Conference PaperPDF Available

Optimisation and Continuous Improvement of Antamina Comminution Circuit

Edward Rybinski1, Jorge Ghersi1, Frank Davila1, Javier Linares1
Walter Valery2, Alex Jankovic2, Roberto Valle2, Serkan Dikmen2
1Compañía Minera Antamina S.A
Derby 055, torre 1, piso 8, oficina 801, Santiago de Surco
Lima 33, Perú
2Metso Process Technology and Innovation
1/8-10 Chapman Place
Eagle Farm, Brisbane, QLD, Australia
Compañía Minera Antamina is a polymetallic mining complex that produces copper and zinc
concentrates as primary products and molybdenum and lead/bismuth/silver concentrates as by-products. In
2007, Metso Process Technology and Innovation (PTI) was contracted to increase SAG mill throughput by
integrating and optimising blast fragmentation, crushing and grinding for the harder M4/M4A (CuZn) ores.
Following the implementation of recommendations provided by Metso PTI, the SAG mill throughput
increased from 2750tph to reach an average of 3600tph for the remainder of 2007. Due to continued
application of mine to mill integration and optimisation philosophy, by 2010, the concentrator consistently
exceeded the target of 4400tph while processing harder CuZn ores.
Antamina, SAG, blasting, crushing, grinding, optimisation, simulation, comminution
Antamina, a polymetallic mining complex that produces copper and zinc concentrates as primary
products and molybdenum and lead/bismuth/silver concentrates as by-products, represents one of the main
complexes of the Peruvian mining industry. It is situated in the central Peruvian Andes around 4,300
meters above sea level.
The project, which commenced operation in 2001, had worked on initiatives which achieved the
design capacity of 70,000tpd after five months of operation with the copper ore (M1/M2). There were
several issues associated with the copper/zinc ores (M4/M4A), which limited the throughput for these ores
types to just half of the M1/M2 throughputs (i.e., 2,000-2,500tph vs. 4,000-5,000tph respectively).
Achieving similar throughputs for both ore types became a challenge for Antamina, and in 2007 Antamina
contracted Metso Process Technology & Innovation (PTI) for the implementation of a Mine to Mill project.
Ore classification at Antamina considers six principal ore types which are processed based on
grinding and flotation requirements. One of these ore types includes sub categories, M4 and M4A, which
are blended together prior to processing in order to get the most stable grades for the downstream
separation. This paper describes the work carried out by Antamina and Metso PTI with the objective to
increase the throughput exclusively for the copper/zinc ore types (M4/M4A).
Metso PTI started the project in early 2007 and continued until the end of 2009. Work conducted
during this period included: comprehensive review of existing operations at the mine and in the
comminution circuits, characterisation of ore sources into domains of similar blasting/fragmentation
properties, measurement of blast design implementation and resulting fragmentation, measurement of
vibration and definition of main constraints, monitoring of blasted material movements using passive
radio-frequency transponders or SmartTagTM system, review of current operating practices of the primary
crusher, the SAG and Ball mills grinding circuits, a comprehensive analysis of historical data, modeling of
current drill and blasting, crushing and milling operations, and mine to mill simulations of optimum
operating strategies which were finally put into practice in combination with other initiatives led by
Antamina personnel.
Metso PTI methodology is based on mechanistic and forward analysis that involves the following
a) Characterisation and delineation of blast domains based on rock structure and strength. The
measurement techniques that have been developed over many years present results that represent the
conditions in which the mines operate in a production environment. The techniques are practical and
statistically representative requiring only a few other tests to validate the results and confirm the
correlation (e.g., Drop Weight and SMC tests). They are inexpensive and can be easily carried out
by mine personnel.
b) Establish process constraints such as wall stability, damage and control in the pit and in the
tailings dam, presence of water, ore dilution, muckpile characteristics, size of mining equipment,
size and installed power of crushing and milling equipment and other process bottlenecks.
c) Definition of the key downstream requirements and development of drilling/blasting strategies for
each domain to suit the defined downstream requirements (milling of ore or waste).
d) Use of proven software tools, predictive models and simulations in conjunction with the
experience and collaboration of mine and plant personnel to establish optimum operating and
control strategies to maximise overall profit from the blast to the mill.
e) Implementation and monitoring of the defined integrated operating strategies (suitable blast
designs for each ore domain followed by respective optimal crushing and grinding strategies) and
establishment of standards, quality assurance and control mechanisms.
f) Analysis and management of comprehensive data and results.
g) Long term implementation and maintenance of benefits.
The steps above are focused on the development of an integrated mining and milling optimisation
strategy to increase profitability of the operation. Part of this strategy is to increase plant throughput with
little or no capital investment. By optimising rock breakage and fragmentation from blasting through
crushing and grinding the following benefits could be achieved:
Increased excavation and loading efficiencies;
Reduced ROM topsize allowing the primary crusher to achieve greater throughput while operating
at a smaller gap to produce mill feed with a finer topsize;
Reduced SAG mill feed topsize and increased proportion of fines (-10mm material) to increase
mill throughput with existing installed power;
Minimal adverse impacts such as dilution or ore loss, structural damage or environmental
This paper summarises the work performed by Antamina and Metso PTI and includes
recommendations of changes in blasting, crushing and grinding operation and also control in order to
improve mill performance during the treatment of M4/M4A ores.
Metso PTI Process Integration & Optimisation methodology involves rock characterisation in
terms of its strength and structure. One of the objectives of rock characterisation is to determine domains
within the orebody that will generate similar blast fragmentation. By understanding what domains exist, the
mine can establish standard blast patterns for each of the different domains that will generate similar
fragmentation distributions. This may require either an increase in blasting energy (e.g., powder factor or
explosive weight per tonne of rock) or perhaps a decrease in energy depending on the rock properties.
Measurement of Rock Strength & Structure
Fracture Frequency (FF) and Rock Quality Designation (RQD) give a good indication of the rock
mass structure, which in turn will drive the proportion of coarse material in the blast fragmentation. Fines
generation in the blast is mainly related to rock strength as well as explosive/rock interaction.
