Search Documents

By Jack White

W. I. Garms-—The authors state that when they added 11 tons of balls to the 45 pct volume ball load, the power needle did not budge. The question arises as to whether any increase in capacity accompanied the 11-ton addition. It is hardly conceivable that all of the balls making up the l l-ton addition were sterile. However, as the work done in the ball mill is not by kilo-watthours per se or balls per se but by kilowatthour-balls, it is possible that the kilowatthourballs in the mill before the ll-ton addition had a different amperage, voltage or robustage than those in the mill after the addition and that the overall outcome was that equal work was done before and after the addition. The authors can keep the efficiency that they got with the 45 pct volume 1-in. ball charge and get additional capacity by putting more kilowatthourballs to work in the form of the addition of 11 tons of 2-in. balls to the present ball charge, and subsequent makeup additions by weight of 20 pct 2-in. balls and 80 pct 1-in. balls. J. F. Myers (authors' reply)-—Mr. Garms points his finger at the 11 tons of "sterile" balls added to the Tri-cone mill. A fact so interesting to us that we felt it should be reported. Clearly, ball slippage at the existing mill speed prevented any further transfer of power from the shell to the ball mass. We agree that the kilowatthourballs would increase, were we charging rough 2-in. balls as suggested by Mr. Garms. The assumption is, of course, that we would get a corresponding increase in work done. Our reported data (CIME meeting, Vancouver, 1946) shows a gain of work accomplished of 7.6 pct as we decreased the ball size from 2 in. to 1% in. and a further gain of work accomplished of 2.1 pct when we went from 1 % in. to 1 in. The small ball results reported by H. R. Banks at Chapman Camp are also very convincing. In the case of 2-in. balls they would take the conventional ball path and start cascading. They would thus destroy the classifying pool over the "foot" of the

Jan 1, 1951

By Jack White

W. I. Garms-—The authors state that when they added 11 tons of balls to the 45 pct volume ball load, the power needle did not budge. The question arises as to whether any increase in capacity accompanied the 11-ton addition. It is hardly conceivable that all of the balls making up the l l-ton addition were sterile. However, as the work done in the ball mill is not by kilo-watthours per se or balls per se but by kilowatthour-balls, it is possible that the kilowatthourballs in the mill before the ll-ton addition had a different amperage, voltage or robustage than those in the mill after the addition and that the overall outcome was that equal work was done before and after the addition. The authors can keep the efficiency that they got with the 45 pct volume 1-in. ball charge and get additional capacity by putting more kilowatthourballs to work in the form of the addition of 11 tons of 2-in. balls to the present ball charge, and subsequent makeup additions by weight of 20 pct 2-in. balls and 80 pct 1-in. balls. J. F. Myers (authors' reply)-—Mr. Garms points his finger at the 11 tons of "sterile" balls added to the Tri-cone mill. A fact so interesting to us that we felt it should be reported. Clearly, ball slippage at the existing mill speed prevented any further transfer of power from the shell to the ball mass. We agree that the kilowatthourballs would increase, were we charging rough 2-in. balls as suggested by Mr. Garms. The assumption is, of course, that we would get a corresponding increase in work done. Our reported data (CIME meeting, Vancouver, 1946) shows a gain of work accomplished of 7.6 pct as we decreased the ball size from 2 in. to 1% in. and a further gain of work accomplished of 2.1 pct when we went from 1 % in. to 1 in. The small ball results reported by H. R. Banks at Chapman Camp are also very convincing. In the case of 2-in. balls they would take the conventional ball path and start cascading. They would thus destroy the classifying pool over the "foot" of the

