3906 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
INTRODUCTION
Mineral deposits are found in many types showing dis-
tinctive geological and geo-mechanical properties, but all
exhibit some degree of heterogeneity. To mineral engineers,
it is vital to know how heterogeneity, e.g., grade, valuable
and deleterious minerals, competence, etc., is distributed
within an ore deposit and its key domains—because it
determines mill feed quality.
Distribution of ore competence within a deposit and
its key domains governs the energy and cost required to
liberate valuable minerals from the gangue. Figure 1 pro-
vides a conceptual illustration of variability within an ore
deposit which includes three key domains, named as A, B
and C and the ore deposit consists 30% of A, 50% of B and
20% of C. Each ore domain depending on its degree of het-
erogeneity exhibits an extent of variation in competence.
Average-based ore breakage tests generate several mean val-
ues that may describe competence heterogeneity within a
deposit ‘to some extent’. However, this level of resolution
may not be adequate because of (Faramarzi, 2020):
• Missing information at the extremities of a deposit:
The average-based methods describe competence by
relying on mean values that describe the difference
between key ore domains. However, the extent of
variability within a given ore domain remains inad-
equately quantified. This deficiency can impose risks
to an operation especially if the competence distribu-
tion is bi-modal and the variation between modes is
significant.
• Missing information regarding the extent of variabil-
ity within key ore domains: This type of variability
translates into variation in process performance.
Quantifying the spread of competence within key
ore domains can facilitate making operational deci-
sions. Additionally, having knowledge beyond the
average description of key ore domains should assist
in budgeting contingencies.
If a sample could preserve intrinsic heterogeneity with
an ore type, then it would be possible to quantify distribu-
tion of competence within it. With this proviso in mind,
the extent of competence variability within the mill feed
may be limited to an ore type or become stretched for
blends of diverse competences sourced from different geo-
logical units. Regardless of operational challenges for keep-
ing the process optimal, when a variable feed is introduced
to an AG/SAG mill, different components exhibit different
breakage characteristics inside the mill. Harder compo-
nents preferentially accumulate and dominate the mill con-
tent, limiting the throughput. Softer components empty
the mill quickly, drawing less power and increasing the risk
for liner breakage. Both extremes can pose problems to mill
operation. To integrate this effect into predictions, one of
the key requirements is to provide a description of the ore
types that predominantly constitute mill feed. To further
understand the implications of ore variability, the possibil-
ity of hardness/grindability changes through the commi-
nution-classification process was examined by Faramarzi et
al. (2022). The results from their laboratory experiments
on three different ore types suggested that depending on
ore characteristics, up to a 10% difference in the Bond Ball
Work index (BBWi) between circuit feed and the actual
ball mill feed is possible. However, the distinction might
become more noticeable with highly variable ore types.
Faramarzi et al. (2020) detailed development of the
Extended Drop Weight Test (ExDWT) which measures
Figure 1. Conceptual description of the variability within an ore deposit (Faramarzi, 2020)
INTRODUCTION
Mineral deposits are found in many types showing dis-
tinctive geological and geo-mechanical properties, but all
exhibit some degree of heterogeneity. To mineral engineers,
it is vital to know how heterogeneity, e.g., grade, valuable
and deleterious minerals, competence, etc., is distributed
within an ore deposit and its key domains—because it
determines mill feed quality.
Distribution of ore competence within a deposit and
its key domains governs the energy and cost required to
liberate valuable minerals from the gangue. Figure 1 pro-
vides a conceptual illustration of variability within an ore
deposit which includes three key domains, named as A, B
and C and the ore deposit consists 30% of A, 50% of B and
20% of C. Each ore domain depending on its degree of het-
erogeneity exhibits an extent of variation in competence.
Average-based ore breakage tests generate several mean val-
ues that may describe competence heterogeneity within a
deposit ‘to some extent’. However, this level of resolution
may not be adequate because of (Faramarzi, 2020):
• Missing information at the extremities of a deposit:
The average-based methods describe competence by
relying on mean values that describe the difference
between key ore domains. However, the extent of
variability within a given ore domain remains inad-
equately quantified. This deficiency can impose risks
to an operation especially if the competence distribu-
tion is bi-modal and the variation between modes is
significant.
• Missing information regarding the extent of variabil-
ity within key ore domains: This type of variability
translates into variation in process performance.
Quantifying the spread of competence within key
ore domains can facilitate making operational deci-
sions. Additionally, having knowledge beyond the
average description of key ore domains should assist
in budgeting contingencies.
If a sample could preserve intrinsic heterogeneity with
an ore type, then it would be possible to quantify distribu-
tion of competence within it. With this proviso in mind,
the extent of competence variability within the mill feed
may be limited to an ore type or become stretched for
blends of diverse competences sourced from different geo-
logical units. Regardless of operational challenges for keep-
ing the process optimal, when a variable feed is introduced
to an AG/SAG mill, different components exhibit different
breakage characteristics inside the mill. Harder compo-
nents preferentially accumulate and dominate the mill con-
tent, limiting the throughput. Softer components empty
the mill quickly, drawing less power and increasing the risk
for liner breakage. Both extremes can pose problems to mill
operation. To integrate this effect into predictions, one of
the key requirements is to provide a description of the ore
types that predominantly constitute mill feed. To further
understand the implications of ore variability, the possibil-
ity of hardness/grindability changes through the commi-
nution-classification process was examined by Faramarzi et
al. (2022). The results from their laboratory experiments
on three different ore types suggested that depending on
ore characteristics, up to a 10% difference in the Bond Ball
Work index (BBWi) between circuit feed and the actual
ball mill feed is possible. However, the distinction might
become more noticeable with highly variable ore types.
Faramarzi et al. (2020) detailed development of the
Extended Drop Weight Test (ExDWT) which measures
Figure 1. Conceptual description of the variability within an ore deposit (Faramarzi, 2020)