot - Coninuity Formula (Based on Cake Thickness) m-dot [ts/hm3] 224.39 226.36 225.83 240.18 220.37 265.85
Mdot (Cont. For. Average: oper. gap & cake thick. ) m-dot [ts/hm3] 210.78 217.63 215.25 216.24 209.37 246.93
Mdot (Calculated based on test reporting data) m-dot [ts/hm3] 210.90 216.51 218.86 212.85 210.94 243.55

151
 HPGR Test #1
Pressure: 4.0 N/mm2

Weight (Kg)
Feed 297
Waste 152.1 HPGR Center Product HPGR Edge Product
Center 96.4 Particle Screen Weight Cum. Screen Weight Cum.
Edge 45.2 31.9% Size Weight Perc. Pass. Weight Perc. Pass.
[mm] [g] [%] [%] [g] [%] [%]
Moisture 1.63% 32 0.0 0.0% 100.0% 0.0 0.0% 100.0%
BOND BMWi 19.3 kw-hr/tonne 25 0.0 0.0% 100.0% 0.0 0.0% 100.0%
19 7.6 0.1% 99.9% 22.7 0.2% 99.8%
HPGR Feed #1 to#3 Feed 16 8.4 0.1% 99.9% 153.3 1.4% 98.4%
Screen Weight Cum. 12.5 107.4 0.9% 98.9% 460.1 4.1% 94.4%
Screen Size Particle Size Weight Perc. Pass. 8 772.3 6.8% 92.1% 2072.7 18.3% 76.1%
[mm] [mm] [g] [%] [%] 5.6 920.7 8.1% 84.0% 1572.4 13.9% 62.2%
-35.5 to +32 32 0.0 0.0% 100.0% 4 1646.4 14.5% 69.4% 1549.4 13.7% 48.5%
-32 to +25 25 347.0 3.1% 96.9% 2.8 1302.6 11.5% 57.9% 1083.4 9.6% 39.0%
-25 to +19 19 2842.5 25.4% 71.5% 2 1067.8 9.4% 48.5% 830.0 7.3% 31.6%
-19 to +16 16 1716.8 15.3% 56.2% 1.4 976.3 8.6% 39.9% 637.7 5.6% 26.0%
-16 to +12.5 12.5 1566.3 14.0% 42.3% 1 733.6 6.5% 33.4% 515.8 4.6% 21.5%
-12.5 to +8 8 2811.4 25.1% 17.2% 0.71 609.7 5.4% 28.0% 436.4 3.9% 17.6%
-8 to +5.6 5.6 1211.9 10.8% 6.4% 0.5 504.9 4.5% 23.5% 344.6 3.0% 14.6%
-5.6 to +4 4 349.2 3.1% 3.3% 0.355 420.2 3.7% 19.8% 280.2 2.5% 12.1%
-4 to +2 2 175.6 1.6% 1.7% 0.25 369.7 3.3% 16.6% 228.0 2.0% 10.1%
-2 to +1 1 60.2 0.5% 1.2% 0.18 306.9 2.7% 13.8% 200.7 1.8% 8.3%
-1 to +0.5 0.5 28.7 0.3% 0.9% 0.125 285.8 2.5% 11.3% 167.5 1.5% 6.8%
-0.5 to +0.25 0.25 24.7 0.2% 0.7% 0.09 181.1 1.6% 9.7% 113.8 1.0% 5.8%
-.25 to +.125 0.125 23.0 0.2% 0.5% 0.063 210.8 1.9% 7.9% 122.9 1.1% 4.7%
-.125 to +.045 0.045 24.9 0.2% 0.2% 0.045 134.1 1.2% 6.7% 73.9 0.7% 4.1%
-0.045 Pan 27.8 0.2% 0.0% Pan 756.1 6.7% 0.0% 463.5 4.1% 0.0%
Total 11210.0 100.0% 11322.4 100.0% 11329.0 100.0%

Calc. P80 [mm] 21.00 Calc. P80 [mm] 5.16 Calc. P80 [mm] 8.96
Calc. P50 [mm] 14.44 Calc. P50 [mm] 2.13 Calc. P50 [mm] 4.17
-6.35mm % 9.7% -6.35mm % 86.5% -6.35mm % 66.5%
-1.0 mm % 1.2% -1.0 mm % 28.0% -1.0 mm % 17.6%