The Point Load test provides an Is50 index for rock strength classification that is well known and
widely used in the industry, mainly by geotechnical personnel. It is commonly used as a quick and simple
method to predict Unconfined Compressive Strength (UCS).
Antamina categorises their ore into material classifications or ore types: copper-only (M1, M2,
and M2A), copper-zinc (M3, M4, M4A) and bornite (M5, M6) depending on the Cu, Zn and Bi assays.
However, this ore type classification is based on assays alone and does not necessarily reflect any distinct
rock type or lithological variation.
Much higher mill tonnages experienced for M1 compared with M4/M4A material are due to
differences in the SAG mill feed size distribution. Figure 1 shows a range of M4A SAG feed size
distributions compared with an M1 feed size distribution.
Figure 1 Comparison of SAG mill feed size & throughputs
It is clear that the M1 material fragments very finely in the blast producing about 70% minus
10mm and an 80% passing size (P80) of 20 to 25mm after crushing, and is processed at 5,200tph. In
contrast, the M4A material blasts much coarser after crushing it has a P80 of 50 to 150mm and
represents a typical SAG mill feed. This material is processed between 2,350 and 2,850tph.
110 100 1,000
Particle Size (mm)
Cumulative Weight % Passing
5,200 tph
2,850 tph
2,600 tph
2,350 tph
Rock Strength Point Load Index
Based on the Point Load Index (PLI) Is50 value, the Unconfined Compressive Strength (UCS)
was estimated. The results, presented in Figure 2, show that all ore types and survey samples exhibited a
wide range of rock strengths from below 50MPa to more than 250MPa. The wide range of strengths
indicates that each ore type is not consistent in rock properties; most likely due to a variety of rock types
and structures.
Figure 2 Summary of estimated UCS from Point Load Index measurements (Jan 2007)
Metallurgical Hardness Testing
A summary of the Bond Ball Mill Work Index (BWi) results since 2002 is shown in Table 1.
These results appear to be quite consistent with M1 reporting a BWi of around 10kWh/t (relatively soft)
and M4/M4A between 12 and 14kWh/t (moderate). It appears that the variability in hardness between ore
types is lessened when the material is below 3mm and suitable to be processed by ball mills.
From a SAG mill‟s perspective, the UCS and BWi results indicate a wide variability in rock
hardness with both M1 and M4A ore types exhibiting soft and hard behaviour. However, once reduced to
ball mill feed size; the materials have similar properties with M1 exhibiting a slightly lower energy
requirement compared with M4A.
With a constant blast pattern used for all these ore types, it would be expected that the variable
rock properties would respond differently and result in a finer and coarser fragmentation which finally
results in a range of mill throughputs.
<25 <50 <75 <100 <125 <150 <175 <200 <225 <250 <275 <300
No of Obs ervat ions
M4A Survey 1
M4A Survey 2
Overall, the M1 material fragments very well whereas M4/M4A ore types do not fragment as well
and result in lower mill production. However, the variable production figures indicate that fragmentation is
never consistent a consequence of not adjusting the blast pattern to suit the rock conditions. By
identifying the rock strength and structure associated with each rock type, and adjusting the blast design
accordingly, it is expected that the variability would be reduced and making it easier to predict.
Table 1 Summary of Bond Ball Work Index by ore type
Blasting Domain Definitions
It was recommended that Antamina initiate a drill core-testing program measuring both rock
strength and structure based on PLI and RQD. In core sections of known ore types, samples of different
rock types should be logged for RQD and then tested for PLI. Very quickly, Antamina could generate an
updated database of rock strength and structure for rock types associated with M1 and M4A (the two main
ore types requiring domain definition). Table 2 lists possible rock types to be tested for M1 and M4A.
Table 2 Rock Types to be characterised (Based on 2007 mill feed)
Ore Type
Rock Type
Café Endoskarn
Rosa Endoskarn
Café Verde Exoskarn
Verde Exoskarn
Café Verde Exoskarn
Café Endoskarn
When designing a blast pattern, it was recommended to estimate the amount of Endoskarn and
Exoskarn in the blast and set the burden and spacing according to a tighter pattern for Exoskarn and wider
pattern for Endoskarn. Metso PTI recommended that Antamina design the blast pattern for the most
difficult rock conditions encountered i.e., presence of M4A material, Exoskarn rock type or high
Blastability Index whatever the measurement used. Rather than design the blast for the average rock
conditions or for the softest material, blasts should be designed to adequately fragment the most
challenging material. This will ensure a minimum quality of fragmentation (and mill throughput) is
Ore type
Feb 2002
Aug 2002
Oct 2002
Feb 2006
M1 intrusive
Feb 2006
Jan 2002
Jan 2003
Jan 2002
Nov 2002
Feb 2003
Mar 2002
Feb 2003
May 2003
Feb 2006
Feb 2006
Metso PTI carried out an audit of Blast 3-SP-4283-08 detonated on January 10th 2007. The blast
polygon was a mix of ore (domains M4, M4A, M1 and M2) and waste. The copper zinc ore (M4, M4A)
from this blast was monitored very closely with passive radio frequency transponders (SmartTags™)
placed both in the stemming of some of the blast holes and on the surface of the muckpile after the blast to
track this particular ore “parcel” from blast to the mill. The SmartTags™ were detected using an antenna
placed over the primary crusher product conveyor and another over the SAG mill feed belt. The following
data was collected as a part of the blast audit:
Blasthole locations (design & actual)
Explosive loading
Measurement of velocity of detonation (VOD) and far-field vibration
Fragmentation images for size analysis
Samples for Point Load Testing
As a part of the audit carried out on Blast 3-SP-4283-08, Metso PTI reviewed „as designed‟ and
„as drilled‟ drill hole parameters to determine accuracy of the design implementation. The parameters
included burden, spacing, blasthole length and stemming length. Figure 3 overlaps the location of the
designed versus the „as drilled‟ hole patterns, which generally matched very well.