Jan 1, 1951

By E. B. Fitch, E. C. Johnson

The operating behavior of liquid-solid cyclones is outlined, together with the nature and range of the process results obtainable, to serve as a background for engineers wishing to consider application of this new process tool. BY now most engineers are familiar with the liquid-solid or Dutch State Mines cyclone. However, it should be helpful to know exactly what it is that the equipment does and what its limits are. Without going into cyclone theory, this paper will describe the operating characteristics of Dutch State Mines cyclones. These are manufactured under license in this country and sold under the trade-marked name of DorrClone. The physical construction of the liquid-solid cyclone has been covered in many papers,'-' the DorrClone in particular being described in some detail by Weems. Fig. 1 shows the unit in cross-section. The feed enters at C. The coarse, heavy particles are thrown centrifugally to the periphery and make their way down the wall to the apex where their rate of discharge as underflow is controlled by an adjustable rubber apex valve. As the apex diameter is decreased the solids build up behind the valve, producing a denser underflow. Meanwhile the fine particles are swept into the upward flowing vortex stream which exits as overflow through the vortex finder, F. Flexibility to produce the specific result desired in a particular process is achieved by providing means for varying the areas of the entrance, vortex discharge, and apex discharge. The entrance area may be varied by insertion of special shims. Vortex discharge area may be changed by use of different-sized vortex finders which are interchangeable. Similarly, the different sizes of apex valves are interchangeable and in addition each apex valve is variable down to about 60 pct of its maximum diameter. A most significant primary distinction to make is that although liquid-solid cyclones have been sometimes called thickeners, they actually are classifiers, and very potent ones. They are almost never thickeners in the special sense that many metallurgical engineers understand the term. There would be no profit in quibbling over the definition of a word, but when the application of cyclones is considered, it will help to understand the difference be- tween two mechanisms, one of which will be called classification, and the other thickening. In what is called thickening the fine solid particles present in the feed hold together by surface attraction during the sedimentation process. The loose network of particles thus held together constrains all particles to settle at approximately the same rate, the larger ones dragging the smaller ones down. As a result, pulp settles with a sharp line of demarcation between solids and a relatively clear supernatant liquid. Essentially all the solids, regardless of their fineness, pass into the thickened underflow, and a clear overflow is separated. In classification, on the other hand, the interparti-cle forces are relatively insignificant as compared to the settling force on the individual particles, and are insufficient to prevent independent movement of the particles. The coarsest, heaviest particles settle most rapidly through the pulp, passing more slowly settling fines. Particles coarser than the mesh-of-separation essentially all settle into the underflow, but if the feed contains any particles finer than the mesh of separation, at least part of them will appear in the overflow. A clear supernatant or overflow can be obtained only if there are no undersize particles present in the feed. Thus it will be seen that classification is impossible under ideal thickening conditions. The finer particles are pulled down at essentially the same rate as coarser particles, and there is no separation on the basis of particle size. The surface attraction holding the particles together in a thickener is usually feeble. Whenever the sedimentation force on any particle is strong with respect to the interparticle forces, that particle can pass through the tenuous structure and settle independently. There are at least four ways of making the sedimentation force strong, with respect to the interparticle forces, and obtaining classification. First, and most obvious, the particles may be large and heavy. Thus coarse sands settle out in a beaker or Dorr thickener ahead of the rest of the thickening solids. Second, the interparticle forces may be altered by physicochemical means; i.e., it is often necessary to add dispersing agents to destroy the interparticle forces and permit classification to take place. Third, the interparticle forces may be reduced by dilution of pulp. It is well known that to obtain the most efficient separation of

Jan 1, 1954

By J. W. Snavely

THE high tension belt conveyor is introducing a new and tremendously expanded era of low cost bulk material handling. High tension belt conveyors are generally those installations involving very long centers, high lifts, or drops, in which the belts are stressed up to their maximum tension values, and further, where the belt construction provides tension capacity far beyond what is possible with conventional belt constructions. With these high tension installations, the magnitude of the forces involved demands careful refinement of accepted design practice in order to achieve optimum balance of all factors. No attempt will be made to evaluate the relative merits of belt conveyor haulage with other means of transportation. For present purposes, it is assumed this has already been done in favor of belt conveyor. Neither will any attempt be made to evaluate the various conveyor belt constructions now available or to balance the advantages of various types of mechanical equipment. It is also assumed that the basic haulage information on which the conveyor design is based is accurate and complete. A sustained maximum, uniform load on the belt at all times must be achieved through proper feed control and the use of adequate surge storage to level the peaks and valleys of any varying demand for the material being handled. General Belt Capacity Considerations The belt conveyor capacity tables published by various belting and conveyor equipment manufacturers vary to a considerable degree, and the ratings given are quite conservative. Of necessity, these published ratings are based on the handling of average materials under average conditions. In applying a high tension belt, all possible capacity from the belt must be obtained in order to hold its width to a minimum and thereby limit the initial cost. Two factors are involved, loading to maximum cross section area and traveling at a maximum practical speed. Belt Loading: Proper treatment of the loading of the belt will result in maximum cross section to the load, and published capacity ratings can be exceeded, sometimes by appreciable margins. On the 10-mile conveyor haul used in the construction of Shasta Dam, California, although the rated capacity of the belt line was 1100 tons per hr, at times the system handled peak loads of 1400 tons per hr, almost 25 pct better than the rated capacity. One of the large coal companies has been able to exceed rated capacity by as much as 50 pct. Loading conditions which must be controlled are: 1. Large lumps must be scalped off and rejected or the load must be primary crushed before being placed on the belt. 2. The material weight per cubic foot must be accurate, must be known for all the materials being handled, and must be known for the complete range of conditions of the individual material being handled. Long centers and high lifts magnify small differences into serious proportions. 3. Uniform feeding to the belt is most important. Various types of feeders are available, which can be used to place a constant predetermined volume of material on the belt, or, where an appreciable range of material weight exists, through electrical control actuated by current demand, to place a predetermined uniform tonnage on the belt. One long slope belt in a coal mine in Pennsylvania is being fed at three separate stations with the controls so arranged that whenever the maximum load is going onto the belt from the first station, the other two stations automatically cut out. Whenever the load from the first station drops back, the other two stations again automatically cut in. 4. Careful design of the chutes and skirts is necessary to get the load centered on the belt with a minimum of free margin along each edge. Some free margin at the edge of the belt is necessary to prevent spillage, but if the load can be kept accurately centered, this free margin area can be reduced, and more material can be carried on the belt. What can be accomplished in this respect will vary widely, depending on the nature of the material being hauled. The chute and skirt design must also protect the belt. 5. The design of chutes and skirts should also get the load traveling in the same direction and close to belt speed, so that the load comes to rest on the belt as quickly as possible. The design of the chutes and skirts is worthy of careful study, and after a system is put into operation it should be experimented with to get the best results. Belt Speed: High belt speeds should be used in high tension work. Obviously, high belt speeds enable haulage on a narrower belt, reducing initial cost. The major portion of belt wear takes place at the loading point and around the terminal pulleys. The

