SAG MILL GRINDING DESIGN VS GEOMETALLURGY - GETTING IT RIGHT FOR COMPETENT ORES by J.H. Starkey, Paul Scinto.  Presented at IMPC XXV Congress.  Brisbane, Australia  Sept. 9, 2010.

SAG kWh/t MEASURED USING A STANDARD TEST - 53 Projects in 6 Years by J.H. Starkey, Paul Scinto.  Presented at Comminution 2010 in Cape Town, South Africa April 12-15, 2010

Abstract - At SAG 2006, the SAGDesign Test was introduced. After 6 years and 50 successful mill design and 3 accurate benchmark projects, this technology has emerged as a proven mill design method based on the direct measurement of SAG pinion energy in kWh/t using standard commercial grinding conditions for the test in a 0.5 m diameter SAG mill.

SUCCESSFUL DESIGN OF THE NICO GRINDING CIRCUIT FOR UNUSUALLY HARD ORE by J.H. Starkey, Michael Samuels.  Presented at the 2010 CMP in Ottawa, ON January 19-21, 2010

Abstract - This paper examines test results and grinding mill design for the recent Fortune Minerals’ NICO Project, located in the Northwest Territories. SAG milling and rod milling options are compared. Rod milling is clearly shown to be the best option because of the ore’s extreme resistance to breakage in a SAG mill environment.

Surprisingly, the Bond Ball Mill Wi was 13 kWh/t for this ore. The Bond Rod Mill Wi was 20 kWh/t, whereas the SAG pinion energy to grind the ore to T80 1.7 mm was 30 kWh/t. The existence of ore with this ratio of SAG hardness to Bond BM Wi has not been previously observed. If not considered, this critical relationship may adversely impact the accuracy of mill design. This paper substantiates the significance of these findings to achieve successful grinding design on ore that is macro-competent during primary size reduction, but relatively normal during secondary grinding.

IMPACT on GRINDING MILL DESIGN on RECENT NEW DISCOVERIES by J.H. Starkey.  Presented at the SME 2009 in Tucson, AZ OCT 6, 2009

Abstract - At the CIM conference in May 2009, previously unseen differences between macro and micro grindability relationships contained in the SAGDesign database were presented.  This database included SAG and ball mill design energy measurements on many samples.  These were compared with the extremes of SAG hardness variation from two recent projects.  These extremes demand different considerations to be used in the design of the SAG and ball mills needed for these ores.  The term ‘new discoveries’, relates to the wide variation revealed by the new measurements.  Based on those findings it is now evident that historic grindability measurements only reveal the information that the tests were designed to reveal.  Bond Rod Mill, Ball Mill, and Crushing Work Index tests have been studied for years to find how to get a good SAG mill design from these (and other) measurements.  These efforts have been partially successful.  Other researchers have used breakage parameters to develop SAG mill design data.  These are not rotary grinding tests but do give a good estimation of the relative point hardness in an ore body with respect to SAG hardness.  These breakage tests do not include ball mill grinding data.  Based on the 2009 CIM paper (Starkey and Samuels, 2009), the ratio of SAG to ball mill grindability is an important factor in achieving a reliable SAG mill circuit design because that ratio reveals the best and most practical split for SAG and ball mill energy for the ore being tested.  The reason the SAGDesign test gives an acceptable grinding circuit design result for any ore relates to the 19mm F80 of the ore charge fed to the SAGDesign SAG mill and to the test procedure which includes a Bond Ball Mill Work Index test on SAG ground ore from the first stage of the test.  The ratio of SAG pinion energy (to grind to T80 1.7mm) to Bond BM Wi is a key parameter in a SAG mill grinding circuit design, because this ratio controls the power split between SAG and ball mills at the design T80 (transfer size). 

IMPACT on MILL DESIGN and FLOTATION CONTROL of NEW DISCOVERIES in the RELATIONSHIP BETWEEN MACRO and MICRO GRINDABILITY by J.H. Starkey.  Presented at COM 2009 in Sudbury, ON AUG 26, 2009

Abstract - Large differences between macro and micro grindability relationships have been discovered for mill design measurements using the patented SAGDesign^TM test.  Crushing, Rod and Ball Mill Wi tests, and breakage parameter measurements have previously been used to design SAG mills but have not revealed the true magnitude of these differences.  A SAGDesign test measures the SAG/Bond Wi Ratio (defined below) for every sample tested and gives a good design result because the F80 of the SAG stage test feed is 19mm and because the test includes a Bond Ball Mill Work Index on SAG ground ore. Good flotation operation demands properly sized grinding mills and steady feed to the flotation circuit.

New Discoveries in the Relationship between Macro and Micro Grindability by J.H. Starkey, M. Samuels, Fortune Minerals Ltd.  Presented at the CIM in Toronto May 10-13, 2009.