Figure 3 Design and actual drill hole locations
273850 273900 273950 274000 274050 274100 274150
As designed
As Drilled
Estimates of ROM or blast fragmentation distributions were made using image analysis software.
Digital photographs were taken of muckpile, shovel working face (digline) and haul truck beds while
dumping at the primary crusher. An example of the image delineation used to determine the particle size
distribution is presented in Figure 4.
Figure 4 of Truck bed image and resulting particle delineation
Blast fragmentation modelling has been used to identify the key parameters affecting
fragmentation. Each of these parameters has a natural variation (e.g., rock strength) and error (e.g., hole
position) associated with it. Therefore, a stochastic approach is taken in modelling blast fragmentation and
these variables are input with a mean and standard deviation. This modelling approach uses Monte Carlo
sampling followed by model simulation giving envelopes of ROM size distributions.
Figure 5 shows the impact of variations in blast design parameters (explosive, rock, drill pattern,
stemming length, and subdrill) on the fragmentation of Blast 3-SP-4283-08. The envelope of ROM ore size
distributions is shown by the lower 95% confidence level, the upper 95% confidence level and the mean.
Based on the variation in design parameters measured in the audited blast, it is expected that significant
variations in blast fragmentation are common at Antamina. For these simulations, the 80% passing size
varied from 157 to 476mm.
Figure 5 Stochastic fragmentation modelling
110 100 1000 10000
Cumulative % passing
Size (mm)
lower -95% confidence
upper +95% confidence
Survey 1
Survey 2
The fragmentation model was calibrated for a range of ore types (i.e., M1, M2 and M4). It
appeared that fragmentation in the M4 domain is much coarser than M1 and M2. Based on the SAG mill
feed surveys conducted, the variations in cumulative percent passing at 25mm size in M1 and M4/M4A are
70 to 80% and 35 to 60%, respectively. Figure 6 shows typical ROM fragmentation curves for M1, M2 and
M4/M4A ores.
Figure 6 Blast fragmentation modelling of different ore types
The blast fragmentation model was used to predict the ROM fragmentation size. Two additional
blast designs were reviewed for the M4/M4A domain using a new powder factor design. The fragmentation
model estimates a reduction in ROM P80 from 401 to 129mm. The results showed that the blast design
with the smaller blasthole diameter generates the finest fragmentation. The explosive energy distribution
was better with this design. Thus, Metso PTI recommended a blast design with 251mm drill holes.
In any PIO project, it is crucial that the material being fed to the mill during a plant survey is the
same material that has been characterised in the mine. Metso PTI has developed an ore block tracking
system named SmartTag™ that allows parcels of ore to be tracked, from the mine, through the crusher and
finally into the grinding mills. The SmartTags™ are built around robust passive radio frequency
transponders. Not having an internal power source, they can remain in stockpiles and ROM pads for
extended periods of time.
Figure 7 shows a typical SmartTag™ installation. The tag passes under the antenna along with the
ore where it is energised by an electromagnetic field. The tag then transmits its unique ID back to the
antenna where it is stored along with a time stamp in the recording device or Tag Reader. For permanent
installations, a wireless communications link or direct network connection can be used to transmit the tag
information back to a central database.
Tags can be placed in the stemming column of blastholes and a high percentage will survive the
detonation. Alternately, they can be placed on the muckpile surface post-detonation. The tags are small
enough to pass through a primary crusher and be detected on the product conveyor belt. They can then be
detected on the SAG mill feed belt before being destroyed in the mill. SmartTags™ offer an inexpensive
and versatile tool for material tracking and can provide information on blast movement during detonation,
dilution or ore loss, stockpile inventories and blending practices.
110 100 1000 10000
Cumulative % passing
Size (mm)
Figure 7 Elements of SmartTagTM installation
By locating antennas at critical points in the process flow ahead of the milling circuit,
SmartTags™ can be detected a number of times and provide valuable information on material movements.
In particular, they provide a means to link the spatial information associated with the mine to the time-
based or temporal information of the concentrator. Two antennas were installed at Antamina one on the
primary crusher product belt and another on the SAG mill feed belt as shown in Figure 8a and b.
(a) Primary crusher product.
(b) SAG mill feed
Figure 8 SmartTag TM temporary installation
For audited blast 3-SP-4283-08 discussed previously, 101 SmartTags™ were placed in the stemming of the
blast holes, with a further 48 on the muckpile surface post-detonation.
Figure 9 illustrates the tag distribution in each case.
(a) stemming columns of blast holes
(b) muckpile surface post detonation
Figure 9 SmartTag TM locations
Figure 9a shows that all the blastholes that were tagged were in the southwest portion of the blast.
The southeast corner was known to be in waste and was not tagged. The x,y,z locations of all tags placed
on the muckpile surface post-detonation were picked up by a surveyor. The audited blast was quite
confined with the material throw inwards. As a result the muckpile heave was measured to be a few metres
as shown in Figure 9b.
The SmartTagTM detection times are presented at the bottom of Figure 10, with blue icons
representing tags that passed through quickly (<2hrs) and red icons representing tags that passed through
slowly (>32hrs). In the first cluster of blue icons, it appears that the coarse ore pile height was low for an
extended period. This also corresponded with an increase in the SAG feed size F80 (mm) as measured by
the on-line image analysis system and a decrease in mill tonnage. January 16th recorded periods where the
tag residence time was long. It is hypothesised that the tags that had detection times greater than 32 hours
were placed on the pile at high stockpile levels and were consequently carried to the outer perimeter
increasing residence time.
Knowledge of shovel locations over time combined with the SmartTag™ detection times
enables material movements during a blast to be estimated. From these movements, the potential for ore
loss and dilution can be quantified. This is particularly important at Antamina where the polygons of
different ore types and waste can be very small perhaps only tens of metres squared. Using the tag
detection times at the primary crusher, the dispatch records were used to determine which truck was
dumping in addition to the shovel location when that truck was filled.