Jan 1, 1952

By Robert M. Funk, James E. Lawver

The development of a laboratory and pilot-scale high-tension electrostatic flowsheet for the production of iron-ore super concentrates having silica contents in the range of 0.1% is presented, A variety of hematite and magnetite ores were tested with this flowsheet, and the pertinent metallurgical data are given. The applications and limitations of this process are brought out in terms of both practical and theoretical considerations. There have been many new developments in the past few years that have required mineral beneficiation engineers to consider new techniques for iron-ore concentration. Developments in iron powder metallurgy, and particularly the making of sheet metal directly from high-purity iron-ore concentrates, and the concept of using metallized pellets as a substitute for scrap in electric-furnace steelmaking have created demands for what is commonly referred to as iron-ore superconcen-trates. In most cases, specifications for superconcentrates are that they contain less than 0.1% total gangue, with some applications requiring less than 0.05% gangue. To produce these concentrates, new and modified processes are being developed, and the three methods receiving attention as the primary concentrating methods are: 1) Magnetic separation in which dry rotor-type equipment is used for magnetite and the high-intensity magnets are used for hematite. 2) Amine flotation. 3) "High-tension" electrostatic separation (or, elec-trodynamic concentration). This paper examines the high-tension electrostatic process in terms of its limitations and abilities for the production of iron-ore superconcentrates. A technically effective and simple flowsheet was developed through laboratory and 5 and 100-ton pilot-plant studies for the production of concentrates containing from 0.02 to 0.20% SiO, with iron recoveries between 85 and 90%. The process is also shown to be economically feasible. Laboratory Test Program Separation Characteristics: The suitability of the electrostatic high-tension method for iron-ore supercon-centration stems from the fundamental mechanism governing the process, i.e., particles are separated on the basis of differences in electrical conductivities. This conductivity is a characteristic property of minerals, and if the difference between the conductivities of two min- erals is great enough, then their separation can be performed by a high-tension machine. As shown in Figs. 1 and 2, the critical factor is the time necessary for the particle charge density, which is established by a corona discharge field, to decay to a level such that the centrifugal force acting on the particle exceeds its electrostatic, or image, force. At this point, the particle can leave the rotor at some trajectory and fall into an appropriate product receptacle. This decay time is directly proportional to electrical conductivity, and the process variables, such as feed rate, rotor speed, electrode configuration, and rotor composition, are selected to optimize the ability of the separator to resolve different conductivities. Table 1 lists the measured conductivities for some of the iron and gangue minerals used in developing the flowsheet. These results were obtained by using the conductivity cell described by Lawver and Wright.' The discharge curve of most minerals that lend themselves to super concentration by high-tension is so fast that the

