4
Minera Nacional, S.A., gold plant in the Andes mountains
of Peru. He was joined there by his family. At start-up he
ran the grinding and flotation circuits. He visited numer-
ous mines in the area, including Bolivian tin plants, on
behalf of Allis-Chalmers. He helped clients’ make major
improvements to gold cyanidation operations, at which he
had become recognized as an expert. He turned down an
offer to extend his stay in South America to continue doing
such work, and returned to Milwaukee in late 1936. This
was largely for family reasons, but also to cultivate a grow-
ing interest in comminution.
Bond’s first published article related to comminution
was entitled “Calculating Average Particle Size from Screen
Analysis” in Engineering and Mining Journal (E&MJ), July,
1931. It provides a method to represent a complete size dis-
tribution by its “50% median grain diameter” by interpola-
tion between the sieve openings passing above and below
the 50% cumulative passing. Although not suggested in the
article, this could be considered necessary to relate energy
used to measured breakage, the “average” particle size before
and after comminution.
In his first major coauthored publication on grinding,
the 1933 “Grindability of Various Ores,” Bond first men-
tions a timed, laboratory, batch ball mill grinding test that
was used early on at the Allis-Chalmers laboratory to com-
pare ores, similar to the method of Lennox (1919). Mill
sales engineers (including Walter Maxson) used this infor-
mation and data they collected (and zealously guarded, in
their “little black books”) on plant mills’ grinding perfor-
mances to help size mills for new customers’ plants. But they
discovered that a locked-cycle test was needed to deal with
a hard to grind fraction in the ore. The timing of the tran-
sition to the locked-cycle test is not precisely mentioned.
However, Bond prepared the paper on the new method,
along with example calculations and more than seventy test
results on identified ores and other materials, and presented
it to the AIME meeting in New York in February, 1933. He
listed W.L. Maxson, F. Cadena and himself, in that order, as
co-authors. The Allis-Chalmers grindability test that they
described in 1933 remains today as the universal standard
for the industry. Although later renamed the Bond Work
Index Test, given Bond’s relatively short tenure at Allis-
Chalmers at the time, beyond writing the paper Bond likely
played a lesser role in developing the test compared to the
others listed as first and second authors.
Bond suggested that this publication “marked the
beginning of the end of secret little black books of grinding
results.” But it provided only the half of the data related
to the ore needed for mill sizing. The second half was the
(unidentified) plant grinding mill performance associated
with each of the grindabilities. The latter would be revealed
publicly over a decade later.
The paper also included tables of the test feed, product,
and circulating load particle mean diameters, as calculated
by the method in the previous reference. No analysis of
these was attempted. They also reported on running the test
wet, vs. dry, showing a 1.5 to 2 times higher net production
rate, depending on percent solids. They suggested that the
same test could be used to study numerous other grinding
variables like mill speed, media load and size or shape, mill
lining and circulating load.
It was at that meeting in New York that Bond also first
met Prof. A. F. Taggart. Taggart later (1945) described the
Allis-Chalmers test as a variation of the Yancy developed
ASTM procedure for the relative grindability of coal (Yancy
et al, 1934). The shorter and simpler, yet locked-cycle,
method for coal was being developed at the same time as
the Allis-Chalmers method. However, there is no sugges-
tion that the two laboratories ever discussed their respec-
tive work. No test on coal was listed by Allis-Chalmers in
1933. Later, the U.S. Department of the Interior (Obert et
al, 1946) adopted the Allis-Chalmers method as the sole
means to suitably characterize the ball mill grindability of
rocks and minerals.
It should also be emphasized that both these, and other,
test results were recognized and referred to as “relative
grindabilities” of ores, correctly indicating that changing
the test conditions, and resulting interactions with differ-
ent ores, could change even the relative outcomes. Batch vs.
locked-cycle test results clearly exemplified this.
Despite his lengthy assignments abroad from 1933
to 1936, Bond, along with Maxson, produced “Crushing
and Grinding Characteristics as Determined from Screen
Analyses,” submitted it to the AIME in August of 1934, and
it was published in the transactions in 1935. They described
how examination of screen analyses could reveal both ore
characteristics (e.g., a hard grinding fraction accumulating
at a certain size) and comminution conditions (e.g., a nar-
rower product sizing from less overgrinding). They noted
that Gaudin’s (1926) observation of log-log linearity for
the fine fractions of ground material size distributions was
highly prevalent. They also expressed that “practical effi-
ciencies” should compare relative grindabilities to equip-
ment power used at the same grind sizes, and that these
did not follow Rittinger based “absolute power efficiencies”
estimated from calculations of new surface areas.