152
 HPGR Test #2
Pressure: 3.0 N/mm2

Weight (Kg)
Feed 296
Waste 116.8 HPGR Center Product HPGR Edge Product
Center 120 Particle Screen Weight Cum. Screen Weight Cum.
Edge 56.8 32.1% Size Weight Perc. Pass. Weight Perc. Pass.
[mm] [g] [%] [%] [g] [%] [%]
Moisture 1.63% 32 0.0 0.0% 100.0% 0.0 0.0% 100.0%
BOND BMWi 19.3 kw-hr/tonne 25 0.0 0.0% 100.0% 18.9 0.2% 99.8%
19 0.0 0.0% 100.0% 71.8 0.7% 99.1%
HPGR Feed Feed 16 82.9 0.8% 99.2% 209.8 2.1% 97.1%
Screen Weight Cum. 12.5 197.5 1.9% 97.3% 701.1 6.9% 90.2%
Screen Size Particle Size Weight Perc. Pass. 8 928.7 9.1% 88.2% 1937.1 19.0% 71.2%
[mm] [mm] [g] [%] [%] 5.6 1012.6 9.9% 78.3% 1380.1 13.5% 57.7%
-35.5 to +32 32 0.0 0.0% 100.0% 4 1375.6 13.4% 64.8% 1310.0 12.8% 44.8%
-32 to +25 25 347.0 3.1% 96.9% 2.8 1139.7 11.1% 53.7% 928.8 9.1% 35.7%
-25 to +19 19 2842.5 25.4% 71.5% 2 909.5 8.9% 44.8% 654.1 6.4% 29.3%
-19 to +16 16 1716.8 15.3% 56.2% 1.4 817.2 8.0% 36.8% 589.2 5.8% 23.5%
-16 to +12.5 12.5 1566.3 14.0% 42.3% 1 617.8 6.0% 30.8% 453.7 4.4% 19.1%
-12.5 to +8 8 2811.4 25.1% 17.2% 0.71 524.7 5.1% 25.6% 276.3 2.7% 16.4%
-8 to +5.6 5.6 1211.9 10.8% 6.4% 0.5 440.2 4.3% 21.3% 287.3 2.8% 13.6%
-5.6 to +4 4 349.2 3.1% 3.3% 0.355 342.0 3.3% 18.0% 235.1 2.3% 11.3%
-4 to +2 2 175.6 1.6% 1.7% 0.25 313.3 3.1% 14.9% 199.3 2.0% 9.3%
-2 to +1 1 60.2 0.5% 1.2% 0.18 251.6 2.5% 12.5% 156.2 1.5% 7.8%
-1 to +0.5 0.5 28.7 0.3% 0.9% 0.125 237.5 2.3% 10.1% 146.7 1.4% 6.4%
-0.5 to +0.25 0.25 24.7 0.2% 0.7% 0.09 156.4 1.5% 8.6% 101.7 1.0% 5.4%
-.25 to +.125 0.125 23.0 0.2% 0.5% 0.063 180.1 1.8% 6.9% 96.3 0.9% 4.4%
-.125 to +.045 0.045 24.9 0.2% 0.2% 0.045 104.7 1.0% 5.8% 73.2 0.7% 3.7%
-0.045 Pan 27.8 0.2% 0.0% Pan 596.6 5.8% 0.0% 376.8 3.7% 0.0%
Total 11210.0 100.0% 10228.6 100.0% 10203.5 100.0%

Calc. P80 [mm] 21.00 Calc. P80 [mm] 6.02 Calc. P80 [mm] 10.09
Calc. P50 [mm] 14.44 Calc. P50 [mm] 2.47 Calc. P50 [mm] 4.64
-6.35mm % 9.7% -6.35mm % 81.4% -6.35mm % 61.9%
-1.0 mm % 1.2% -1.0 mm % 25.6% -1.0 mm % 16.4%