The audited blast was quite confined where the material was thrown internally and this was
reflected in the resulting tag movement vectors. The length of the vector also reflected the distance the tags
moved during the blast (and possibly during clean-up and excavation). This movement averaged 25m
indicating that some of the material moved in excess of 25m. While some of this may be due to the tags
rolling down the free face as they are exposed during excavation, it is estimated that the tags experienced at
least 15m of movement in the direction of the free face. Based on the SmartTag™ results from the audited
blast, there was evidence of significant dilution and ore loss.
Figure 10 SmartTag™ detection times with coarse ore stockpile conditions
Operation of the primary crusher is very important in optimising mill feed size for maximum
throughput. While it has already been stated that the majority of the fines in the feed are generated in the
„crushing‟ zone during the blast, the top size of the mill feed is dictated by the crusher setting and operation.
Maximising the average power draw of a primary crusher will result in a finer product. In addition, choke
feeding of a primary crusher results in multiple breakage events as the material passes down the chamber
rather than only a few events for a non-choked crusher. The ROM topsize and crusher gap have the
following interdependence, which will affect “Mine-to-Mill” optimisation:
Crusher throughput for a given CSS is largely controlled by the largest rocks in the ROM.
Reducing the top size of ROM ore allows the crusher to operate at the same throughput (and same
power draw) with a smaller CSS.
Finer ROM allows the CSS to be reduced and results in the production of a finer crusher product.
Finer ROM and smaller CSS allows better choke feeding in the crusher, which produces finer feed
to the SAG mills and also increases the life of the crusher mantle and concaves.
For a given crusher CSS, variations in the crusher product P80 will result from differences in ore
hardness as well as variations in the blast fragmentation.
12-Jan 12:00 13-Jan 12:00 14-Jan 12:00 15-Jan 12:00 16-Jan 12:00 17-Jan 12:00 18-Jan 12:00
Fe ed Size (mm) , Pile Height (%)
Tonna ge (tph)
F80 (mm)
COS Pile
SAG Tonnage
Detected Tags
Blue: <2hr
residence time
Red: >32hr
residence time
Survey 1
Survey 2
Full Stockpile
Primary Crusher Product Sample
In order to evaluate the performance of the primary crusher, a sample of the product material was
collected from the conveyor belt on January 15th following the Metso PTI methodology. At the same time,
trucks dumping into the crusher were photographed for image analysis of the crusher feed material. These
data were used to develop a primary crusher model for the simulation study investigating the impact of
ROM ore size and crusher setting on mill throughput. The ROM and crusher product size distributions are
provided in Figure 11.
Figure 11 ROM and crusher product size distribution
Using a tighter setting did not produce a higher power draw and therefore there was more capacity
available in the primary crusher to operate at a tighter setting (4.5 to 5 inches) without negatively affecting
the throughput.
In January 2007, two complete grinding circuit surveys were performed followed by a SAG mill
crash stop and grind-out. After stopping the mill, a SAG mill feed sample was also collected from
Conveyor 1. Figure 12 shows the circuit and sampling points of the grinding circuit.
Figure 12 Flow sheet and sampling points of the milling circuit
Due to lower throughputs when processing M4/M4A ores, only two of the three ball mills are
operated when treating these ores. For instance, during the first survey, ball mill 2 was not operating and
during the second survey, ball mill 1 was not operating. The summary of the surveys conditions are
presented in Table 3.
Table 3 Summary of SAG mill survey conditions
Variable Survey 1
Survey 2
Power, MW 14.8
Bearing Pressure, kPa 5347
F80, mm 89.7
Throughput, tph 2366
Total charge, % 23
Mass Balancing
Mass balancing showed that the survey data were of good quality and consistent. Table 4
summarises the key stream tonnages, percent solids and P80 values for both surveys.
The cyclone feed density was lower for Survey 1 compared to Survey 2 (60% vs. 65%). All the
ball mill circuits had relatively low circulating loads of approximately 230%.
Table 4 Grinding circuit survey balanced results for Survey 1 and Survey 2
Mass Balance results Survey 1
Mass Balance results Survey 2
Solids (tph)
% solids
% solids
Fresh Feed
SAG discharge
trommel oversize
BM 1 product
BM 3 product
cyclone 1 feed
cyclone 1 o/f
cyclone 1 u/f
cyclone 3 feed
cyclone 3 o/f
cyclone 3 u/f
SAG disch. BM1
SAG disch. BM3
Based on the mass balanced results, the cyclone efficiency curves were plotted. It was observed in
Figure 13a that the sharpness of classification for Survey 1 was relatively high and fraction of fines by-
passing to the underflow was relatively low (<20%). This can be explained by a favourable cyclone feed
density of around 60%.
Figure 13b shows the cyclone efficiency curves for Survey 2. The cyclone feed density was higher,
therefore, the cyclone efficiency was slightly lower and the amount of fines by-passing to the underflow
had increased to at least 25%.
(a) Survey 1
(b) Survey 2
Figure 13 Cyclone efficiency curves
Model Fitting
Following data balancing, mathematical models of all the process units were calibrated in
JKSimMet. The purpose was to model the grinding circuit and primary crusher as they were operating at
the time of the surveys.
Feed Size Effect on SAG Mills
SAG mill feed is supplied from the primary crusher, and the product from the primary crusher is
influenced by the size distribution achieved from blasting. Significant effort has been spent at a number of
operations to relate SAG mill throughput to SAG mill feed size. Particularly for the Antamina case, a
correlation was obtained, derived from measurements and data analysis, indicating that finer M4/M4A
topsize and F80 mill feed gives higher SAG mill throughput as presented in Figure 14.
Figure 14 Antamina SAG feed size F80 vs. throughput
Reduction at the coarse end of the feed size distribution to the SAG mill is only part of the
optimisation effort. In operations such as Antamina, increasing the amount of fines (-10mm) should
increase the SAG mill throughput proportionally. Such material is smaller than the mill discharge grates
and trommel screen apertures and can be considered as „free grind‟.
