AUTOGENOUS MILLS

In this section, we make extensive use of Bond’s review of early autogenous grinding sys­tems, using his terminology for autogenous grinding (Bond 1964).

Pebble milling or secondary autogenous grinding is the process in which feed pass­ing 19 mm or finer is ground by pebbles of ore with a maximum size of 76 mm or more that are extracted by screens from a coarse rock stream. This process was the first type of autogenous grinding used in plants.

Primary autogenous grinding is the process in which run-of-underground-mine ore or the product from primary crushers is the feed for mills. Both dry and wet primary autogenous grinding processes are used.

SAG grinding is the process in which 6%-12% of the primary mill volume is occu­pied by 100-150 mm (4-6 in.) in diameter steel balls to break the “critical-size” pebbles— the ones too large to be broken by the large rocks and too small to break finer particles. SAG milling has become a generic term to describe all types of primary mills that use rock or a mixture of rocks and balls to break rock.

The ABC (autogenous-ball-crusher) circuit is a particular type of SAG mill circuit that involves removing the critical-size particles from the SAG mill, screening them, crushing the screen oversize, returning the product to the SAG mill, and sending the screen undersize to the ball mill. Figure 7.20 shows a flow sheet for an ABC circuit.

Bond developed a media competency test which involved 100-160 mm rocks being rotated in a 1.93 x 0.32 m drum for a set time, sizing the product and testing the size fractions for grindability and resistance to impact. There is now little interest in this test, because media competency is not an issue with the SAG mills that are common today.

The first tumbling mills used hard pebbles from the beaches of France and Denmark on the North Sea to do the grinding. These pebbles were expensive, but the alternative grinding media was steel balls that were even more expensive, although they were heavier and broke particles faster. The term autogenous was apparently first used by Hard — inge in 1940 during a discussion with mill workers about a mill he had designed that did not use steel balls (Robinson 1980). Autogenous grinding has since been defined as the “…grinding of ore by itself rather than by special grinding bodies distinct from the ore” (Bond 1964).

Autogenous grinding started in a small way at a gold mine on the Rand in South Africa in 1907. Tube mills used in the cement industry were being introduced then into South Africa to grind the stamp mill discharge finer for cyanidation, and a metallurgist decided to compare the grinding properties of pebbles made from the ore with pebbles from Denmark. The early tests showed that pebbles made from ore were successful, and autogenous grinding has been part of metallurgical practice on the Rand ever since. The process worked because the ore was a hard conglomerate cemented by silica and the conglomerate made ideal pebbles. The deposit itself was relatively homogeneous, which is always an advantage for autogenous grinding.

AUTOGENOUS MILLS

FIGURE 7.20 Flow sheet for an ABC circuit (JKMRC 2002; reprinted by permission from University of Queensland)

When tumbling mills were first used to grind gold ores in South Africa and Austra­lia, the wear rate of the grinding media (the pebbles from the beaches of Denmark or France) was remarkably low at 2 kg/ton. But the cost of the pebbles was high at $5/ton, providing an incentive to develop pebble milling. In 1907, pebbles were used as grinding media on two occasions and these were the start of autogenous milling.

K. L. Graham at the Geldenhuis Deep Mine near Johannesburg ran a test using two mills 1.7 m in diameter and 7.1 m long to compare Danish pebbles and pieces of gold ore as grinding media. The 81-day test showed little variation between them. Engineers from other mines on the Rand verified the results in their own mills, and further experiments showed that 125-mm pebbles could be used to grind -0.5-in. feed. As a result, pebble milling became common practice in South Africa and was supported by development work. By 1932, SAG grinding was being used with up to 33% of 75-mm balls being added to the rock pebbles.

Hardinge, inventor of the conical mill and a later inventor and manufacturer of wet and dry autogenous mills, described a pebble mill at an AIME meeting:

(The feed) was 1 inch and finer. The mill contained 2000 pounds of 2 and 3 inch pebbles. The ore was crushed to 80 mesh size at the rate of 4 tons per hour at an expenditure of 17 hp. This type of mill appears to be specially adapted to the use of lumps of ore instead of the usual foreign bodies for grinding. (Hardinge 1955)

The crops of birds filled with small pebbles, successfully break up whole grains pre­paratory to digestion. This is the natural prototype of the pebble mill. (Fischer 1944)

Pebble milling was tested in North and Central America from 1912 to 1917 with results that were regarded as satisfactory at the time, but the idea did not catch on as it did in South Africa. Perhaps “the dam of professional hesitancy” (Robinson 1980) was a factor. Unlike South Africa, where many concentrators were clustered around Johannes­burg, the concentrators in North and Central America were widely scattered and metal­lurgists had few opportunities to discuss ideas with their peers. Understandably, then, they were cautious about a major change that might go wrong and shut down a produc­tion system. But some early tests did give promising results, and it was established that pebble milling was successful for a variety of ore types. For some reason, however, the

momentum was lost and little happened with wet pebble milling in North America for the next 30 years.