153
HPGR Test #3
Pressure: 3.5 N/mm2

Weight (Kg)
Feed 297
Waste 145.3 HPGR Center Product HPGR Edge Product
Center 102.9 Particle Screen Weight Cum. Screen Weight Cum.
Edge 48.8 32.2% Size Weight Perc. Pass. Weight Perc. Pass.
[mm] [g] [%] [%] [g] [%] [%]
Moisture 1.63% 32 0.0 0.0% 100.0% 0.0 0.0% 100.0%
BOND BMWi 19.3 kw-hr/tonne 25 0.0 0.0% 100.0% 0.0 0.0% 100.0%
19 11.1 0.1% 99.9% 0.0 0.0% 100.0%
Feed test #3 (1st cycle) Feed 16 24.6 0.3% 99.6% 168.4 1.9% 98.1%
Particle Screen Weight Cum. 12.5 145.2 1.5% 98.1% 467.1 5.3% 92.7%
Screen Size Size Weight Perc. Pass. 8 833.0 8.5% 89.6% 1822.0 20.8% 71.9%
[mm] [mm] [g] [%] [%] 5.6 966.0 9.9% 79.7% 1277.8 14.6% 57.3%
-35.5 to +32 32 0.0 0.0% 100.0% 4 1391.9 14.2% 65.5% 1196.9 13.7% 43.6%
-32 to +25 25 347.0 3.1% 96.9% 2.8 1139.5 11.7% 53.8% 850.3 9.7% 33.9%
-25 to +19 19 2842.5 25.4% 71.5% 2 883.8 9.0% 44.8% 596.7 6.8% 27.1%
-19 to +16 16 1716.8 15.3% 56.2% 1.4 800.7 8.2% 36.6% 429.9 4.9% 22.2%
-16 to +12.5 12.5 1566.3 14.0% 42.3% 1 580.4 5.9% 30.7% 359.4 4.1% 18.1%
-12.5 to +8 8 2811.4 25.1% 17.2% 0.71 434.8 4.4% 26.2% 246.9 2.8% 15.2%
-8 to +5.6 5.6 1211.9 10.8% 6.4% 0.5 432.8 4.4% 21.8% 245.1 2.8% 12.4%
-5.6 to +4 4 349.2 3.1% 3.3% 0.355 334.0 3.4% 18.4% 197.4 2.3% 10.2%
-4 to +2 2 175.6 1.6% 1.7% 0.25 319.2 3.3% 15.1% 173.3 2.0% 8.2%
-2 to +1 1 60.2 0.5% 1.2% 0.18 247.0 2.5% 12.6% 139.1 1.6% 6.6%
-1 to +0.5 0.5 28.7 0.3% 0.9% 0.125 232.6 2.4% 10.2% 129.3 1.5% 5.1%
-0.5 to +0.25 0.25 24.7 0.2% 0.7% 0.09 148.5 1.5% 8.7% 189.0 2.2% 3.0%
-.25 to +.125 0.125 23.0 0.2% 0.5% 0.063 168.3 1.7% 7.0% 85.9 1.0% 2.0%
-.125 to +.045 0.045 24.9 0.2% 0.2% 0.045 107.2 1.1% 5.9% 55.2 0.6% 1.4%
-0.045 Pan 27.8 0.2% 0.0% Pan 572.2 5.9% 0.0% 119.3 1.4% 0.0%
Total 11210.0 100.0% 9772.8 100.0% 8749.0 100.0%

Calc. P80 [mm] 21.00 Calc. P80 [mm] 5.66 Calc. P80 [mm] 9.75
Calc. P50 [mm] 14.44 Calc. P50 [mm] 2.46 Calc. P50 [mm] 4.75
-6.35mm % 9.7% -6.35mm % 82.8% -6.35mm % 61.9%
-1.0 mm % 1.2% -710um % 26.2% -710um % 15.2%

154
HPGR Test #5 (1st Recirculation)
Pressure: 3.5 N/mm2
Weight (Kg)
Feed 266
Waste 118.7
Center 98.5
Edge 46 31.8%
Moisture 1.56%