Jan 1, 1971

By D. F. Kaufman, H. R. Spedden, A. M. Gaudin

MANY studies of comminution have been made to ascertain the size distribution of the product and to evaluate the work of comminution in the light of the size distributions of the feed and product. Up to now, these studies have been essentially statistical in character, that is, a certain lot of feed was subjected to comminution in some specified way, and the aggregate product was fractionated into sizes, thereby losing all knowledge of individual relationship of feed to product pieces. Radioactive tracers offer a means to do something in this respect which could not be done before, namely, to follow the rupturing of some particular piece in its normal environment of other pieces. That is, it permits going beyond the usual statistical limitations of size distribution studies to what may be termed a personalized or individualized study. The purpose of this paper is to present some preliminary experiments conducted with this tool. The method employed was to mark radioactively some constituent of a feed. It is possible, of course, to consider the preparation of two lots of material of which one is radioactive and the other is not, and to blend the two ahead of the comminuting step; but to do so is open to the objection that the two preparations may not be identical. Therefore a technique has been chosen that removes this objection by merely taking out a size fraction of a comminution feed, rendering that fraction radioactive by exposure to a neutron flux, and then by returning it to Table I. Size Distribution of Offspring Albite Particles Originally 28/35 Mesh and in Admixture with Other Sizes After Grinding 2 min in a Steel Ball Mill Specific Activity ' Cumu- Corrected Distrl- latlve Size for Back- butlon In Distri- Fractlon ground, Weight, Product, button, of Product, cpm/gm g Pctb Pct Mesh (A). (W) (P) (ZP) + 28 0 56.0 0 100.1 28/35 62.6 54.0 24.8 75.3 35/48 62.8 59.4 27.7 47.6 48/65 41.1 53.0 16.2 31.4 65/100 29.6 45.7 10.2 21.2 100/150 23.7 37.0 6.6 14.6 150/200 23.3 25.1 4.4 10.2 200/270 20.1 19.0 2.9 7.3 270/400 17.8 21.2 2.9 4.4 -400 22.9 25.2 4.4 — 100.1 a These activity determinations were made in rapid succession in the order given. The specific activity (Ao) of the active 28/35 mesh fraction of the feed was measured at the beginning, after the measurement on the 65/100 mesh size fraction of the product, and; The end. The decay-corrected activities at those times were 246.7, 241.0. and 236.9 cpm per gm. The weight (W0) of the active 28/35 mesh fraction in the feed was 55.0. b Example of calculation for P in the 65/100 mesh oroduct frac- A W tion; A = 29.6, W = 45.7, Ao = 242.7, Wo = 55.0: P = — x — Ao Wo = 0.102 = 10.2 pet. the remainder of the charge for the comminution experiment. A relatively simple procedure was developed by which albite, containing sodium, was activated in the M.I.T. cyclotron. The cyclotron makes highspeed deuterons which impinge on a beryllium target, thereby producing a concentrated neutron flux. The mineral was exposed to this flux for 2 hr. This treatment changed enough of the sodium to sodium 24 (14.8 hr half-life, 1.4 mev ß) as to make detection and measurement easy. The nuclear reactions taking place were: 11Na23 (n,?) 11Na24 (irradiation) 11Na24 ß,?,? 12Mg24 (decay) The detailed technique of the experimentation was as follows: 40 kg of hand-sorted, lump albite were crushed to pass 10 mesh. After careful mixing of the lot, a screen analysis was made. The whole lot of material was fractionated on standard Tyler screens from 14 down to 200 mesh. Samples for experiments were compounded from these fractions in accordance with the screen analysis. When it was desired to make an experiment in which, for example, the 28/35 mesh size fraction was to be studied, the blend of size fractions was made as indicated above, except that the 28/35 mesh size fraction was added only after irradiation in the cyclotron. The blended charge containing the activated albite was ground for 2 min in a laboratory ball mill with a steel ball charge of controlled size distribution. The ground product was carefully sized on a set of Tyler screens in a Ro-tap. Each size was analyzed for radioactivity by the use of an end-window Geiger-Mueller counter and standard scaling circuit. This analysis was carried out in detail as follows: a 20-g sample was placed in a Petri dish, packed carefully to obtain reproducible geometric distribution with reference to the Geiger-Mueller tube, and the activity was counted for a 2-min period. Several determinations of the activity of the active size fraction in the feed were made at various times to establish the decay in activity with time. Linear interpolation was used to evaluate the activity that the active size fraction in the feed would have had at any given instant. The ratio of the observed activity in a size fraction of the product to the activity that the active size fraction in the feed would have had at the same time gives the fraction in the product size that came from the irradiated size in the feed. The general formula for finding the distribution, P, of a specific individual size fraction in the feed

Jan 1, 1952

By F. M. Lewis, J. F. Myers

The paper reports the development of a large, slow speed ball mill closed circuited with a hvdro-scillator. This increased grinding efficiency 28 pct over conventional units. AS the title indicates, this is a progress report for the first year of experimental operation of a relatively large diameter, slow speed ball mill with 1-in. balls, with and without various classifiers, and including a hydroscillator. Several years ago the authors came to the conclusion that if any further progress were to be made in our flotation process, an understanding of the chemical and physical reactions and the control thereof, in the grinding circuit, was essential. It may be stated, at this point, that the preparation of a correct feed for the flotation process is still the number one objective of our grinding studies. This progress report deals with grinding "per se," and not with the chemical and physical reactions as the result of grinding. Prior to this time we had taken the naive position, that since grinding had been practiced by the industry for some fifty years or more, it had been well exploited and developed by a host of investigators. Consequently, we had given the matter only superficial attention. At the start of our study, we were aware of the fact that some difference in opinion existed, but we were not prepared for the controversial evidence that developed nor for the extent to which the experts differed in their viewpoints. There seemed nothing else to do but to tour the continent and decide for ourselves what was right and what was wrong, read all the published literature and visit with anyone who had ideas about the matter. This we did over the next few ensuing years. Since our study has met with some measure of success, we report, herewith, these data that ultimately they may be of help to other investigators. In 1944 it became necessary to increase our grinding capacity and at the same time to grind the sulphide portion a little finer. This provided the impetus to put our ideas into effect. As has been previously described, our grinding flowsheet at that time consisted of a 6x12 rod mill, followed by two 5x10 ft, and one 6x12 ft trunnion overflow ball mills, with conventional rake classifiers. Naturally, we first considered all the possibilities of altering or adding to the existing equipment. It became evident that if any appreciable increase in grinding efficiency could be obtained, it would pay to scrap the existing small mills and classifiers and install one big mill with adequate classification. Trying to determine just how much more efficient a big mill could be, in which were incorporated all

