XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 1429
recognition of the economic “monetary” impact of the coal
losses, and (iii) aggressive steps taken by plant operators to
correct/mitigate operational problems.
OPERATION AT CONSTANT
INCREMENTAL QUALITY
Operation at a constant incremental quality is the old-
est governing principle in preparation plant optimization
(Mayer 1950 Dell 1956 Rayner 1987). For coking plants
constrained by dry ash, this concept dictates that all circuits
be operated at the same incremental ash regardless of parti-
cle size and washability. Likewise, energy plants constrained
by a calorific specification should operate at the same incre-
mental inerts (i.e., ash +moisture) to maximize overall yield
(Bethell et al 2006 Luttrell et al 2009 Luttrell et al 2014).
In layman’s terms, incremental quality is the effective grade
specification of the last unit of material recovered/discarded
in a circuit when the yield is increased/decreased by an
infinitesimally small amount. Unfortunately, incremen-
tal quality is a mathematical term that cannot be directly
monitored. However, it can be estimated for ideal/efficient
separations if the parameter of interest is ash since the ash
content of a particle can be directly correlated with particle
density (Abbott and Miles 1990). Due to the ash-density
correlation, total plant yield can be maximized when all
parallel circuits are operated at the same density cutpoint
(Clarkson 1992). This statement holds regardless of the size
or washability characteristics of the feed coal, provided that
the contract constraint is dry ash and very efficient separa-
tions are maintained in each circuit. For less efficient pro-
cesses, such as fine coal density-based separators, slightly
higher RD cutpoints are needed to counteract the effects
of misplaced coal/middlings on incremental ash (Luttrell
et al 2003a). The required RD increase can be estimated
from simulation software. This optimization concept also
dictates that density cutpoints should not be adjusted in
real-time and upstream or downstream blending should
be used instead to regulate quality (Luttrell et al 2003b
Bratton et al 2014).
The benefit of the incremental quality concept can be
demonstrated using the washability data summarized in
Table 7. The data correspond to parallel dense medium cir-
cuits consisting of a dense medium vessel that treats 150
tph of 100 × 6 mm feed solids and a dense medium cyclone
that treats 350 tph of –6 +1 mm feed solids.
First consider the case where both circuits are set to
produce the same clean coal ash of 7.0%. Since the coarser
coal is less liberated, the vessel RD cutpoint must be set to
a low value of 1.35 RD to meet cumulative 7.0% ash limit.
In contrast, the coal fed to the dense medium cyclones is
much better liberated and, as such, can operate at a much
higher RD cutpoint of 1.80 while still maintaining the
same cumulative 7.0% product ash. While these operat-
ing points appear to be suitable, they cannot be optimal.
When operating at a cumulative ash of 7.0% (RD=1.35),
the last particle recovered in the vessel contained 8.3% ash.
To get the same cumulative clean coal ash of 7.0%, the
cyclone circuit had to be operated at a much higher cut-
point (RD=1.80) and the last increment recovered in this
circuit contained 44.5% ash. The combined result is infe-
rior since one circuit is sending 8.3% ash solids to reject,
while the other is sending 44.5% ash solids to clean coal.
To correct this shortcoming, it would be prefer-
able to operate both circuits at the same density cutpoint
Table 7. Washability data for two parallel dense medium separators (dry basis)
Float
RD
Dense Medium Vessel Dense Medium Cyclones
Individual Cumulative Individual Cumulative
Mass, %Ash, %Mass, %Ash, %Mass, %Ash, %Mass, %Ash, %
1.30 11.4 3.9 11.4 3.9 59.2 3.5 59.2 3.5
1.35 27.5 8.3 38.9 7.0 6.1 7.6 65.3 3.9
1.40 16.3 13.7 55.2 9.0 3.7 13.1 69 4.4
1.45 3.8 20.1 59 9.7 2.3 18.5 71.3 4.8
1.50 2.5 25.8 61.5 10.4 1.9 23.8 73.2 5.3
1.55 1.3 29.7 62.8 10.8 1.0 29.0 74.2 5.6
1.60 0.8 33.9 63.6 11.0 0.9 33.2 75.1 6.0
1.65 0.3 37.1 63.9 11.2 0.5 37.2 75.6 6.2
1.70 0.3 40.1 64.2 11.3 0.6 40.4 76.2 6.5
1.80 0.4 45 64.6 11.5 1.1 44.5 77.3 7.0
1.90 0.6 52.7 65.2 11.9 1.0 51.6 78.3 7.6
2.00 0.5 62.1 65.7 12.3 1.0 60.8 79.3 8.2
Sink 34.3 88.5 100.0 38.4 20.7 87.2 100.0 24.6
recognition of the economic “monetary” impact of the coal
losses, and (iii) aggressive steps taken by plant operators to
correct/mitigate operational problems.