The difference between ball milling and pebble milling is that a ball mill with a ball charge weighing 4,700 kg/m3 and occupying 40% of the mill volume draws approxi­mately twice the power drawn by the same mill with a 40% charge of pebbles weighing 1,500-2,300 kg/m3. The ball mill has a correspondingly higher capacity for the same reduction ratio. Converting pebble mills to ball mills will give higher plant capacity and requires only that the mill drive be changed to deliver the power needed.

Dry pebble milling was investigated in 1938 by the American Nepheline Company, which operated a dry 2.4 x 0.9 m conical mill to grind 4.5 tph of -7 mm syenite to -1 mm using 75-mm pieces of syenite as the grinding medium. The circuit was successful, and the use of dry grinding opened up questions about the merits of dry and wet grinding, gravity discharge, and air sweeping. These questions would only be answered by years of test work in plants.

Bunting Crocker, a metallurgist at Lake Shore Mines in Ontario, gave new impetus to pebble milling in 1948. The gold ore being processed at Lake Shore required grinding to 80% passing 30 pm, and steel balls (20 mm) were expensive. Crocker converted a mill (1.6 m x 5.2 m) to a pebble mill and operated it successfully. This led to the conversion of eight mills (2 m x 5 m) to pebble mills. He used 3-4 tpd of 75 x 50 mm lumps of screened ore as grinding media, and each mill ground 100-120 tpd. The rock media had a much lower cost than the steel balls, and the successful experiment created confidence in the use of pebble mills, particularly in the Canadian gold and uranium companies.

Pebble milling has been used continuously in South Africa from its introduction in 1907, with many improvements. By 1960, common practice was to use rod or ball mills as the first stage of grinding and pebble mills up to 3.9 m in diameter and 5.2 m long as the second stage. Metallurgists then became interested in the extension of the technol­ogy to primary autogenous grinding, which will be discussed in the next section.

In the 1960s in the United States, rod mill-pebble mill circuits were installed at the Anaconda Company’s Kelly plant in Montana and Twin Buttes plant in Arizona. Pebble mills in the Kelly plant were 3.85 m in diameter by 6.45 m long, and in the Twin Buttes plant, they were 4.7 m in diameter by 9 m long. There were problems at each plant because the pebbles fed to the mill were not “competent” pebbles; that is, they were not satisfactory grinding media.

At one installation the nature of the ore was too fragile, the crushing plant did not produce enough pebbles, and the pebbles it did produce immediately broke in the 3.85-m (12.5-ft) diameter mills.

At the other installation pilot plant testing had shown that the ore made good peb­bles. The crushing plant produced an adequate amount of properly sized pebbles. The pebbles were screened out of the ore in the crushing section of the plant and were then conveyed to the pebble storage area where they were dropped onto the pebble stockpile from a 19.4-m (60-ft) high stacking conveyor. There was breakage of the pebbles, and about 50% of the pebbles fed to the 4.7-m diameter mills were smaller than 25 mm and too small to be suitable media.

In the mid-1960s, the Duval Corporation installed four 6.45-m (20-ft) diameter mills in its copper ore concentrator near Kingman, Arizona. They were designed as inter­mediate autogenous mills to replace rod mills, and their feed was tertiary crusher prod­uct. The grinding media were 100-mm (+4-in.) products from a primary crusher but the circuit failed for lack of sufficient ore media, so it was converted to a conventional three — stage crushing circuit. Because of the structural design of the mill and the lack of power, the ball charge was limited to 29% of the mill volume. These were the first balls mills (6.45 m [20-ft] in diameter) used to grind copper ore. In 1967, an intermediate autogenous mill (2.3 m in diameter) was installed at the Western Mining Corporation nickel ore plant in Kambalda, western Australia, and it operated successfully for years, but interest in this form of autogenous grinding has faded.

Updated: 24.03.2016 — 12:06