Jan 1, 1951

By J. F. Myers, F. M. Lewis

The paper reports the development of a large, slow speed ball mill closed circuited with a hvdro-scillator. This increased grinding efficiency 28 pct over conventional units. AS the title indicates, this is a progress report for the first year of experimental operation of a relatively large diameter, slow speed ball mill with 1-in. balls, with and without various classifiers, and including a hydroscillator. Several years ago the authors came to the conclusion that if any further progress were to be made in our flotation process, an understanding of the chemical and physical reactions and the control thereof, in the grinding circuit, was essential. It may be stated, at this point, that the preparation of a correct feed for the flotation process is still the number one objective of our grinding studies. This progress report deals with grinding "per se," and not with the chemical and physical reactions as the result of grinding. Prior to this time we had taken the naive position, that since grinding had been practiced by the industry for some fifty years or more, it had been well exploited and developed by a host of investigators. Consequently, we had given the matter only superficial attention. At the start of our study, we were aware of the fact that some difference in opinion existed, but we were not prepared for the controversial evidence that developed nor for the extent to which the experts differed in their viewpoints. There seemed nothing else to do but to tour the continent and decide for ourselves what was right and what was wrong, read all the published literature and visit with anyone who had ideas about the matter. This we did over the next few ensuing years. Since our study has met with some measure of success, we report, herewith, these data that ultimately they may be of help to other investigators. In 1944 it became necessary to increase our grinding capacity and at the same time to grind the sulphide portion a little finer. This provided the impetus to put our ideas into effect. As has been previously described, our grinding flowsheet at that time consisted of a 6x12 rod mill, followed by two 5x10 ft, and one 6x12 ft trunnion overflow ball mills, with conventional rake classifiers. Naturally, we first considered all the possibilities of altering or adding to the existing equipment. It became evident that if any appreciable increase in grinding efficiency could be obtained, it would pay to scrap the existing small mills and classifiers and install one big mill with adequate classification. Trying to determine just how much more efficient a big mill could be, in which were incorporated all

Jan 1, 1951

By F. M. Lewis, J. F. Myers

W. I. Garms-—The authors state that when they added 11 tons of balls to the 45 pct volume ball load, the power needle did not budge. The question arises as to whether any increase in capacity accompanied the 11-ton addition. It is hardly conceivable that all of the balls making up the 11-ton addition were sterile. However, as the work done in the ball mill is not by kilo-watthours per se or balls per se but by kilowatthour-balls, it is possible that the kilowatthourballs in the mill before the 11-ton addition had a different amperage, voltage or robustage than those in the mill after the addition and that the overall outcome was that equal work was done before and after the addition. The authors can keep the efficiency that they got with the 45 pct volume 1-in. ball charge and get additional capacity by putting more kilowatthourballs to work in the form of the addition of 11 tons of 2-in. balls to the present ball charge, and subsequent makeup additions by weight of 20 pct 2-in. balls and 80 pct 1-in. balls. J. F. Myers (authors' reply)-—Mr. Garms points his finger at the 11 tons of "sterile" balls added to the Tri-cone mill. A fact so interesting to us that we felt it should be reported. Clearly, ball slippage at the existing mill speed prevented any further transfer of power from the shell to the ball mass. We agree that the kilowatthourballs would increase, were we charging rough 2-in. balls as suggested by Mr. Garms. The assumption is, of course, that we would get a corresponding increase in work done. Our reported data (CIME meeting, Vancouver, 1946) shows a gain of work accomplished of 7.6 pct as we decreased the ball size from 2 in. to 11/2 in. and a further gain of work accomplished of 2.1 pct when we went from 11/2 in. to 1 in. The small ball results reported by H. R. Banks at Chapman Camp are also very convincing. In the case of 2-in. balls they would take the conventional ball path and start cascading. They would thus destroy the classifying pool over the "foot" of the

