1398 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
provided the most statistically significant and reliable cor-
relations. Hence 10g was selected for the SFH test protocol
and moved to the next phase of testing.
Comparison of SFH Test Size Distributions
The question of which size fraction below 2 mm to use in
the SFH test protocol was addressed by comparing the size
distributions of four separate size fractions and three sam-
ples, as follows:
Size Fractions
• 2.0×1.7 mm
• 1.7×1.4 mm
• 1.4×1.18 mm
• 1.18×1.00 mm
Samples
• 11202603 (Axb=32)
• 11202604 (Axb=331)
• 11202606 (Axb=96)
The size distributions shown in Figure 10 indicate that
smaller fractions are not as good as far as distinguishing the
samples in terms of the impact hardness (or Axb index).
It appears the 1.7×1.4 mm size fraction is the best choice
for the SFH protocol, based on the difference between size
distributions for the three samples selected.
APPLICATION OF SFH TESTS TO ALL 7
SAMPLES
The selected SFH test protocol of 10g using 1.7×1.4 mm
material was applied to all seven samples and results com-
pared in charts of HIT Axb versus %Unbroken and HIT
Axb versus %Passing 0.5 Initial Mean Size. Figure 11 shows
the X-Y plot comparisons.
The plots suggest the %Unbroken metric and fractions
smaller than 1.4 mm are not as reliable as 1.7×1.4 mm. The
most significant correlation is for the 1.7×1.4 mm size frac-
tion and %Passing 0.5 Initial Mean Size, showing an R^2
value of 0.97, expressed as follows:
HIT Axb Index =1126.1 × %P – 332.4 (2)
where %P =%Passing 0.5 Initial Mean Size (1.54 mm)
Clearly the dataset lacks softer samples, with Axb val-
ues in the 150 to 300 range. Hence more work is required
to confirm the robustness of the relationship between Axb
and %Passing 0.5 Initial Mean Size, if other metrics are
more relevant, and to see if the relationship is ore specific
or universal.
The initial results are very encouraging, suggesting the
SFH tests provide a simple and quick method of testing the
breakage properties of small particles which can be linked
to the standard measures of impact hardness such as Axb.
Figure 9. Charts for 2.0×1.7 mm samples (5, 10, and 15g), showing HIT Axb vs %Unbroken and HIT Axb vs %Passing 0.5 RS
(10g highlighted in red as most reliable)
provided the most statistically significant and reliable cor-
relations. Hence 10g was selected for the SFH test protocol
and moved to the next phase of testing.
Comparison of SFH Test Size Distributions
The question of which size fraction below 2 mm to use in
the SFH test protocol was addressed by comparing the size
distributions of four separate size fractions and three sam-
ples, as follows:
Size Fractions
• 2.0×1.7 mm
• 1.7×1.4 mm
• 1.4×1.18 mm
• 1.18×1.00 mm
Samples
• 11202603 (Axb=32)
• 11202604 (Axb=331)
• 11202606 (Axb=96)
The size distributions shown in Figure 10 indicate that
smaller fractions are not as good as far as distinguishing the
samples in terms of the impact hardness (or Axb index).
It appears the 1.7×1.4 mm size fraction is the best choice
for the SFH protocol, based on the difference between size
distributions for the three samples selected.
APPLICATION OF SFH TESTS TO ALL 7
SAMPLES
The selected SFH test protocol of 10g using 1.7×1.4 mm
material was applied to all seven samples and results com-
pared in charts of HIT Axb versus %Unbroken and HIT
Axb versus %Passing 0.5 Initial Mean Size. Figure 11 shows
the X-Y plot comparisons.
The plots suggest the %Unbroken metric and fractions
smaller than 1.4 mm are not as reliable as 1.7×1.4 mm. The
most significant correlation is for the 1.7×1.4 mm size frac-
tion and %Passing 0.5 Initial Mean Size, showing an R^2
value of 0.97, expressed as follows:
HIT Axb Index =1126.1 × %P – 332.4 (2)
where %P =%Passing 0.5 Initial Mean Size (1.54 mm)
Clearly the dataset lacks softer samples, with Axb val-
ues in the 150 to 300 range. Hence more work is required
to confirm the robustness of the relationship between Axb
and %Passing 0.5 Initial Mean Size, if other metrics are
more relevant, and to see if the relationship is ore specific
or universal.
The initial results are very encouraging, suggesting the
SFH tests provide a simple and quick method of testing the
breakage properties of small particles which can be linked
to the standard measures of impact hardness such as Axb.
Figure 9. Charts for 2.0×1.7 mm samples (5, 10, and 15g), showing HIT Axb vs %Unbroken and HIT Axb vs %Passing 0.5 RS
(10g highlighted in red as most reliable)