OPERATION AT CONSTANT
INCREMENTAL QUALITY
Operation at a constant incremental quality is the old-
est governing principle in preparation plant optimization
(Mayer 1950 Dell 1956 Rayner 1987). For coking plants
constrained by dry ash, this concept dictates that all circuits
be operated at the same incremental ash regardless of parti-
cle size and washability. Likewise, energy plants constrained
by a calorific specification should operate at the same incre-
mental inerts (i.e., ash +moisture) to maximize overall yield
(Bethell et al 2006 Luttrell et al 2009 Luttrell et al 2014).
In layman’s terms, incremental quality is the effective grade
specification of the last unit of material recovered/discarded
in a circuit when the yield is increased/decreased by an
infinitesimally small amount. Unfortunately, incremen-
tal quality is a mathematical term that cannot be directly
monitored. However, it can be estimated for ideal/efficient
separations if the parameter of interest is ash since the ash
content of a particle can be directly correlated with particle
density (Abbott and Miles 1990). Due to the ash-density
correlation, total plant yield can be maximized when all
parallel circuits are operated at the same density cutpoint
(Clarkson 1992). This statement holds regardless of the size
or washability characteristics of the feed coal, provided that
the contract constraint is dry ash and very efficient separa-
tions are maintained in each circuit. For less efficient pro-
cesses, such as fine coal density-based separators, slightly
higher RD cutpoints are needed to counteract the effects
of misplaced coal/middlings on incremental ash (Luttrell
et al 2003a). The required RD increase can be estimated
from simulation software. This optimization concept also
dictates that density cutpoints should not be adjusted in
real-time and upstream or downstream blending should
be used instead to regulate quality (Luttrell et al 2003b
Bratton et al 2014).
The benefit of the incremental quality concept can be
demonstrated using the washability data summarized in
Table 7. The data correspond to parallel dense medium cir-
cuits consisting of a dense medium vessel that treats 150
tph of 100 × 6 mm feed solids and a dense medium cyclone
that treats 350 tph of –6 +1 mm feed solids.
First consider the case where both circuits are set to
produce the same clean coal ash of 7.0%. Since the coarser
coal is less liberated, the vessel RD cutpoint must be set to
a low value of 1.35 RD to meet cumulative 7.0% ash limit.
In contrast, the coal fed to the dense medium cyclones is
much better liberated and, as such, can operate at a much
higher RD cutpoint of 1.80 while still maintaining the
same cumulative 7.0% product ash. While these operat-
ing points appear to be suitable, they cannot be optimal.
When operating at a cumulative ash of 7.0% (RD=1.35),
the last particle recovered in the vessel contained 8.3% ash.
To get the same cumulative clean coal ash of 7.0%, the
cyclone circuit had to be operated at a much higher cut-
point (RD=1.80) and the last increment recovered in this
circuit contained 44.5% ash. The combined result is infe-
rior since one circuit is sending 8.3% ash solids to reject,
while the other is sending 44.5% ash solids to clean coal.
To correct this shortcoming, it would be prefer-
able to operate both circuits at the same density cutpoint
Table 7. Washability data for two parallel dense medium separators (dry basis)
Float
RD
Dense Medium Vessel Dense Medium Cyclones
Individual Cumulative Individual Cumulative
Mass, %Ash, %Mass, %Ash, %Mass, %Ash, %Mass, %Ash, %
1.30 11.4 3.9 11.4 3.9 59.2 3.5 59.2 3.5
1.35 27.5 8.3 38.9 7.0 6.1 7.6 65.3 3.9
1.40 16.3 13.7 55.2 9.0 3.7 13.1 69 4.4
1.45 3.8 20.1 59 9.7 2.3 18.5 71.3 4.8
1.50 2.5 25.8 61.5 10.4 1.9 23.8 73.2 5.3
1.55 1.3 29.7 62.8 10.8 1.0 29.0 74.2 5.6
1.60 0.8 33.9 63.6 11.0 0.9 33.2 75.1 6.0
1.65 0.3 37.1 63.9 11.2 0.5 37.2 75.6 6.2
1.70 0.3 40.1 64.2 11.3 0.6 40.4 76.2 6.5
1.80 0.4 45 64.6 11.5 1.1 44.5 77.3 7.0
1.90 0.6 52.7 65.2 11.9 1.0 51.6 78.3 7.6
2.00 0.5 62.1 65.7 12.3 1.0 60.8 79.3 8.2
Sink 34.3 88.5 100.0 38.4 20.7 87.2 100.0 24.6