55 60 65 70 75 80
Feed Size F80 (mm)
SAG Tonnage (tph)
Ball Mill Media Size
Photographs along the mill axis of the three mills were taken for image analysis to determine the
size distribution of grinding media within the mills. Antamina was trialling different ball make-up sizes in
each mill: a different combination of 2½ and 3 inch balls are added to the three ball mills.
Figure 15a shows the ball size distribution results from the photos of the surface of the ball charge
in the three ball mills. Less than 5% of balls were found to be smaller than 50mm.
(a) In three ball mills
(b) Along ball mill 1
Figure 15 Ball Size Distribution
The analysis of the ball size distribution along the mill axis was also carried out to determine if
larger balls tend to concentrate at the feed end and smaller balls towards the mill discharge. This effect
may be observed in Figure 15b for ball mill 1 where the proportion of large balls at the mill discharge is
lower than near the feed or middle of the mill. For the other two ball mills this effect was not observed and
therefore segregation along the mill axis could not be proven definitively.
Models for blasting, crushing and grinding operations were developed based on data collected on-
site and results of the samplings during trials. These models can be combined to simulate the overall
“Mine-to-Mill” Process Integration & Optimisation. Inputs to these models include rock properties (UCS,
RQD), blast design parameters (burden, spacing, stemming column, explosive type, etc.), primary crusher
setting, SAG mill operating conditions (ball size, ball charge, total charge, grate size, mill speed, etc.) and
ball mill circuit conditions (ball size, ball charge, mill speed, cyclone geometry, pressure, etc.)
These simulations quantify the possible throughput increases. When changes indicate higher
tonnage, it is quite likely that combining two or more of these conditions will give an even higher increase
in plant throughput. Conditions such as different blast designs, primary crusher setting, SAG mill grate
design (worn vs. new), SAG mill ball charge level were considered in the simulation study.
Total Distribution (%)
+0 -40 +40 -45 +45 -50 +50 -55 +55 -60 + 60 -65 +65 -70 +70
Ball Dia meter Range (mm)
Ball Weight Percentage (%)
BM#1 (%)
+0 -40 +40 -45 +45 -50 +50 -55 +55 -60 +60 -65 +65 -70 +70
Ball Dia meter Range (mm)
Ball Weight Percentage (%)
Near Feed
Near Discharge
ROM Size Distribution Simulations
The results of the simulations of the different ROM size distribution are presented in Table 5. The
Base Case followed the conditions of the audited blast while the D2 blast design and a 5 inch CSS on the
primary crusher resulted in a 12% increase in tonnage which translates to 2,750tph of the same material.
The effect of blast fragmentation variability was demonstrated as the coarser end of the 95% envelope of
ROM size reduced mill throughput by 10% and the finer end increased throughput by 22% or 3,000tph.
Table 5 Simulations of different ROM size distribution
Operating Condition
Run of Mine P80 (mm)
Primary Crusher CSS (inch)
SAG Feed Rate (tph)
SAG Feed F80 (mm)
SAG Feed % -10mm
Scats Rate (tph)
No. of Cyclones
Cyclone O/F % solids
Cyclone O/F P80 (μm)
Tonnage Increase (%)
SAG Mill Operating Condition Simulations
Simulations were carried out to investigate the impact of changing grate open area and SAG mill
ball charge level. The simulations indicated that as the grates wear the scats rate increases marginally
and the mill tonnage increases by 4%. Due to additional ball mill capacity being available under low
tonnage conditions (i.e., M4A material), the option of increasing the trommel screen aperture and sending
coarser material to the ball mills should be considered. Due to the fineness of the M1 (high tonnage) SAG
mill feed, this should have little impact on the ball mill circuits.
At the time of the grinding circuit survey, the SAG mill was operating with a 13.5% volumetric
ball charge. A simulation of 15% ball charge level was done to estimate the impact on mill throughput.
This result indicated that mill throughput could increase by 5% as a result of the greater ball load. While
changes in feed size and material hardness may reduce the benefit of an increased ball charge, it was
recommended that Antamina trial a 15% ball load during M4/M4A campaigns.
Implementation of the recommendations made by Metso PTI during the 2007 study resulted in an
increase in SAG mill throughput from 2750tph up to an average of 3600tph. Following this, Compañía
Minera Antamina and Metso PTI entered into a technical services support contract for 2008-09, to
implement continuous improvement projects to further improve the mine and mill efficiencies. As a part
of this contract two projects were conducted in 2008: 1) Mine to Mill Blast Fragmentation and 2) Blast
Damage and Wall Control.
The main objective of the “Mine to Mill Blast Fragmentation” project was to improve blast
fragmentation in M4/M4A ore blasts in order to increase the SAG mill throughput even further. Three
high intensity blasts were conducted in M4/M4A ore and the ore was campaigned through the plant to
assess the SAG mill performance. These high intensity blasts produced finer ROM fragmentation and
feed to the SAG mill. The SAG feed F80 was reduced to around 50-60mm compared to 100-120mm
previously and fines (-10mm) percentage was increased from 25% to 45%. Primary crusher power peaks
were reduced significantly during these campaigns. SAG mill throughput averaged 4100tph for the three
blasts and throughput above 4500tph was achieved for several hours during these trials.
A full plant survey was conducted during the third trial in December 2008, and used to update
site specific blasting, crushing and grinding models as pebble crushers were installed in the meantime.
Modelling and simulations indicated that modified blast designs and proper utilisation of pebble crushers
can increase the SAG mill throughput up to 4500tph for the softer ore and 4250tph for the harder ore.
Simulation results shown in Figure 16 indicate that blast design changes increased the mill throughput by
25% (14% from the first design change and 11% from the second design change). Simulations also show
that pebble crushers with larger grates can potentially increase the throughput by 11% and the increase in
ball load would contribute about 4%. M4/M4a ores have become softer compared to the 2006 base case.
Where the ore is very soft, ore hardness has contributed to 11% increase whereas for relatively harder ore,
its contribution is only 4%.