Jan 1, 1951

By J. F. Myers, F. M. Lewis

W. I. Garms-—The authors state that when they added 11 tons of balls to the 45 pct volume ball load, the power needle did not budge. The question arises as to whether any increase in capacity accompanied the 11-ton addition. It is hardly conceivable that all of the balls making up the 11-ton addition were sterile. However, as the work done in the ball mill is not by kilo-watthours per se or balls per se but by kilowatthour-balls, it is possible that the kilowatthourballs in the mill before the 11-ton addition had a different amperage, voltage or robustage than those in the mill after the addition and that the overall outcome was that equal work was done before and after the addition. The authors can keep the efficiency that they got with the 45 pct volume 1-in. ball charge and get additional capacity by putting more kilowatthourballs to work in the form of the addition of 11 tons of 2-in. balls to the present ball charge, and subsequent makeup additions by weight of 20 pct 2-in. balls and 80 pct 1-in. balls. J. F. Myers (authors' reply)-—Mr. Garms points his finger at the 11 tons of "sterile" balls added to the Tri-cone mill. A fact so interesting to us that we felt it should be reported. Clearly, ball slippage at the existing mill speed prevented any further transfer of power from the shell to the ball mass. We agree that the kilowatthourballs would increase, were we charging rough 2-in. balls as suggested by Mr. Garms. The assumption is, of course, that we would get a corresponding increase in work done. Our reported data (CIME meeting, Vancouver, 1946) shows a gain of work accomplished of 7.6 pct as we decreased the ball size from 2 in. to 11/2 in. and a further gain of work accomplished of 2.1 pct when we went from 11/2 in. to 1 in. The small ball results reported by H. R. Banks at Chapman Camp are also very convincing. In the case of 2-in. balls they would take the conventional ball path and start cascading. They would thus destroy the classifying pool over the "foot" of the

Jan 1, 1951

By A. Chakravarti, C. C. Harris

First-order kinetics adequately describe the recovery of a single floatable species. The conditions under which cell behazior can be simulated by a single-phase model are discussed; deviations from the multiphase model are accountable. The rate coefficient depends upon operating conditions besides particle size and mineralogy and the physicochemical properties of the pulp constituents. The model for a real flotation system combines the first -order, single-phase model with a distribution of noninteracting species. Methods for determining the distributions are discussed. Applications of the procedure to flotation studies are outlined. The mathematical description of flotation as a rate process can be traced to Garcia1 who reported cases in which the fraction of solid remaining in a cell in semi-batch operation, F (for Notation used throughout, see Table l), decays exponentially with experimental time T. F = e-k>r. (1 The differential equation leading to Eq. 1 is dF ----= -k,F (2) dT which is recognized as the algebraic analog of the chemical kinetics first-order reaction equation.' This has led to suggested mechanisms for flotation and mathematical relationships other than first order: with very little foundation other than analogy. The flotation rate coefficient kl can be accounted for as the product of several probabilities concerning bub-ble-particle attachment and detachment, aggregate entry from pulp into froth and return from froth to pulp, and removal from froth as concentrate. These factors clearly contribute to the magnitude of the overall flotation rate, and there are others also, but describing flotation merely as the overall probability of consecutive gross events hardly appears a significant advance in understanding. More recently, it has been described in terms of a stochastic process, which promises to provide a more satisfactory approach? The present paper will be more concerned with a phenomeno-logical description of the flotation process than with an account of the internal mechanics of the process. This is not a statement of priorities, or a judgement of the relative importance of these options but rather an assessment of what is presently possible. Previous investigators'" have developed methods for interpreting semi-batch flotation data in terms of species distribution. These methods have been used in an attempt to explain flotation behavior in response to changes in operating conditions." Inspection of examples given in the literature" reveals discrepancies between theory and experiment, and data obtained in the Colum-

Jan 1, 1971

By Norman F. Schulz

The technical excellence of separation achieved in a mineral concentration process, or any other process where two constituents of any kind are physically separated from each other, is expressed uniquely and quantitatively by the separation efficiency (E,): E. = (R - R,), where R is the percentage of the valuable constituent and R, the percentage of the waste appearing in the concentrate. The separation efficiency meets all the usual requirements of a true efficiency and is independent of the nature of the process used. It can be visualized as the percentage of the feed material that, in effect, actually undergoes complete separation while the rest of the feed is distributed randomly and unchanged into the two separation products. The problem of quantitatively assessing the results of a mineral concentration process recurs whenever one tries to decide which of two similar test results is more desirable. Although economic factors may greatly influence the eventual choice of process, they usually have very little to do with the technical excellence of a separation and are, therefore, not a consideration at this point. Concentrate grade and recovery, used simultaneously are the most widely accepted measures of performance, and if both the grade and the recovery are greater for one case than for another, the choice is simple. However, when the results of closely competitive operations are compared, it often occurs that one set of test results shows a higher grade and a lower recovery than another and the choice is no longer obvious. Laboratory investigations of a mineral concentration process applied to a given feed material can often be designed to yield successive paired values of concentrate grade and recovery over the practical range of the course of a separation process.' These results may be plotted as some combination of the cumulative values of grade, weight recovery, or unit recovery in the concentrate. The resulting correlations for two slightly different sets of conditions can then be compared qualitatively, or even quantitatively in particular cases.2 Such comparisons are not applicable, however, to individual concentration test results. The reduction of a separation test result to a single efficiency figure, preferably as a percentage of ideality, would afford the simplest and most precise means for comparing the results of competitive tests. The effective proportion of feed material that undergoes complete separation in any real concentration process, hereafter referred to as "separation efficiency," is such a figure. It is essily calculated from the usual feed and product data and can even be evaluated, in many cases, from data in publications describing various concentration processes. Mineral Concentration Processes and Their Evaluation Mineral Concentration Processes: The overall concentration process may be subdivided into a liberation process which severs the bonds between unlike mineral species, and a separation process, which sorts the resulting liberated particles into two products—a concentrate rich in mineral A and a tailing lean in mineral A. A perfect mineral concentration process is represented by the idealized scheme shown in Fig. 1 in which liberation and separation are not differentiated. The feed mixture consists of two mineral types which are completely liberated early in the process and then sorted by type. All the mineral particles of type A which respond uniquely to the selection forces acting in the ideal separator, report to the concentrate, while all the mineral particles of type non-A report to the tail-

