XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 3911
model parameters (A, b values) generated from these fit-
tings are given in Table 4, to describe the competence of the
three components of hard, medium and soft within each
sample. These model parameters will be used in the multi-
component AG/SAG mill model to run several simulation
scenarios.
Jacobson and Fabbri (2019) reported that pebbles could
be 24% harder than the fresh ore, on average. Referring to
Table 1, at 50% percentile curve, Sample D was 29% more
resistant compared to its raw ore (Sample C). The ExDWT
results should explain why this is the case for a characterised
sample. Figure 7 compares the distribution of three com-
ponents of hard, medium and soft within Sample C and its
pebbles (Sample D).
The results from the ExDWT-based multi-component
breakage modelling and description approach confirm
more competent nature of oversized pebbles with the hard
component of more than doubled, which were sorted by
the SAG mill. This is an implication of SAG mill limit in
breaking particles with certain competence component—if
the amount of these material exceeds a certain percentage,
then it could limit throughput. It is because the available
specific energy in the mill is not sufficient to cause any
damage to these hard particles, hence they are accumu-
lated in the mill. Usage of large balls in the SAG mills can
increase the specific energy and improve the breakage of
these hard pebbles in the mill. However larger ball size may
reduce the number of balls for a given ball charge hence
reduce the number of collisions capable of exceeding the
EMin threshold within the mill. Hence, it can be expected
that a slight improvement in the probability of collisions
with energies enough to achieve EMin may contribute to
breakage of many more particles (Morrison et al., 2007).
SIMULATION SCENARIOS,
CONSTRAINTS AND CONSIDERATIONS
The comminution circuit at Barrick Cortez gold mine
is SAG mill (with inside liners diameter and length of
7.7×3.2 m) in closed circuit with a screen followed by a
ball mill in closed-circuit with hydrocyclone classifica-
tion (SAB). The SAG mill was operating in closed circuit
with a screen and the oversized material was returned to
the mill. At the time of survey, volumetric mill filling of
the SAG mill was 21.3% (charged with 12.1% grinding
steel balls) and it was operating at 75% of its critical speed.
A JKSimMet ® model of the SAG milling circuit at this
operation was transferred into the Integrated Extraction
Simulator (IES) developed by CRCORE (2020) in col-
laboration with The Julius Kruttschnitt Mineral Research
Centre, the University of Queensland. A range of widely
used JKMRC models were coded into this platform includ-
ing the JKMRC variable rates SAG mill model (Morrell et
al., 2001), which was used in this study to study the impact
of three components of competence as hard, medium and
Table 2. Distribution of three components of competence within each sample from the ExDWT data
Description Sample A Sample B Sample C Sample D*
Hard %(A×b38) 24.00 31.75 50.25 72.75
Medium %(38A×b67) 42.75 33.25 36.00 21.50
Soft %(A×b67) 33.25 35.00 13.75 5.75
*Pebbles of Sample C
Table 3. Model parameters of the split functions
Split Function
Sample A Sample B Sample C Sample D *
A b A b A b A b
A×b=38 57.11 0.67 60.62 0.63 53.75 0.71 61.18 0.62
A×b=67 62.04 1.08 60.59 1.11 66.10 1.01 56.24 1.19
*Pebbles of Sample C
Table 4. Breakage model parameters for different components of four samples tested by the ExDWT
Description
Sample A Sample B Sample C Sample D*
A b A×b A b A×b A b A×b A b A×b
Hard (A×b38) 56.24 0.54 30 61.92 0.45 28 50.39 0.53 27 73.90 0.31 23
Medium (38A×b67) 57.74 0.89 52 59.33 0.86 51 59.54 0.79 47 58.12 0.77 45
Soft (A×b67) 64.15 1.36 88 60.96 1.56 95 69.71 1.24 86 58.62 1.32 78
*Pebbles of Sample C
model parameters (A, b values) generated from these fit-
tings are given in Table 4, to describe the competence of the
three components of hard, medium and soft within each
sample. These model parameters will be used in the multi-
component AG/SAG mill model to run several simulation
scenarios.
Jacobson and Fabbri (2019) reported that pebbles could
be 24% harder than the fresh ore, on average. Referring to
Table 1, at 50% percentile curve, Sample D was 29% more
resistant compared to its raw ore (Sample C). The ExDWT
results should explain why this is the case for a characterised
sample. Figure 7 compares the distribution of three com-
ponents of hard, medium and soft within Sample C and its
pebbles (Sample D).
The results from the ExDWT-based multi-component
breakage modelling and description approach confirm
more competent nature of oversized pebbles with the hard
component of more than doubled, which were sorted by
the SAG mill. This is an implication of SAG mill limit in
breaking particles with certain competence component—if
the amount of these material exceeds a certain percentage,
then it could limit throughput. It is because the available
specific energy in the mill is not sufficient to cause any
damage to these hard particles, hence they are accumu-
lated in the mill. Usage of large balls in the SAG mills can
increase the specific energy and improve the breakage of
these hard pebbles in the mill. However larger ball size may
reduce the number of balls for a given ball charge hence
reduce the number of collisions capable of exceeding the
EMin threshold within the mill. Hence, it can be expected
that a slight improvement in the probability of collisions
with energies enough to achieve EMin may contribute to
breakage of many more particles (Morrison et al., 2007).
SIMULATION SCENARIOS,
CONSTRAINTS AND CONSIDERATIONS
The comminution circuit at Barrick Cortez gold mine
is SAG mill (with inside liners diameter and length of
7.7×3.2 m) in closed circuit with a screen followed by a
ball mill in closed-circuit with hydrocyclone classifica-
tion (SAB). The SAG mill was operating in closed circuit
with a screen and the oversized material was returned to
the mill. At the time of survey, volumetric mill filling of
the SAG mill was 21.3% (charged with 12.1% grinding
steel balls) and it was operating at 75% of its critical speed.
A JKSimMet ® model of the SAG milling circuit at this
operation was transferred into the Integrated Extraction
Simulator (IES) developed by CRCORE (2020) in col-
laboration with The Julius Kruttschnitt Mineral Research
Centre, the University of Queensland. A range of widely
used JKMRC models were coded into this platform includ-
ing the JKMRC variable rates SAG mill model (Morrell et
al., 2001), which was used in this study to study the impact
of three components of competence as hard, medium and
Table 2. Distribution of three components of competence within each sample from the ExDWT data
Description Sample A Sample B Sample C Sample D*
Hard %(A×b38) 24.00 31.75 50.25 72.75
Medium %(38A×b67) 42.75 33.25 36.00 21.50
Soft %(A×b67) 33.25 35.00 13.75 5.75
*Pebbles of Sample C
Table 3. Model parameters of the split functions
Split Function
Sample A Sample B Sample C Sample D *
A b A b A b A b
A×b=38 57.11 0.67 60.62 0.63 53.75 0.71 61.18 0.62
A×b=67 62.04 1.08 60.59 1.11 66.10 1.01 56.24 1.19
*Pebbles of Sample C
Table 4. Breakage model parameters for different components of four samples tested by the ExDWT
Description
Sample A Sample B Sample C Sample D*
A b A×b A b A×b A b A×b A b A×b
Hard (A×b38) 56.24 0.54 30 61.92 0.45 28 50.39 0.53 27 73.90 0.31 23
Medium (38A×b67) 57.74 0.89 52 59.33 0.86 51 59.54 0.79 47 58.12 0.77 45
Soft (A×b67) 64.15 1.36 88 60.96 1.56 95 69.71 1.24 86 58.62 1.32 78
*Pebbles of Sample C