Figure 16 Simulated effect of blast and plant changes on SAG mill throughput
Based on the results achieved in the project, Metso PTI concluded that Antamina can consistently
achieve SAG mill throughput above 4000tph in M4/M4a ores by implementing the mine to mill process
integration and optimisation philosophy.
Compañía Minera Antamina engaged Metso Process Technology and Innovation (PTI) to provide
technical assistance for implementation of continuous improvement projects. The main objective of the
support contract was to increase the SAG mill throughput by improving blast fragmentation and
comminution for the harder M4/M4A (CuZn) ores. This project started in 2007 and proceeded through a
continuous improvement contract to completion in 2009. The project consisted of a complete review and
optimisation of existing operations, including:
Characterisation of ore sources into domains of similar blasting/fragmentation properties;
Measurement of blast design implementation and resulting fragmentation; measurement of
vibration and definition of main constraints;
Monitoring of blasted material movements using passive radio-frequency transponders or
SmartTagTM system;
Review of current operating practices of the primary crusher and grinding circuits (SAG and Ball
Modeling of current drill and blasting, crushing and milling operations;
Simulations of optimum operating strategies from the mine to mill;
Implementation of recommendations and operating strategies.
The production budget in 2007 was 2750tph for the harder CuZn ores. After the implementation
of changes recommended by Metso PTI during the first phase of the project the production increased to
3600tph on average for the remainder of 2007. By continued application of the mine to mill integration
and optimisation philosophy, the concentrator exceeded its target of 4400tph for harder CuZn ores by 2010.
Metso PTI thanks Compañía Minera Antamina for permission to publish this paper and acknowledge
contribution of all company employees involved in this project.
Dechert, C., 2006, Antamina “Design through to Operation”, SAG Conference 2006.
Valery, W., Sedat, E., Colacioppo, J., La Rosa, D., Jankovic, A., Metso Process Technology & Innovation
“Mine to Mill Process Integration & Optimisation at Compañía Minera Antamina”, 2007.
Valery, W., Valle, R., Corsini, S., Kanchibotla, S., Colacioppo, J., Dikmen, S., Baguley, P. Metso
Process Technology & Innovation “Mine to Mill Blast Fragmentation Project at Compañía Minera
Antamina”, 2009.
... Because of this, continuous improvement processes are sought within mining operations under a holistic approach [2]. Fragmentation of blasting rocks has been studied because it can add great value by reducing costs, as it has a high impact on the productive and economic performance of loading, carrying (50-60% mine costs), crushing, and milling [3]. Several studies have been conducted to measure the relationship of these parameters with P80 (an indicator of the degree of fragmentation). ...
... Therefore, the area of each drill is represented in (2). The influence area of the explosive per drill is represented in (3). The coverage percentage of the explosive per drill is I1=90.69% ...
Full-text available
Blasting is one of the most important stages in the productive process of a mine due to its direct impact on rock fragmentation, which determines the degree of productivity of operations and the extraction costs generated. In this scenario, an optimized methodology is presented for designing blasting meshes by using mathematical models that help calculate the geometric parameters of a blasting mesh, such as burden, considering the variables of the rock mass and the type of explosive to measure its impact on rock fragmentation and loading productivity (tons/hour). The main advantage of this method is the reliability of the design, which takes into account a greater number of variables that influence fragmentation and uses the principle of distribution and amount of energy in an optimal way. The results obtained in the case of application show that a change in design (2.7 x 2.7 square mesh to 2.2 x 2.5 triangular mesh) reduces P80 by 65%, from 17 to 6 inches, approximately. Additionally, the results show that greater operational efficiency was achieved by increasing excavator productivity by approximately 15.6%.
... Several researchers introduced the concept of Mine to Mill by applying it at several open pit mines (Kanchibotla et al., 1998;Valery et al., 2001;Workman and Eloranta, 2003;Adams, 2004;Dance et al., 2006;Rybinski et al., 2011;Hakami et al., 2015;Kanchibotla et al., 2015). The researchers used established blast fragmentation models and comminution models to understand the impact of the blast design and ore characteristics on crushing and grinding circuit performance. ...
In hard rock mining, blasting is generally the first operation of rock fragmentation from its in-situ state. Therefore, blast design and geotechnical ore properties have a significant impact on downstream processes such as crushing and grinding. This work aims to understand this impact at an open pit case study by applying the simulations of the Mine to Mill methodology. JKMRC Crushed Zone blast fragmentation model (CZM) estimates the Run of Mine (ROM) fragmentation using rock properties (such as density, hardness and in-situ fragmentation) and blast design characteristics. This ROM fragmentation feeds the JKSimMet crushing and grinding models to predict mill throughput and the product size distribution. This study demonstrates the relevance of rock hardness and powder factor on mill throughput and the size specific energy (SSE). The simulation results have shown that the Uniaxial Compressive Strength (UCS) has an inverse relationship with throughput and a direct relationship with SSE. The results also indicate that higher powder factor increases the mill throughput and coarsen mill grind size (P80). Since coarser grind size can have a negative effect on recovery, there is a trade-off between throughput and recovery. Higher powder factor also decreases the SSE of the whole circuit. The rock fragmentation parameter, Rock Quality Designation (RQD), mainly impacts the coarser end of ROM size distribution; as the primary crusher breaks those coarser particles, no significant RQD effect on the plant throughput and the Size Specific Energy was observed.
... The next level of optimisation is to control the stockpile flow so that the mill feed maintains an even and high grindability or pre-emptive regulation of the grinder in cases where variations are unavoidable. In [33], a 60% increase in the plant throughput was achieved this way, by using physical tracers to track the material from different sources to the mill via crushers and storages. That is, however, not sustainable over time, and a digital tracking system must be used in place. ...