Jan 1, 1971

By R. B. Walker, R. Palmer, L. C. Bryner

The purpose of this investigation was to further study some of the factors affecting the air oxidation of sulfide minerals. Data obtained from laboratory studies on the effect of temperature, concentration of iron salts, pH, and other catalysts on the biological and nonbiological oxidation of sulfide minerals are presented. The oxidation of copper sulfide minerals and subsequent leaching of the soluble copper has long been a profitable operation. The chief oxidizing agent is oxygen from the air. This reaction is influenced by a number of different factors such as temperature, light, hydrogen ion concentration, oxidizing agents, and various catalysts. Sullivan1 found that copper sulfide minerals are slowly oxidized by the air and that dilute sulfuric acid and iron (111) increased their solubility. At that time it was not known that bacteria were involved in the oxidation of the sulfide minerals. It has been shown that pyrite, molybdenite, free sulfur, and the copper sulfide minerals found in the waste rock dumps in Bingham Canyon, Utah, are largely oxidized by chemosynthetic autotrophic microorganisms found in the leaching streams.2-6 Similar organisms have been isolated in other mining areas7,8 and from the effluents from coal rnines.9 The following are some of the questions that have arisen regarding the mechanisms of the reactions involved in the leaching process. How dependent is the leaching operation upon the biological reactions? How does the temperature effect the biological and non-biological reactions? How does the concentration of iron (II) and (111) effect the rate of oxidation of copper (II) sulfide? Does the iron act as a catalyst? Does activated carbon act as a catalyst on the oxidation of sulfide minerals? What effect does the hydrogen ion concentration (pH) have on the above reactions? The purpose of this paper is to report the results of laboratory studies on these problems. EXPERIMENTAL The apparatus used in this investigation was an air lift percolator described in our earlier work.' The percolator consisted essentially of an air lift which enabled the leaching solution and microorganisms to pass freely through a suspension of mineral (substrate) in Ottawa sand. The air lift served the additional purpose of maintaining an ample concentration of oxygen and carbon dioxide in the nutrient (leaching) solution. (See Fig. 1.) The percolators were so constructed that they could be immersed in liquid constant temperature baths for temperature studies. The thermostats were controlled to +0.5°C. The reactions were all run in the dark. The percolators were charged with one to three g of mineral, dispersed in 100 g of Ottawa sand and 100 ml of nutrient or leaching solution. The sand was purified to remove traces of iron by digesting with concentrated hydrochloric acid and washng with distilled water until free from chloride. After drying, the sand was ready for use. The compressed air supply was regulated to 5 psi and bubbled through an aerator containing distilled water, which was maintained at the same temperature as the bath used. The air was filtered through cotton plugs to maintain sterile conditions. The tops of the

