Page 1
Optimized Iron Ore Blast Designs for SAG/AG Mills
Jack Eloranta
Eloranta & Associates Inc
Abstract
Few mine to mill issues inspires as heated debates as blasting for SAG & AG mills. Increased blast energy may produce more surface area at a lower cost than mechanical milling. Decreased blast energy may, on the other hand, create more large pieces needed for grinding media. Rational allocation of energy in the comminution of ores is the essence of profitability and survival of taconite producers.
A case study shows that lower powder factor does not always to an increase in production of ‘media rock’. Comprehension of rock strength and existing jointing is essential to changes in blast design. If weak bedding planes are less than 9 inches apart; lowering the powder factor may only slow mill throughput while increasing energy consumption.
Page 2
Introduction
The current environment for energy usage is ripe for optimization of energy usage in mining and milling. The 1988 paper by Revnitsev, “We Really Need Revolution in Comminution” offered a number of observations that demand close examination given the rising cost of electricity and petroleum fuels. Mineral processors have watched the advancement of products and methods that can now reduce energy usage and operating costs. However, commodity prices have been dropping and capital for new equipment and plant flowsheet changes has not been available. Now, ‘push has come to shove’. In a backhanded way, higher energy bills have made it easier to justify outlays for changes. Many iron ore companies need payback periods of less than 24 months to get funding approval. Today’s higher energy bills now shorten up the payback period.
The
golden age of flowsheet optimization may well come to fruition in the next five
years. If “one measurement is worth a thousand opinions”, then the myriad of
measurements that technology now affords us; will equal trillions of opinions.
Previous Work
Modern comminution theory goes back to 19th century Germany where Rittinger (1867) and Kick (1885) proposed models based on surface area and particle volume respectively. Bond (1951) proposed a third theory of comminution that is still widely used today. King and Schneider (1995) at the University of Utah have very recently demonstrated improved modeling of grinding circuits.
Overall blast optimization has more recent roots. MacKenzie (1966) reported on costs in iron ore from drilling through crushing. Udy and Thornley (1977) reviewed optimization through crushing. Gold (1987) tabulated and modeled overall mining cost related to blasting at Fording Coal. LeJuge and Cox (1995) reported overall costs in quarrying. Eloranta (1995) published costs in iron ore from blasting through grinding. Moody et al (1996) related dig times, crusher speeds and particle size to fragmentation in quarry operations. Furstenau (1995) used single-particle roll mill crushing to demonstrate a 10% energy savings in the drilling through grinding process by increasing powder factor by 25%.
Recent
laboratory work has been aimed at tying mine and processing size reduction to
common factors. These efforts include the work of Revnivtsev (1988) who related
micro-cracks from blasting to energy use in subsequent crushing and grinding.
McCarter (1996) has quantified blast preconditioning through the use of an
ultra fast load cell. Nielsen (1996) has done extensive grinding tests on
preconditioned rock and demonstrated changes in Bond work indices of nearly 3
to 1.
Page 3
|
Figure 1 |
Fragmentation is controlled by bedding
Exploration
drillers placed sections of drill core in a two-foot long core box for
transportation and storage. These pieces came out of the core barrel in a
various lengths. An unknown number of core pieces had to be broken to fit in
the boxes; resulting in a shortening bias of length. However, a great deal of
detail concerning bedding thickness is available for beds less than two-feet in
thickness.
To relate core lengths to size distributions in run of mine rock involves a number of assumptions. If beam bending is used as the failure mode of blasted rock, then one might assume that bed thickness represents the shortest dimension of a fragment. Casual observation of muck piles seems to support this, as bedding surfaces are easy to identify in the Biwabik Iron Formation. Using an aspect ratio of 1:2:3, the two-foot long core limit translates to a maximum dimension of 6 feet. Six feet or less would account for all but a small fraction of run of mine rock. However, core pieces are broken to even shorter lengths to fill out core boxes. Because of intentionally broken core, geologic fragmentation should be augmented with other field data. The attached plot (Fig. 1) compares core lengths to resulting fragmentation measured by digital image analysis at the primary crusher. The thinner bedded areas result in better fragmentation.
Page 4
Penetration rate vs. bed thickness
Yin
& Thompson have led the implementation of the Stratalogger™ drill
monitoring system in iron ore. Dance has described the implementation of the
Aquila™ system in open pit copper. Without a firm understanding in changes in
geology and ore characteristics; optimization efforts are difficult at best. In
flat-lying ore deposits, such as the Mesabi ores, closely spaced bedding planes
can profoundly increase drill penetration rates. This may be due to spalling,
as the tungsten carbide projection on the bit breaks larger than normal pieces
as the weak bedding plane is encountered. If the piece is yet small enough to
pass up the annular space between the drill steel and the wall of the hole;
energy is saved that would otherwise be wasted in creating additional surface
area in the cuttings, rather than advancing the hole. In this manner,
penetration rate can be controlled by mean bed thickness. Mill operators are
interested in mean bed thickness as it, in turn, has a dominating effect on ROM
size distribution.
