3820 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
(AI) obtained was 1,042 g/t, thus classifying the chromium
ore as having medium abrasion (600 ≤ AI ≤ 1200 g/t). The
found Macon crushability (Br) was 47.58%, which indi-
cates an ore easy to crush (40 ≤ Br ≤ 50%). The lamellarity
of the ore was 7.85%, indicating that it was an ore with a
shape tending to cubic.
Figure 3 presents the particle size distribution of the
feed and product of the FERBASA’s bar mill. It is possible
to note that the ore after comminution had d80 =4,100 µm
and d20 =180 µm. Only 45% of the mass produced was
suitable for secondary grinding (below 1.0 mm).
Figure 4 shows the results obtained with Metso’s
HRC D1000. It is possible to note that the product’s d80
increased to 6,000 µm and the d20 to 1,500 µm, when com-
pared to the current circuit using a bar mill. After closing
the comminution circuit with a sieve, the d80 (gray curve)
was reduced to 640 µm and the d20 to 73 µm, 100% pass-
ing at 1.0 mm.
Another key point to note is the amount of material
produced below 250 µm (or 60#), since material that is
too fine compromises the production of chromite sand,
whose specification in said mesh must be between 25 and
48% through. .The final product generated with the HRC
D1000, after sieving, obtained 55.6% passing at 250 µm.
Figure 5 presents a comparison between the final product
obtained with the HRC D1000 (real) and the objective
particle size curve initially defined (theoretical). It is pos-
sible to note that for coarser particle sizes the results were
remarkably close to straight identity, indicating a strong
correlation between the theoretical and the real. However,
it is possible to notice a bias in the results obtained for fine
particle sizes (420 µm), where the product obtained was
finer than initially predicted, indicating a greater break-
down of the chromium ore than expected.
When the ore passes through the roller crusher, it can
form small agglomerates, called cakes in English. The mate-
rial produced in bench tests with the HRC D1000 could
be easily broken manually, with no caking problems being
noted in the product. Therefore, it is estimated that if there
is adequate pre-pulping in the sieve feed and the use of
spray, the agglomerates that may be formed will be com-
pletely dissolved.
The simulations carried out with the Bruno software
indicated that a Metso HRC 800 crusher is sufficient to
meet a production of 60 t/h of final product, operating
with a circulating load of 100% of the value of the new
feed. For a production of 120 t/h, one HRC 1000 or two
units of the HRC 800 is required. However, attention must
be paid to screening at the 1.0 mm mesh in order to keep
the circulating load stable.
Figure 2. Target particle size distribution for the comminution of the chromium ore
(AI) obtained was 1,042 g/t, thus classifying the chromium
ore as having medium abrasion (600 ≤ AI ≤ 1200 g/t). The
found Macon crushability (Br) was 47.58%, which indi-
cates an ore easy to crush (40 ≤ Br ≤ 50%). The lamellarity
of the ore was 7.85%, indicating that it was an ore with a
shape tending to cubic.
Figure 3 presents the particle size distribution of the
feed and product of the FERBASA’s bar mill. It is possible
to note that the ore after comminution had d80 =4,100 µm
and d20 =180 µm. Only 45% of the mass produced was
suitable for secondary grinding (below 1.0 mm).
Figure 4 shows the results obtained with Metso’s
HRC D1000. It is possible to note that the product’s d80
increased to 6,000 µm and the d20 to 1,500 µm, when com-
pared to the current circuit using a bar mill. After closing
the comminution circuit with a sieve, the d80 (gray curve)
was reduced to 640 µm and the d20 to 73 µm, 100% pass-
ing at 1.0 mm.
Another key point to note is the amount of material
produced below 250 µm (or 60#), since material that is
too fine compromises the production of chromite sand,
whose specification in said mesh must be between 25 and
48% through. .The final product generated with the HRC
D1000, after sieving, obtained 55.6% passing at 250 µm.
Figure 5 presents a comparison between the final product
obtained with the HRC D1000 (real) and the objective
particle size curve initially defined (theoretical). It is pos-
sible to note that for coarser particle sizes the results were
remarkably close to straight identity, indicating a strong
correlation between the theoretical and the real. However,
it is possible to notice a bias in the results obtained for fine
particle sizes (420 µm), where the product obtained was
finer than initially predicted, indicating a greater break-
down of the chromium ore than expected.
When the ore passes through the roller crusher, it can
form small agglomerates, called cakes in English. The mate-
rial produced in bench tests with the HRC D1000 could
be easily broken manually, with no caking problems being
noted in the product. Therefore, it is estimated that if there
is adequate pre-pulping in the sieve feed and the use of
spray, the agglomerates that may be formed will be com-
pletely dissolved.
The simulations carried out with the Bruno software
indicated that a Metso HRC 800 crusher is sufficient to
meet a production of 60 t/h of final product, operating
with a circulating load of 100% of the value of the new
feed. For a production of 120 t/h, one HRC 1000 or two
units of the HRC 800 is required. However, attention must
be paid to screening at the 1.0 mm mesh in order to keep
the circulating load stable.
Figure 2. Target particle size distribution for the comminution of the chromium ore