Feed test #5 (2nd cycle) Feed HPGR Center Product HPGR Edge Product
Screen Size Particle Size Weight Cum. Particle Screen Weight Cum. Screen Weight Cum.
Screen Weight Perc. Pass. Size Weight Perc. Pass. Weight Perc. Pass.
[mm] [mm] [g] [%] [%] [mm] [g] [%] [%] [g] [%] [%]
-35.5 to +32 32 0.0 0.0% 100.0% 32 0.0 0.0% 100.0% 0.0 0.0% 100.0%
-32 to +25 25 549.3 4.4% 95.6% 25 0.0 0.0% 100.0% 0.0 0.0% 100.0%
-25 to +19 19 2399.6 19.1% 76.6% 19 0.0 0.0% 100.0% 0.0 0.0% 100.0%
-19 to +16 16 1254.3 10.0% 66.6% 16 7.7 0.1% 99.9% 163.7 1.5% 98.5%
-16 to +12.5 12.5 1380.9 11.0% 55.7% 12.5 83.6 0.8% 99.1% 419.0 3.9% 94.6%
-12.5 to +8 8 3510.3 27.9% 27.8% 8 668.8 6.3% 92.9% 1948.5 17.9% 76.7%
-8 to +5.6 5.6 2083.9 16.5% 11.2% 5.6 967.3 9.1% 83.8% 1590.5 14.6% 62.1%
-5.6 to +4 4 713.8 5.7% 5.6% 4 1547.6 14.5% 69.3% 1568.2 14.4% 47.6%
-4 to +2.8 2.8 187.0 1.5% 4.1% 2.8 1277.2 12.0% 57.4% 1077.1 9.9% 37.7%
-2.8 to +2 2 108.8 0.9% 3.2% 2 990.6 9.3% 48.1% 741.8 6.8% 30.9%
-2 to +1.4 1.4 82.6 0.7% 2.6% 1.4 933.0 8.7% 39.3% 697.1 6.4% 24.5%
-1.4 to +1 1 72.2 0.6% 2.0% 1 685.2 6.4% 32.9% 468.0 4.3% 20.2%
-1 to +.71 0.71 45.1 0.4% 1.6% 0.71 594.5 5.6% 27.4% 385.4 3.5% 16.6%
-.71 to +.5 0.5 47.8 0.4% 1.3% 0.5 482.1 4.5% 22.8% 315.0 2.9% 13.7%
-.5 to +.355 0.355 38.0 0.3% 1.0% 0.355 392.6 3.7% 19.2% 252.8 2.3% 11.4%
-.355 to +.25 0.25 31.6 0.3% 0.7% 0.25 348.7 3.3% 15.9% 218.6 2.0% 9.4%
-.25 to +.18 0.18 23.0 0.2% 0.5% 0.18 283.8 2.7% 13.2% 177.6 1.6% 7.7%
-.18 to +.125 0.125 1.6 0.0% 0.5% 0.125 261.2 2.4% 10.8% 154.1 1.4% 6.3%
-.125 to +.09 0.09 7.6 0.1% 0.5% 0.09 167.0 1.6% 9.2% 110.7 1.0% 5.3%
-.09 to +.063 0.063 6.3 0.1% 0.4% 0.063 193.6 1.8% 7.4% 97.8 0.9% 4.4%
-.063 to +.045 0.045 9.9 0.1% 0.3% 0.045 119.1 1.1% 6.3% 75.4 0.7% 3.7%
-0.045 Pan 40.8 0.3% 0.0% Pan 673.1 6.3% 0.0% 402.7 3.7% 0.0%
Total 12594.5 100.0% 10676.7 100.0% 10864.0 100.0%
Calc. P80 [mm] 20.08 Calc. P80 [mm] 4.59 Calc. P80 [mm] 8.83
Calc. P50 [mm] 11.59 Calc. P50 [mm] 2.17 Calc. P50 [mm] 4.26
-6.35mm % 16.4% -6.35mm % 86.7% -6.35mm % 66.6%
-1.0mm % 2.0% -710um % 27.4% -710um % 16.6%
-710um % 1.6%