Full-text available
Systems for transport and processing of granular media are challenging to analyse, operate and optimise. In the mining and mineral processing industries, these systems are chains of processes with a complex interplay among the equipment, control and processed material. The material properties have natural variations that are usually only known at certain locations. Therefore, we explored a material-oriented approach to digital twins with a particle representation of the granular media. In digital form, the material is treated as pseudo-particles, each representing a large collection of real particles of various sizes, shapes and mineral properties. Movements and changes in the state of the material are determined by the combined data from control systems, sensors, vehicle telematics and simulation models at locations where no real sensors could see. The particle-based representation enables material tracking along the chain of processes. Each digital particle can act as a carrier of observational data generated by the equipment as it interacts with the real material. This make it possible to better learn the material properties from process observations and to predict the effect on downstream processes. We tested the technique on a mining simulator and demonstrated the analysis that can be performed using data from cross-system material tracking.
... Energy savings in comminution have become an increasingly important topic in mineral processing research (Bearman, 2012). Research into this subject has followed a number of different routes, including modelling to better understand dynamics during comminution (Weerasekara et al., 2013), improvements to existing equipment design and operation to improve performance (Rybinski et al., 2011), mine-to-mill optimisation (Djordevic & Michaux, 2005) and design/implementation of new technologies such as HPGR and Isamill. High voltage breakage falls into the latter category, being a novel technology that offers significant potential for energy savings as well as improvements in liberation. ...
Conference Paper
High voltage breakage is an emerging comminution technology with demonstrated advantages in terms of particle weakening and improved liberation. However, up till now a holistic optimisation assessment was lacking. In this work a number of relationships are presented to describe high voltage breakage behaviour of an ore. Based on these relationships a standardised methodology is outlined for single particle testing that allows for assessment of weakening and size reduction and the optimum generator setup. A relationship similar to the drop weight test breakage function allows determination of the weakening-energy relation, and the pre-weakening index helps ascertain the most efficient pulse energy. Weibull distributions can be used to describe the probability of electrical breakdown occurring. It is shown 69% weakening can be achieved using 2 kWh/t at an optimum pulse energy of 385 Joule and an electrical field of 5.5 kV/mm. Scale-up from single particle to continuous tests is also demonstrated.
Conference Paper
Full-text available
Hatch and Enaex conducted a joint Mine-to-Mill optimization project at Lundin Mining's Chapada operation in Brazil. The main objective was to maximize plant throughput with existing equipment while maintaining the product size of the comminution circuit. The project included a review of the available ore characterization data, ore domain definition, drill and blast operating practices, mining and processing operations, in-depth data analysis of historical process variables, planning and supervision of a full Mine-to-Mill survey, rock characterization for the surveyed ore, and the development of mathematical models for blasting and comminution, or updates to existing models where available. The calibrated, site-specific models were used in conjunction with comprehensive data analysis and ore domain definition to conduct simulations and integrated optimization of the 2 drill and blast and comminution processes. A set of blasting guidelines were proposed, which consisted of optimized drill and blast designs for each ore domain, as defined by a combination of site lithology classifications, rock strength, and rock structure. These blast guidelines were tailored to each domain to provide an optimized and consistent Run-of-Mine (ROM) fragmentation for downstream processes. The comminution circuit would receive finer feed with the optimized blasting, with further opportunities for improvement identified in the comminution circuits themselves. These opportunities included changes to the operating conditions of the primary and pebble crushers, as well changes to the Semi-Autogenous Grinding (SAG) and ball mills-including optimising the design of the shell lifters/liners, SAG grates and pulp lifters, ball charge level and media size-and cyclones. Taken together, these improvements ensure that the benefits of the improved ROM PSD are fully realised, and throughput maximised. The predicted benefit from the full implementation of the recommendations was an increase in plant throughput of 13-22 % (for the Soft-Fine to Hard-Coarse domains, respectively) while maintaining the final grinding product size below the target. The recommendations are currently being implemented, with throughputs of between 3,300-3,500 tph achieved to date. This is consistent with the predictions of the study for the current extent of implementation. If the recommendations are implemented in full, sustained throughput rates above 3,500 tph are expected to be achieved, all while maintaining the product size. The recommendations provided were practical to implement for both the mining and processing groups at Chapada, with no significant capital investment involved in the project.
Full-text available
Buzwagi Gold Mine (BGM) is operated by Acacia Mining and located in the Lake Victoria Goldfields of central Tanzania. The mine commenced its operation since April 2009 and treats a sulphide copper-gold ore to produce gold in form of doré bars and a concentrate containing gold, copper and silver. The BGM comminution circuit includes a primary crushing stage with a gyratory crusher and a two grinding circuits using a Semi-Autogenous Grinding (SAG) mill and a ball mill. The SAG mill circuit also includes a single-deck screen and a cone crusher while the ball mill circuit utilizes hydrocyclones. Currently, the grinding circuits are inefficient in achieving the aspired product fineness of xP,80 = 125 μm even at low to normal throughputs (450-600 t/h). An evaluation and optimization study of the circuit performance was conducted to improve the product fineness through circuit surveys, experimental lab work and simulations. In three full scale sampling campaigns, size distributions and solids contents of the samples were determined at selected points in the circuit. Further, several types of breakage tests were conducted; standard Bond tests to determine ore grindability and work indices, batch grinding tests to determine parameters for breakage and selection functions , and standard ball mill tests for mineral liberation characterization by an automated mineral liberation analyzer (MLA).The tests were conducted in a size range from 0.063 to 2 mm. Then, mass balance of the circuit was calculated and the models for mills, screens and hydrocyclones were employed in MODSIM (version 3.6.24). Firstly, simulations were conducted to optimize the existing plant. Several options were evaluated such as reduction of SAG screen aperture, adjustment of cyclone feed solids content and reduction of vortex