XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 501
Determining the processing for the cleaning stage was
relatively simpler. In all cases the primary goal was to maxi-
mize the enrichment of Fe, and without a specific yield tar-
get there was no minimum yield that had to be achieved
for the process to remain viable. Thus, the constraints and
goals of the first pass are the same as the second pass clean-
ing. Therefore, the same conditions were used for the sec-
ond pass cleaning as the first pass.
Scavenging offered a different challenge. The second
pass processing of the byproduct was necessary to raise the
yield of the entire process. Therefore, the process conditions
of the first pass that sought to maximize Fe enrichment were
not always applicable. The scavenging process had to maxi-
mize yield at the Fe increase required to match the original
feed concentration. The model constructed for first pass
optimization could also be used for determining second
pass conditions with minimal modification. The response
‘Fe’ could be removed from the model, which left Delta
Fe as the single iron enrichment response. The Prediction
Profiler could then be used to find conditions that maxi-
mized yield at the Delta Fe required to increase the byprod-
uct Fe concentration to that of the original feed. Due to the
relatively low yields of Production 2, the Fe concentration
of their respective byproducts was close to the original feed.
Therefore, only a modest enrichment of Fe was necessary to
match the feed. This made it easier to achieve higher yields
in the scavenging. Production 1s byproduct had the oppo-
site problem, the high yield meant that a larger increase in
Figure 5. Example of the ‘Prediction Profiler’
Table 5. Results of bulk first-pass processing schemes
Run Class
Conditions Results
Belt Speed Gap Voltage Feed Port Polarity
Belt
Material
Delta Fe
(%)
Yield
(%)
Production 1 52 330 10.0 1 Positive 2 4.84 ± 2.03 61.4 ± 1.0
Production 2 65 330 10.0 3 Positive 2 9.61 ± 0.33 32.8 ± 1.2
Determining the processing for the cleaning stage was
relatively simpler. In all cases the primary goal was to maxi-
mize the enrichment of Fe, and without a specific yield tar-
get there was no minimum yield that had to be achieved
for the process to remain viable. Thus, the constraints and
goals of the first pass are the same as the second pass clean-
ing. Therefore, the same conditions were used for the sec-
ond pass cleaning as the first pass.
Scavenging offered a different challenge. The second
pass processing of the byproduct was necessary to raise the
yield of the entire process. Therefore, the process conditions
of the first pass that sought to maximize Fe enrichment were
not always applicable. The scavenging process had to maxi-
mize yield at the Fe increase required to match the original
feed concentration. The model constructed for first pass
optimization could also be used for determining second
pass conditions with minimal modification. The response
‘Fe’ could be removed from the model, which left Delta
Fe as the single iron enrichment response. The Prediction
Profiler could then be used to find conditions that maxi-
mized yield at the Delta Fe required to increase the byprod-
uct Fe concentration to that of the original feed. Due to the
relatively low yields of Production 2, the Fe concentration
of their respective byproducts was close to the original feed.
Therefore, only a modest enrichment of Fe was necessary to
match the feed. This made it easier to achieve higher yields
in the scavenging. Production 1s byproduct had the oppo-
site problem, the high yield meant that a larger increase in
Figure 5. Example of the ‘Prediction Profiler’
Table 5. Results of bulk first-pass processing schemes
Run Class
Conditions Results
Belt Speed Gap Voltage Feed Port Polarity
Belt
Material
Delta Fe
(%)
Yield
(%)
Production 1 52 330 10.0 1 Positive 2 4.84 ± 2.03 61.4 ± 1.0
Production 2 65 330 10.0 3 Positive 2 9.61 ± 0.33 32.8 ± 1.2