Figure 2
SAG energy vs. powder factor
Reduction in capital costs and reduced operating costs through the reduction or wholesale elimination of grinding media led to adoption of SAG and FAG flowsheets in certain iron ore plants Recent developments
Page 5
in the understanding of blast through mill size reduction call into question the actual economies of SAG/FAG mills. Figure 2 compares blast energy to electrical consumption in the SAG mill. The concern at this operation was that excessive blast energy was depleting the source of media rock. Although that problem has been documented for some types of ore; excessive blast energy does not appear to be the culprit in this case.. Note that local maximums and minimums of powder factor and SAG mill energy actually run counter-cyclical.
Figure 3
Consumption of grinding media
Figure
3 compares rod consumption to powder factor. Each point represents one year of
operations in which over 50 million tons of feed were processed. Although, this
is example is from conventional grinding; this phenomenon of rock weakening
through higher powder factors is starkly apparent. This example suggests that
the cost of higher powder factors may be fully offset by rods alone.
Powder factor vs. Total cost
The assumption underlying the upper and lower
envelopes in figure 4 is that an inverse relationship does exist between blasting and downstream costs and that the
purpose of the plot is to shed light on the slope of that line. The broad
scatter is due to seasonal changes, geological variations and large purchases in one month. Validation will
require higher powder factor tests in the lower right area of figure 4.
|
Figure 4 |
Page 6
Figure 5
Page 7
Total size reduction costs
Figure
5 summarizes the distribution of costs in producing concentrate. The
significance of this pie chart lies in the relative costs involved in drilling
and blasting versus those for grinding. Drill and blast costs have been reduced
to where they represent about 13% of concentrate costs. The question is whether
minimizing individual cost leads to an overall minimum.
Management Structures
Close
communication between blast and mill engineers has resulted in breakthroughs in
process operations. In the previously mentioned Red Dog example, it is
significant that the mill and blast engineers are married to one another. In
other instances cited herein, the mill engineer took a special interest in
learning more about blast operations. The authors experience is one where mill
engineers were highly cooperative and willing to openly discuss problems. This
is often no the case, where competition for budgets and blame for ore grade
issues between mining and milling personal curtails constructive dialog.
Cutting the budget for mill operators is often the only means to free up moneys
to test higher powder factors. Once it becomes ‘us against them’; optimization
falls by the wayside.
Conclusions
1)
Greenfield
operations will need to understand the equation relating operating costs and
capital costs of conventional vs. SAG/FAG flowsheets.
2)
Fragmentation
costs may favor blasting over milling wherever feasible.
3)
The
growing shortfall of electrical demand may be eased
through implementation of
increased powder factors.
4)
Blasting
will not create larger pieces of media rock where fissile bedding and highly
jointed predominate.
5)
Drill
monitoring and video image analysis are the two key technologies for optimizing
mine through mill efficiencies.
6)
Blast
engineers and mill engineers need to work together.
Page 8
References
F.C. Bond, 1951, “The Third
Theory of Comminution” Meeting of AIME
in Mexico City, October 1951 in Mining Engineering, May 1952, pp 484-494
A.
Bouajila, G. Bartolacci, N. Kock, J Cayouette and C. Cote, "Toward the
Improvement of Primary Grinding Productivity and Energy Efficiency. Part 1;
Investigation of the Feed Ore Size Effect”
IFAC Workshop, Future trends in automation in mineral and metal
processing (MM'2000), Finland 22-24 August 2000, pp 280-285.
B. Duevel, 1994, “Preliminary Analysis of Drill Core
Slides”, unpublished report to author.
J.W. Eloranta, 1992, “Cap
Testing at the Minntac Mine” proceedings of the eighteenth conference of
Explosives and Blasting Technique, Orlando, Florida, January 19-23,1992.
International Society of Explosives Engineers, Cleveland, Ohio, pp 205-211
J.W. Eloranta, 1993,
“Practical Blast Evaluation at The Minntac Mine” proceedings of the nineteenth
conference of Explosives and Blasting Technique, San Diego, California January
31-February 4, 1993. International Society of Explosives Engineers, Cleveland,
Ohio, pp 101-107
J.W. Eloranta, 1994,
“Stemming Selection for Large-Diameter Blast Holes” proceedings of the
twentieth conference of Explosives and Blasting Technique, Austin, Texas,
January 30-February 3,1994. International Society of Explosives Engineers,
Cleveland, Ohio, pp 255-272
J.W. Eloranta, 1995, The
Effect of Fragmentation on Downstream Processing Costs. Proceedings of Explo95
Conference, Brisbane, Qld, Australia, Sept. 4-7, 1995 pp 25-28
J.W. Eloranta, 1996,
“Geologic Data for Blasting” Proceedings of the twenty-second conference of
Explosives and Blasting Technique, Orlando, Florida, February 4-8, 1996.