finder and apex diameters. Moreover, simulations were also evaluated for a possible modification of the existing circuit and include; partial splitting of the cyclone underflow back to SAG mill, introduction of a second classification stage as well as introduction of a second ball mill. The evaluation of breakage tests and survey data revealed the following; the Bond work index obtained for the current ore ranges between 17.20 - 18.70 kWh/t compared to 14.50 - 16.50 kWh/t which was estimated during plant design.This indicates a change in hardness of the ore during the last 7 years. Harder ore means more energy requirement for an efficient operation, the consequence of which is increased costs. iv Thus, a periodic review of the ore hardness for ongoing mining operation is recommended. This will help in establishing better blends as well as prediction of appropriate tonnages for the existing ore types, so as to be efficiently treated by the available plant design. The work indices of the ore blends treated during survey were correlated with their quartz content and showed a strong linear relationship (R2= 0.95). Therefore, the work index for the BGM ore could be predicted based on known quartz content of the material. Further, the model could be used as a control tool for monitoring hardness variation of the SAG mill feed. The mineral liberation studies indicated that the valuable phase (pyrite-pyrrhotite) could be liberated at relatively coarser particle sizes (200-400 µm). This implies that, there could be no problem with the efficiency of the gravity circuit for the BGM operation, where the gold contained in pyrite-pyrrhotite could be easily concentrated. However, the efficiency of flotation and cyanidation processes will still require finer feed. In overall, the liberation characteristics of the ore blends treated during survey showed minor differences. The Bond efficiency factors of 48-61 % were obtained for the BGM grinding circuit, indicating an inefficient operation. This suggests that the operation could achieve targets by lowering the throughput. Further, the SAG mill circuit was characterized by fluctuating feed size of between xF,80 =102 to 185 mm. A need for control of the feed size as well as blending ratios was recommended for an efficient operation in terms of throughput and final product size. This could be achieved through closer monitoring of the primary crusher performance and proper control of the ratios for the SAG mill feeders drawing the ore from the stockpile. The ball mill grinding efficiency was poor and could be indicated by the fraction < 125 µm of only 5-9 % or xP, 80 : >400 µm in the mill discharge. This was deemed due to poor hydrocyclone performance which was characterized by higher feed solids content, coarser overflow xP,80: >200 µm as well as cut sizes, xT : > 200 µm. An improvement of product fineness up to 327 µm could be achieved during the simulation and optimization of the existing design. This could be achieved by modification of the operating conditions such as reduction of SAG screen aperture from 12 mm to 10 mm, reduction of vortex finder from 280 mm to 270.3 mm, reduction of apex diameter from 150 mm to 145.6 mm as well as adjustment of the cyclone feed solids content from 66.7 to 67.1 %. Based on this result, it was concluded that the current equipment could not achieve the target product quality (i.e. xP,80 = 125 µm ). v Further simulations based on flowsheet modification options showed that a second ball mill (series configuration) can help to achieve the desired product fineness as well as an increase of throughput from 618 t/h to 780 t/h. Although the circulating load increases to approximately 500 % in this configuration, it is outweighed by the benefits. Importantly, this option is cost intensive and hence may be considered as a long term solution and especially after cost-benefit analysis. Finally, the results based on optimization of the existing design is recommended as short term solution for improvement of the BGM operation. Although the fineness achieved is still low (i.e. xP,80 = 327 µm) compared to the target (i.e. xP,80 = 125 µm), this gives additional advantage in the sense that, also better hydrocyclone performance is achieved in terms of overflow product (xP,80 = 105 µm vs. > 240 µm) , cut size (xT =133.1 µm vs. > 220 µm) and circulating load (CL =350 %). The improved overflow fineness will contribute to improved efficiency for the downstream processes.
Conference Paper
Full-text available
Mine to mill optimisation will lead to a systematic reduction of production costs in an open mine. In this paper, mine to mill optimisation has been designed and implemented in the Sarcheshmeh copper mine. At first, databases have been provided. Then, specific drilling, specific charge and specific loading indices have been calculated and measured. With the help of statistical analysis, the relation between proper fragmentation of the blast and performance and efficiency of other production systems was studied. Among the parameters under study, the specific charge and the hole diameter have a very high correlation coefficient with and , so that in most of the proposed models, the correlation coefficient was higher than 0.9. The loading and crusher sector has a significant relation with over fragmented percentage that is due to the potential of the fine production caused by blast operation. The optimisation of the effective parameters on fragmentation was studied by design of experiments (DOE) method, where the specific charge was ranked first in terms of importance.
Conference Paper
Full-text available
Mine to mill optimization will lead to a systematic reduction of production costs in an open mine. In this paper, mine to mill optimization has been designed and implemented in the Sarcheshmeh copper mine. At first, databases have been provided. Then, specific drilling, specific charge and specific loading indices have been calculated and measured. With the help of statistical analysis, the relation between proper fragmentation of the blast and performance and efficiency of other production systems was studied. Among the parameters under study, the specific charge and the hole diameter have a very high correlation coefficient with D_50 and D_80, so that in most of the proposed models, the correlation coefficient was higher than 0.9. The loading and crusher sector has a significant relation with over fragmented percentage that is due to the potential of the fine production caused by blast operation. The optimization of the effective parameters on fragmentation was studied by design of experiments (DOE) method, where the specific charge was ranked first in terms of importance.
Design through to Operation
  • C Dechert
Dechert, C., 2006, Antamina -"Design through to Operation", SAG Conference 2006.
Metso Process Technology & Innovation -"Mine to Mill Process Integration & Optimisation at Compañía Minera Antamina
  • W Valery
  • E Sedat
  • J Colacioppo
  • D La Rosa
  • A Jankovic
Valery, W., Sedat, E., Colacioppo, J., La Rosa, D., Jankovic, A., Metso Process Technology & Innovation -"Mine to Mill Process Integration & Optimisation at Compañía Minera Antamina", 2007.
Process Technology & Innovation -"Mine to Mill Blast Fragmentation Project at Compañía Minera Antamina
  • W Valery
  • R Valle
  • S Corsini
  • S Kanchibotla
  • J Colacioppo
  • S Dikmen
  • P. -Metso Baguley
Valery, W., Valle, R., Corsini, S., Kanchibotla, S., Colacioppo, J., Dikmen, S., Baguley, P. -Metso Process Technology & Innovation -"Mine to Mill Blast Fragmentation Project at Compañía Minera Antamina", 2009.