155
HPGR Test #6 (2nd Recirculation)
Pressure: 3.5 N/mm2
Weight (Kg)
Feed 254.3 HPGR Center Product HPGR Edge Product
Waste 103.8 Particle Screen Weight Cum. Screen Weight Cum.
Center 99.9 Size Weight Perc. Pass. Weight Perc. Pass.
Edge 46.7 31.9% [mm] [g] [%] [%] [g] [%] [%]
32 0.0 0.0% 100.0% 0.0 0.0% 100.0%
Moisture 1.23% 25 0.0 0.0% 100.0% 0.0 0.0% 100.0%
19 0.0 0.0% 100.0% 9.3 0.1% 99.9%
Feed test #6 (3rd cycle) Feed 16 14.3 0.1% 99.9% 134.8 1.2% 98.7%
Screen Size Particle Size Screen Weight Weight %. C. Pass. 12.5 178.3 1.4% 98.4% 456.3 4.0% 94.8%
8 879.1 7.1% 91.3% 2005.9 17.5% 77.2%
[m] [mm] [g] [%] [%] 5.6 1220.6 9.9% 81.4% 1691.1 14.8% 62.5%
-35.5 to +32 32 0.0 0.0% 100.0% 4 1707.0 13.8% 67.6% 1661.1 14.5% 48.0%
-32 to +25 25 362.8 2.9% 97.1% 2.8 1423.1 11.5% 56.1% 1143.2 10.0% 38.0%
-25 to +19 19 1859.8 15.1% 82.0% 2 1140.6 9.2% 46.9% 774.7 6.8% 31.2%
-19 to +16 16 1250.6 10.1% 71.9% 1.4 1048.6 8.5% 38.4% 717.1 6.3% 24.9%
-16 to +12.5 12.5 1317.1 10.7% 61.2% 1 743.2 6.0% 32.4% 503.3 4.4% 20.5%
-12.5 to +8 8 3485.3 28.2% 33.0% 0.71 657.1 5.3% 27.1% 422.8 3.7% 16.9%
-8 to +5.6 5.6 2392.3 19.4% 13.6% 0.5 554.1 4.5% 22.6% 332.4 2.9% 14.0%
-5.6 to +4 4 885.9 7.2% 6.4% 0.355 440.6 3.6% 19.0% 260.4 2.3% 11.7%
-4 to +2 2 357.8 2.9% 3.5% 0.25 399.5 3.2% 15.8% 229.3 2.0% 9.7%
-2 to +1 1 161.3 1.3% 2.2% 0.18 315.6 2.6% 13.2% 184.7 1.6% 8.1%
-1 to +0.5 0.5 85.6 0.7% 1.5% 0.125 300.9 2.4% 10.8% 169.9 1.5% 6.6%
-0.5 to +0.25 0.25 54.1 0.4% 1.1% 0.09 189.4 1.5% 9.3% 115.0 1.0% 5.6%
-.25 to +.125 0.125 42.8 0.3% 0.7% 0.063 226.4 1.8% 7.4% 115.5 1.0% 4.6%
-.125 to +.045 0.045 45.8 0.4% 0.4% 0.045 138.7 1.1% 6.3% 76.1 0.7% 3.9%
-0.045 Pan 44.3 0.4% 0.0% Pan 780.3 6.3% 0.0% 446.4 3.9% 0.0%
Total 12345.5 100.0% 12357.4 100.0% 11449.3 100.0%
Calc. P80 [mm] 18.41 Calc. P80 [mm] 5.43 Calc. P80 [mm] 8.71
Calc. P50 [mm] 10.72 Calc. P50 [mm] 2.27 Calc. P50 [mm] 4.23
-6.35mm % 19.6% -6.35mm % 84.5% -6.35mm % 67.1%
-1.0 mm % 2.2% -710um % 27.1% -710um % 16.9%

156
HPGR Test #7 (Fresh Feed)
Pressure: 3.5 N/mm2
Weight (Kg)
Feed 224
Waste 56.7
Center 110.9
Edge 50.2 31.2%
Moisture 1.72%
BOND BMWi 16.6 kw-hr/tonne