Abstract - This paper examines 5 years of SAGDesign testing data with particular emphasis on the grinding test results from two recent projects, the Fortune Minerals Limited NICO project, and the Aurora Energy Resources Inc., Michelin deposits, located in the Northwest Territories and in Labrador respectively. The context is to compare these results with the SAGDesign database. The required SAG mill pinion energy for these two ores showed vast differences, even though both ores had similar average Bond Ball Mill Work Index values - about 13 kWh/t each for the two deposits respectively. Bond Rod Mill Work Index values for NICO ranged up to 19.7 kWh/t for the hardest ore, confirming the harder nature of the coarse material. SAG mill pinion energy to grind the NICO cobalt-gold-bismuth-copper ore from F80 152 mm to T80 1.7mm varied from 16 to 35 kWh/t, while for Michelin uranium ore, it varied from 2 to 9 kWh/t. NICO ore is harder than any other material encountered to date using SAGDesign testing. SAG pinion energy is more than double the ball mill pinion energy using a transfer size T80 of 1.7mm and a final grind P80 of 74 microns. For NICO, the ore is ‘off the charts’ for simulation design techniques because there is no comparable operating SAG mill treating ore that is this macro-competent. For Aurora, the relationship between SAG and ball mill pinion energy is reversed in that the SAG mill requires less than half of the ball mill pinion energy using a transfer size T80 of 1.7mm and a final grind P80 of 90 microns.

SAGDesign Testing Review - Case Studies by J.H. Starkey, D. Meadows, P. Thompson, A. Senchenko.  Presented at the IMPC XXIV in Beijing September 24-28, 2008.

Abstract - SAGDesign Testing for grinding mill design was introduced commercially in 2004. Up to January 2008, 18 projects have been completed and 99 samples have been tested using this method. Two papers have been published describing the patented test procedure and one of these shows how the results compare with other grinding test methods. This paper will briefly discuss the theory on which this technology is based, and explain the procedures used to develop the calibration equation. Case studies are also included to demonstrate the effectiveness of the test work being done.  These case studies show how the SAGDesign test is being used today: to design new grinding circuits for commercial full scale use; to calculate throughput in existing mills for production planning; and to accurately define grinding mill requirements at the scoping, pre-feasibility or feasibility study stage of project development, in order to economically identify the size and cost of the grinding mills that will be needed for the project. This is important because if the full recommended SAGDesign testing program is done at an early stage in the project, the definitive cost for a guaranteed throughput, grinding mill design (by Outotec if their mills are purchased) is available.

Successfully Going Global by C. Hersey.  CIM Magazine December 2007/January 2008

Abstract - Who doesn't love a story about a small Canadian company making it big?  The leap to becoming a global company is never easy and is guaranteed to come with its pros and cons, benefits and sacrifices, both good and bad alike.  In John Starkey's case, both owner and principal consulting engineer for Starkey & Associates, going global has had very satisfying results.

A Comparison of Ore Hardness Measurements for Grinding Design for the Tenke Project* by John Starkey and Dave Meadows, Proceedings of the CMP Conference, Ottawa, January 2007.

SAGDesign Testing - What is it and Why it Works, by John Starkey, Sami Hindstrom and George Nadasdy. Presented at SAG 2006 in Vancouver, October 2006

Abstract - A robust, accurate laboratory SAG test has been needed for years. The SAGDesign Consulting Group has developed such a test. This paper also presents case studies in how it is used. Feed is prepared from ~10 kg of half core samples by crushing to 80% passing 19-mm. Grinding is then done in a 0.5 m diameter SAG mill to 80% passing 12 mesh US Standard (1.7 mm). The SAG ground product is then used for a Bond BM Work Index test. Reproducibility for SAG grinding is within 3% for duplicate tests done to date. The test has been used for predicting throughput as well as new plant design. 

Accurate, Economical Grinding Design Using SPI and Bond, by John Starkey. Presented at the IMPC XXII Congress in Cape Town, October 2003.

Choosing a SAG Mill to Achieve Design Performance, by John Starkey, Sami Hindstrom and Travis Orser. Presented at the Canadian Mineral Processors Conference in Ottawa, January 2003.

Grinding of Agrium Phosphate Ore in a 3 Ft. Diameter Pilot SAG Mill, by John Starkey, Rene Carre, Richard May, and Lucky Lakshmanan. Presented at the Canadian Mineral Processors Conference in Ottawa, January 2002.

Design of the Agnico-Eagle Laronde Division SAG Mill, by John Starkey, Jean Robitaille, Paul Cousin, James Jordan, & Glenn Kosick. Presented at the SAG Conference in Vancouver, October 2001.

A Critical Review of SAG Design Procedures for the 21st Century, by John Starkey. Presentation was made at. Comminution 01 in Brisbane, Australia in March 2001. A decision was made not to publish this information. It is available as a Professional Development Seminar.

Grinding Circuit Design at Kubaka Using SPI and Bond, by John Starkey and Graham Holmes. Presented at the Canadian Mineral Processors Conference in Ottawa, January 23, 2001.

Getting More From Drill Core - Preliminary SAG Design, by John Starkey. Presented at the Randol Gold Forum, Monterey CA, May 1997.

Getting More Out of Your Drill Core, by John Starkey. Presented at the Annual Meeting of the CIM, Vancouver BC, April 1997.

Application of the Minnovex SAG Power Index at Five Canadian SAG Plants, by John Starkey and Glenn Dobby. Presented at the SAG Conference, Vancouver BC, October 1996.

Water Usage in the Carol Concentrator, by J.H. Starkey and S.E. El-Alfy. Presented at the Canadian Mineral Processors Conference, January 1995.

A New Tool for SAG Hardness Testing, by John Starkey, Glenn Dobby and Glenn Kosick. Presented at the Canadian Mineral Processors Conference, January 1994.

Modernization of the Campbell Red Lake Mill in 1986, by J. Starkey and S. Roberts. Presented at the CIM Gold Conference, Winnipeg, August. 1987.