In 1935 Bond also wrote an article for the June CSM
Magazine entitled “The Theory of Crushing.” This was a
highly theoretical analysis of new surface area produced
with breakage of rectangular cuboids into specific geometric
Minera Nacional, S.A., gold plant in the Andes mountains
of Peru. He was joined there by his family. At start-up he
ran the grinding and flotation circuits. He visited numer-
ous mines in the area, including Bolivian tin plants, on
behalf of Allis-Chalmers. He helped clients’ make major
improvements to gold cyanidation operations, at which he
had become recognized as an expert. He turned down an
offer to extend his stay in South America to continue doing
such work, and returned to Milwaukee in late 1936. This
was largely for family reasons, but also to cultivate a grow-
ing interest in comminution.
Bond’s first published article related to comminution
was entitled “Calculating Average Particle Size from Screen
Analysis” in Engineering and Mining Journal (E&MJ), July,
1931. It provides a method to represent a complete size dis-
tribution by its “50% median grain diameter” by interpola-
tion between the sieve openings passing above and below
the 50% cumulative passing. Although not suggested in the
article, this could be considered necessary to relate energy
used to measured breakage, the “average” particle size before
and after comminution.
In his first major coauthored publication on grinding,
the 1933 “Grindability of Various Ores,” Bond first men-
tions a timed, laboratory, batch ball mill grinding test that
was used early on at the Allis-Chalmers laboratory to com-
pare ores, similar to the method of Lennox (1919). Mill
sales engineers (including Walter Maxson) used this infor-
mation and data they collected (and zealously guarded, in
their “little black books”) on plant mills’ grinding perfor-
mances to help size mills for new customers’ plants. But they
discovered that a locked-cycle test was needed to deal with
a hard to grind fraction in the ore. The timing of the tran-
sition to the locked-cycle test is not precisely mentioned.
However, Bond prepared the paper on the new method,
along with example calculations and more than seventy test
results on identified ores and other materials, and presented
it to the AIME meeting in New York in February, 1933. He
listed W.L. Maxson, F. Cadena and himself, in that order, as
co-authors. The Allis-Chalmers grindability test that they
described in 1933 remains today as the universal standard
for the industry. Although later renamed the Bond Work
Index Test, given Bond’s relatively short tenure at Allis-
Chalmers at the time, beyond writing the paper Bond likely
played a lesser role in developing the test compared to the
others listed as first and second authors.
Bond suggested that this publication “marked the
beginning of the end of secret little black books of grinding
results.” But it provided only the half of the data related
to the ore needed for mill sizing. The second half was the
(unidentified) plant grinding mill performance associated
with each of the grindabilities. The latter would be revealed
publicly over a decade later.
The paper also included tables of the test feed, product,
and circulating load particle mean diameters, as calculated
by the method in the previous reference. No analysis of
these was attempted. They also reported on running the test
wet, vs. dry, showing a 1.5 to 2 times higher net production
rate, depending on percent solids. They suggested that the
same test could be used to study numerous other grinding
variables like mill speed, media load and size or shape, mill
lining and circulating load.
It was at that meeting in New York that Bond also first
met Prof. A. F. Taggart. Taggart later (1945) described the
Allis-Chalmers test as a variation of the Yancy developed
ASTM procedure for the relative grindability of coal (Yancy
et al, 1934). The shorter and simpler, yet locked-cycle,
method for coal was being developed at the same time as
the Allis-Chalmers method. However, there is no sugges-
tion that the two laboratories ever discussed their respec-
tive work. No test on coal was listed by Allis-Chalmers in
1933. Later, the U.S. Department of the Interior (Obert et
al, 1946) adopted the Allis-Chalmers method as the sole
means to suitably characterize the ball mill grindability of
rocks and minerals.
It should also be emphasized that both these, and other,
test results were recognized and referred to as “relative
grindabilities” of ores, correctly indicating that changing
the test conditions, and resulting interactions with differ-
ent ores, could change even the relative outcomes. Batch vs.
locked-cycle test results clearly exemplified this.
Despite his lengthy assignments abroad from 1933
to 1936, Bond, along with Maxson, produced “Crushing
and Grinding Characteristics as Determined from Screen
Analyses,” submitted it to the AIME in August of 1934, and
it was published in the transactions in 1935. They described
how examination of screen analyses could reveal both ore
characteristics (e.g., a hard grinding fraction accumulating
at a certain size) and comminution conditions (e.g., a nar-
rower product sizing from less overgrinding). They noted
that Gaudin’s (1926) observation of log-log linearity for
the fine fractions of ground material size distributions was
highly prevalent. They also expressed that “practical effi-
ciencies” should compare relative grindabilities to equip-
ment power used at the same grind sizes, and that these
did not follow Rittinger based “absolute power efficiencies”
estimated from calculations of new surface areas.
In 1935 Bond also wrote an article for the June CSM
Magazine entitled “The Theory of Crushing.” This was a
highly theoretical analysis of new surface area produced
with breakage of rectangular cuboids into specific geometric