HPGR Feed #7 (FF Drum4) Feed HPGR Center Product HPGR Edge Product
Screen Weight Cum. Particle Screen Weight Cum. Screen Weight Cum.
Screen Size Particle Size Weight Perc. Pass. Size Weight Perc. Pass. Weight Perc. Pass.
[m] [mm] [g] [%] [%] [mm] [g] [%] [%] [g] [%] [%]
-35.5 to +32 32 0.0 0.0% 100.0% 32 0.0 0.0% 100.0% 0.0 0.0% 100.0%
-32 to +25 25 645.8 7.4% 92.6% 25 0.0 0.0% 100.0% 0.0 0.0% 100.0%
-25 to +19 19 1776.3 20.3% 72.4% 19 51.8 0.4% 99.6% 162.6 1.3% 98.7%
-19 to +16 16 1187.8 13.5% 58.8% 16 104.7 0.9% 98.7% 376.8 3.0% 95.7%
-16 to +12.5 12.5 856.4 9.8% 49.1% 12.5 259.4 2.1% 96.6% 938.3 7.5% 88.1%
-12.5 to +8 8 1100.5 12.5% 36.5% 8 963.6 7.9% 88.7% 1933.6 15.6% 72.6%
-8 to +5.6 5.6 655.8 7.5% 29.1% 5.6 980.9 8.0% 80.7% 1271.3 10.2% 62.3%
-5.6 to +4 4 472.8 5.4% 23.7% 4 1680.2 13.7% 67.0% 1563.3 12.6% 49.8%
-4 to +2.8 2.8 399.1 4.6% 19.1% 2.8 1427.9 11.7% 55.3% 1126.7 9.1% 40.7%
-2.8 to +2 2 263.8 3.0% 16.1% 2 1138.2 9.3% 46.0% 890.5 7.2% 33.5%
-2 to +1.4 1.4 256.3 2.9% 13.2% 1.4 1048.8 8.6% 37.5% 655.6 5.3% 28.3%
-1.4 to +1 1 204.6 2.3% 10.8% 1 743.1 6.1% 31.4% 565.0 4.5% 23.7%
-1 to +.71 0.71 156.2 1.8% 9.1% 0.71 656.1 5.4% 26.0% 573.3 4.6% 19.1%
-.71 to +.5 0.5 135.3 1.5% 7.5% 0.5 538.7 4.4% 21.6% 421.1 3.4% 15.7%
-.5 to +.355 0.355 113.9 1.3% 6.2% 0.355 433.1 3.5% 18.1% 330.1 2.7% 13.1%
-.355 to +.25 0.25 91.6 1.0% 5.2% 0.25 392.1 3.2% 14.9% 281.7 2.3% 10.8%
-.25 to +.18 0.18 78.5 0.9% 4.3% 0.18 313.6 2.6% 12.3% 227.8 1.8% 9.0%
-.18 to +.125 0.125 69.4 0.8% 3.5% 0.125 296.2 2.4% 9.9% 212.5 1.7% 7.3%
-.125 to +.09 0.09 42.7 0.5% 3.0% 0.09 185.3 1.5% 8.4% 133.7 1.1% 6.2%
-.09 to +.063 0.063 52.6 0.6% 2.4% 0.063 223.4 1.8% 6.6% 151.4 1.2% 5.0%
-.063 to +.045 0.045 24.3 0.3% 2.1% 0.045 135.8 1.1% 5.4% 97.6 0.8% 4.2%
-0.045 Pan 187.1 2.1% 0.0% Pan 666.6 5.4% 0.0% 520.0 4.2% 0.0%
Total 8770.8 100.0% 12239.5 100.0% 12432.9 100.0%
Calc. P80 [mm] 21.26 Calc. P80 [mm] 5.52 Calc. P80 [mm] 10.15
Calc. P50 [mm] 12.83 Calc. P50 [mm] 2.34 Calc. P50 [mm] 4.03
-6.35mm % 31.4% -6.35mm % 83.2% -6.35mm % 65.5%
-1.0mm % 10.8% -710um % 26.0% -710um % 19.1%

157
APPENDIX F: AG-CRUSHER-HPGR PLANT LAYOUT

158
159
APPENDIX G: SABC PLANT LAYOUT

160
161
 APPENDIX H: POWER CONSUMPTION COMPARISON

Unit Power Average Consumption Specific

Qt. Inst. Simu. Load Unit Total Energy
Description kW kW Factor kW kW kWh/t

AG-Cr-HPGR Option (2 lines)