International Society of Explosives Engineers, Cleveland, Ohio, pp 107-117
M. C. Fuerstenau, G. Chi,
and R. C. Bradt, 1995, “Optimization of Energy Utilization and Production Costs
in Mining and Ore Preparation” XIX International Mineral Processing Congress,
San Francisco, California. Oct. pp 161-164
Page 9
S. Grannes, et al, 1994,
“Recent Advances in Digital Image Analysis” Proceedings of The 67th Annual
Meeting of the Minnesota Section of the SME. pp 105-116, Duluth, Minnesota,
1994
C. Hendricks, J. Peck, M.
Scoble, 1992, “An Automated Approach to
Blast Optimization Through Performance Monitoring of Blast Hole Drills and
Mining Shovels” 4th High Tech Seminar on Blasting, Nashville, Tennessee June
1992
R. T. Hukki, 1975, The
Principles of Comminution: an Analytical Summary, Engineering and Mining
Journal, Vol. 176 pp 106-110
F. Kick, “Das Gesetz der proporionalem Widerstand und
Siene Anwendung” Leipzig, 1885
G. E. LeJuge and N. Cox,
1995, ”The Impact of Explosive Performance on Quarry Fragmentation” Proceedings
of Explo95 Conference, Brisbane, Qld, Australia, Sept. 4 - 7, 1995 pp 445-452
Lownds, C. 1995, “Prediction
of Fragmentation Based on Energy Distribution of Explosives” Proceedings of the
twenty-first conference of Explosives and Blasting Technique, Nashville,
Tennessee, February 5-9, 1995. International Society of Explosives Engineers,
Cleveland, Ohio, pp 286-297
Lownds, C. 1996, “Derivation
of Fracture Intensity from Measured Fragmentation” Proceedings of the
twenty-second conference of Explosives and Blasting Technique, Orlando,
Florida, February 4-8, 1996. International Society of Explosives Engineers,
Cleveland, Ohio,
McCarter, M. K. 1996,
“Effect of Blast Preconditioning on Comminution for Selected Rock Types”
Proceedings of the twenty-second conference of Explosives and Blasting
Technique, Orlando, Florida, February 4-8, 1996. International Society of Explosives Engineers, Cleveland, Ohio,
pp 119-129
L. Moody, C. Cunningham, and
H. Lourens. 1996, “Measuring the effect of blasting fragmentation on hard rock
quarrying operations “ Proceedings of FRAGBLAST5, Fragmentation by Blasting pp
353-359, Montreal, Quebec, Canada
August 25 - 29, 1996
K. Nielsen, and J.
Kristiansen, 1995. “Blasting and Grinding - An Integrated Comminution System
“ Proceedings of Explo95 Conference pp
113-117, Brisbane, Qld, Australia, Sept. 4 - 7, 1995
Page 10
K. Nielsen, and J.
Kristiansen, 1996. Blasting- Crushing-
Grinding: Optimisation of an
Integrated Comminution System. Proceedings of FRAGBLAST5, Fragmentation by
Blasting pp 269-277, Montreal, Quebec, Canada
August 25 - 29, 1996
S.
Morrell
et al, 1992, “The Prediction of Power Draw in Comminution Machines “
Comminution - Theory and Practice, pp 233-247. International Society of Mining
Engineers, Littleton, Colorado
Pastika et al, 1995,
“Improved Fragmentation for Mine Cost Reduction. Proceedings of the 68th
Annual meeting of the Minnesota Section of the AIME and 56th
Symposium, Duluth, MN pp185-192
V. I. Revnivtsev, 1988, “We
really Need a Revolution in Comminution “ Proceedings of XVI International
Minerals Processing Congress, pp 93-114
P. R., Rittinger, 1867,
Lehrbauch der Aufbereitungskunde. Berlin, 1867
C. L. Schneider, and R. P.
King, 1995, “A Comprehensive Simulation of an Industrial Comminution Circuit
Treating Taconite “ XIX International Mineral Processing Congress, San
Francisco, California. Oct. 1995
V. P. Tarasenko, 1996,
“Controlling the patterns of fragmentation in blasting and mechanical crushing
operations“ Proceedings of FRAGBLAST5, Fragmentation by Blasting pp 293-296,
Montreal, Quebec, Canada August 25 -
29, 1996
D. Thompson, 1999, “Modern
Drilling and Blasting Technology at the Minntac Mine” Proceedings of Minnblast
99 Surface Blasting Conference pp 77-96, June 7-11, 1999, Duluth Minnesota,
Available from ISEE Cleveland, OH
R. Simkus and A. Dance,
“Tracking hardness and Size: Measuring and Monitoring ROM Ore Properties at
Highland Valley Copper “ Mine to Mill 1998 - AusIMM Conference, October 11-14,
1998, Brisbane Australia
B. A. Willis, 1988 “Enhancement of Mineral Liberation “, Proceedings of XVI International Minerals Processing Congress, pp 293-297
K. Yin, H. Liu and Yang, 2000, “Extracting
information from drill data”, Fragblast Journal 4(2000), A. Balkema, Postbus
1675, Rotterdam, Netherlands pp. 83-99