4,000 tph per line (simulated instantaneous)
92% Circuit Overall Availability - 176,640 t/d
Comminution equipment
AG Mills - 10.4 m D x 6.1m EGL (34 x 20 ft) 2 8,000 7,411 1 6,818.1 13,636.2 1.853
Sec. Crushers XL-1100 (8 - 6 oper. 2 stdby) 6 820 807 1 742.4 4,454.6 0.605
Tert. HPGR - 2.4 D x 1.7 W m 4 5,000 3,665 1 3,371.8 13,487.2 1.833
Ball Mill - 7.9m D x 12.8m L ( 26 x 42 ft ) 4 18,000 15,563 1 14,318.0 57,271.8 7.782
96576 88,850 12.07

SABC Option (2 lines)

3,080 tph (simulated instantaneous rate)
94% Circuit Overall Availability - 138,970 t/d
Comminution equipment
SAG Mills - 12.2 m D x 6.7m EGL (40 x 22 ft) 2 25,000 23,279 1 21,882.1 43,764.1 7.558
Pebble Crusher XL-1100 4 820 361.4 1 339.7 1,358.9 0.235
Ball Mill - 7.9m D x 12.8m L ( 26 x 42 ft ) 4 18,000 12,636 1 11,877.8 47,511.4 8.205
98547 92,634 16.00

24.5%

AG-Cr-HPGR Option (2 lines)

Comminution equipment
AG Mills - 10.4 m D x 6.1m EGL (34 x 20 ft) 2 8,000 7,411 1 6,818.1 13,636.2 1.853
Sec. Crushers XL-1100 (8 - 6 oper. 2 stdby) 6 820 807 1 742.4 4,454.6 0.605
Tert. HPGR - 2.4 D x 1.7 W m 4 5,000 3,665 1 3,371.8 13,487.2 1.833
Ball Mill - 7.9m D x 12.8m L ( 26 x 42 ft ) 4 18,000 15,563 1 14,318.0 57,271.8 7.782
Sub-total 96576 88,850 12.07
Conveyors and feeders
AG Mill Feed Conveyors 2 857 857 0.75 591.5 1,183.1 0.161
AG Screen Discharge Conveyors 2 149 149 0.75 102.9 205.8 0.028
Sc Crushers Bin Feed Conveyor 2 298 298 0.75 205.8 411.6 0.056
Sec. Crusher Bin Feeder 2 56 56 0.75 38.6 77.2 0.010
Sec. Crusher Feeders No.1 to 8 (2 on standby) 6 56 56 0.75 38.6 231.5 0.031
Crusher prod. to HPGR bin - High Lift Conveyor 2 373 373 0.75 257.3 514.5 0.070
Crusher tramp material by-pass conveyor 2 56 56 0.20 10.3 20.6 0.003
HPGR Feeders No.1 to 4 4 112 112 0.75 77.2 308.7 0.042
HPGRs Discharge - High Lift Conveyor 2 373 373 0.75 257.3 514.5 0.070
Sub-total 5108 3,468 0.47
Pumps
Ball Mill Cyclone Feed Pumps 4 2,983 2983 0.77 2,113.0 8,452.1 1.148
Peb.Cr Area -Scrubber Effluent Return pump 4 37 37 0.8 27.4 109.8 0.015

162
 Sub-total 12080 8,562 1.16
Screens
SAG Discharge vibrating screens 4 30 30 0.75 20.6 82.3 0.011
Sub-total 119 82 0.01
Comminution Eq. - Lube and Cooling systems
Crusher Lube units 6 75 75 0.75 51.5 308.7 0.042
Crusher Lube cooling units 6 37 37 0.75 25.7 154.4 0.021
HPGR Lube units 4 15 15 0.75 10.4 41.4 0.006
HPGR Lube cooling units 4 30 30 0.75 20.7 82.8 0.011
Ball Mill Motor Cooling Systems 4 75 75 0.8 54.9 219.5 0.030
Ball Mill Cycloconv. Cooling Systems 4 75 75 0.8 54.9 219.5 0.030
Ball Mill Lube Cooling Systems 4 22 22 0.8 16.5 65.9 0.009
Sub-total 328 1,092 0.15
Dust and metal collection requirements
Peb.Cr Area -Scrubber Fan 4 89 89 0.80 65.9 263.4 0.036
AG Screen Discharge Conveyor Magnet 2 30 30 0.75 20.6 41.2 0.006
Crusher system self cleaning Magnet 2 30 30 0.75 20.6 41.2 0.006
HPGR system self cleaning Magnet 2 30 30 0.75 20.6 41.2 0.006
Metal detectors 4 10 10 0.75 6.9 27.6 0.004
Ball Mill Trunion Magnets 4 19 19 0.75 12.9 51.5 0.007
Sub-total 208 466 0.06

Heat & Ventilation Systems 1 325 325 1 299.4 299.4 0.041

Sub-total 299 0.04
TOTAL 102,819 13.97

SABC Option (2 lines)

Comminution equipment
SAG Mills - 12.2 m D x 6.7m EGL (40 x 22 ft) 2 25,000 23,279 1 21,882.1 43,764.1 7.558
Pebble Crusher XL-1100 4 820 361.4 1 339.7 1,358.9 0.235
Ball Mill - 7.9m D x 12.8m L ( 26 x 42 ft ) 4 18,000 12,636 1 11,877.8 47,511.4 8.205
98547 92,634 16.00
Conveyors and feeders
SAG Mill Feed Conveyors 2 746 746 0.75 525.7 1,051.4 0.182
SAG Screen Discharge Conveyor 1 261 261 0.75 184.0 184.0 0.032
Pebble Bin Feed Conveyor No. 1 1 186 186 0.80 140.2 140.2 0.024
Pebble Crusher Bypass Conveyor 1 112 112 0.75 78.9 78.9 0.014
Pebble Bin Feed Conveyor No. 2 1 19 19 0.75 13.1 13.1 0.002
Pebble Crusher Discharge Conveyor No. 1 1 112 112 0.75 78.9 78.9 0.014
Pebble Crusher Discharge Conveyor No. 2 1 149 149 0.75 105.1 105.1 0.018
Pebble Crusher Feeders No.1 to 4 4 56 56 0.75 39.4 157.7 0.027
Sub-total 2554 1,809 0.31
Pumps
Ball Mill Cyclone Feed Pumps 4 2,983 2983 0.77 2,159.0 8,635.8 1.491
Peb.Cr Area -Scrubber Effluent Return pump 1 37 37 0.8 28.0 28.0 0.005
Sub-total 11968 8,664 1.50
Screens
SAG Discharge vibrating screens 2 30 30 0.75 21.0 42.1 0.007
Sub-total 60 42 0.01
Comminution Eq. - Lube and Cooling systems

163
Crusher Lube units 4 75 75 0.75 52.6 210.3 0.036
Crusher Lube cooling units 4 37 37 0.75 26.3 105.1 0.018
SAG Lube Cooling Systems 2 22 22 0.8 16.8 33.6 0.006
SAG Cycloconv. Cooling Systems 2 75 75 0.8 56.1 112.2 0.019
SAG Motor Cooling Systems 2 75 75 0.8 56.1 112.2 0.019
Ball Mill Motor Cooling Systems 4 75 75 0.8 56.1 224.3 0.039
Ball Mill Cycloconv. Cooling Systems 4 75 75 0.8 56.1 224.3 0.039
Ball Mill Lube Cooling Systems 4 22 22 0.8 16.8 67.3 0.012
Sub-total 1476 1,089 0.19
Dust and metal collection requirements
Peb.Cr Area -Scrubber Fan 1 89 89 0.80 67.3 67.3 0.012
SAG Screen Discharge Conveyor Magnet 1 30 30 0.75 21.0 21.0 0.004
Metal detectors 1 10 10 0.75 7.1 7.1 0.001
Ball Mill Trunion Magnets 4 19 19 0.75 13.1 52.6 0.009
Sub-total 204 148 0.03

Heat & Ventilation Systems 1 274 274 1 257.9 257.9 0.045

Sub-total 274 258 0.04
TOTAL 104,645 18.07

Total Com. Circ. Specific Energy Difference 22.7%

164