Jan 1, 1968

By Frederick F. Frick

SPONGE iron as produced at Anaconda is a fine, -35 mesh, impure product, about 50 pct metallic iron, obtained from the reduction of iron calcine at a temperature of 1850°F by use of coke resulting from slack coal. The metallic iron particles are bulky and spongey and precipitate copper readily and rapidly from a copper sulphate solution. Investigation of the treatment of Greater Butte Project, Kelley, ore at Anaconda early showed the desirability of using sponge iron as a precipitant for the copper in solution resulting from desliming of the ore in a dilute sulphuric acid solution. Anaconda had done considerable work on the production of sponge iron in 1914 for use as a precipitant of copper from leach solutions. Some success and considerable experilence were attained at the time. indicating that, sponge iron might be successfully made by a modification of the process used in 1914, a batch process in which an iron calcine was reduced by means of soft coke, resulting from noncoking coal, in a Bruckner-type revolving horizontal cylindrical furnace widely used 50 years ago. The coke and calcide formed the bed in the Bruckner furnace, which was rotated at about 1 rpm. The bed was brought to a temperature of about 1800°F by means of an oil flame over the surface. Although results were reasonably satisfactory, they did not warrant full development of the process at that time. A good deal of work has been done in the last 50 years on the production of sponge iron. The objective in some cases has been the production of a precipitant for copper from solution, but the bulk of the work has been done for the production of open-hearth steel furnace stock. The production of an open-hearth stock presents two problems rather than one: first, producticon of the sponge iron, and second, what is perhaps of equal difficulty and importance, conversion of the sponge iron into a form suitable for use in the open-hearth furnace. So far as is known to the writer, none of the sponge iron processes tried in the past have proved to be economically feasible. However, Anaconda had a combination of conditions appearing to justify an attempt to produce sponge iron which would serve for the leach-precipitation-float process. It was thought that the process used in 1914, if changed to a continuous one, might work out satisfactorily. The following favorable conditions at Anaconda justified the investigation: 1—A sufficient tonnage of good grade iron calcine resulting from the roasting of a pyrite concentrate in one of the acid plants, at substantially no cost. 2—Reasonably cheap natural gas. 3-—The fact that there was no need for production of a high grade product. 4— The fact that there was no need for obtaining a consistently high reduction of' the iron in calcine. A small revolving Bruckner-type furnace about 2 ft ID by 4 ft long was set up for early pilot work at the research building. This pilot furnace showed that a satisfactory product could be obtained at reasonable cost. It also indicated a marked advantage in preceding the reduction furnace with a furnace of similar size and capacity for preheating and roasting out any residual sulphur from the feed. The small furnace was operated for several months, various details of the process were worked out. and sponge iron was produced to supply a pilot LPF plant which treated 300 lb of Kelley ore pel- hr. Later a second pilot furnace 5 ft in diam and 12 ft long inside was set up at our reverberatory furnace building. This furnace confirmed the data of the small furnace and gave a basis for design of the final plant. At Anaconda a pyrite concentrate, running about 48 pct S, is recovered from copper concentrator tailings by flotation. This concentrate is roasted to sulphur of 3 pct or less at the Chamber acid plant. The iron calcine contains about 57 pct Fe and 18 pct insoluble. The iron calcine feed, as mentioned before, is re-roasted and preheated in a reroast furnace preceding the reduction furnace. Both are of the Bruckner type. The reroasted calcine is fed into the reduction furnace at 800" to 1000°F along with 30 pct slack coal. In the feed end of the furnace the volatile is burned from the slack, giving a soft coke which readily serves for reduction of the iron. Hard metallurgical coke will not serve the purpose. since it does not reduce CO readily at a temperature of 1850°F. All indications are that the actual reduction of the iron is accomplished by carbon monoxide below the surface of the bed, which is 30 in. deep at its center. Apparently there is a constant interchange: Fe²O³ + 3CO = 2Fe - 3CO², CO² + C = 2CO Actually iron oxide is reduced by CO at somewhat lower temperature than the 1850 °F used in the process. but this temperature is necessary to obtain a satisfactory rate of furnace production. The furnace atmosphere is generally reducing, and typical blue carbon monoxide flames satisfactorily cover the bed. Gas flames from four 3-in. Denver Fire Clay Inspirator burners are played directly on the bed, which is slowly cascaded by the 1 rpm of the furnace. An excess of coke is necessary to assure maintenance of good reducing conditions in the furnace bed. Part of this coke is recovered for re-use.

Jan 1, 1954

By A. W. Jenike

A theory of gravity flow of bulk solids like ores, concentrates, coal is outlined and an example of design for unobstructed flow is given. Mass-flow and plug-flow patterns are defined and related to the geometry of the bin. The effect of these patterns on segregation, first in-first out flow, and caking is described. This paper is a report on the results from the AIME sponsored project "Flow of Bulk Solids" which was carried out under the author's direction at the Utah Engineering Experiment Station, University of Utah, during the period 1956-1962. The theory has been applied to the design of numerous industrial storage plants. A QUANTITATIVE METHOD OF DESIGN FOR FLOW There is hardly an industry which does not store and handle solid materials in bulk form. When the volume of the solids is large, gravity is invariably relied upon to cause the solids to flow out of storage, through channels and chemical reactors. Such materials as ore, coal, cement, flour, plastics, clay, soil, to which a general term of bulk solids is applied, flow by gravity or are expected to flow by gravity in thousands of installations and by billions of tons annually. The chemical and process industry depends on gravity flow of solids in process and out of storage. Mining relies on gravity flow in block-caving, in ore passes, as well as in storage and loading bins. Agriculture uses gravity flow in elevators, in feed plants, and in farm silos. In practice, satisfactory gravity flow is seldom attained. Such problems as doming, piping (rat-holing), lack of live capacity, segregation, deterioration in storage, harass industry. In spite of the universality of these problems, until recently, there has been no theory of solids flow, no method of predicting how a solid would behave, and how the channel, bin, or hopper should be designed. In the late nineteen forties and early fifties, the theory of plasticity came of age and simultaneously the computer became widely available. These two factors permitted the development of a theory of flow of bulk solids. The work was conducted during the years 1953-1961. In conjunction with the development of the theory a method to measure the flowability of solids was also developed1-7 and the theory was ap-plied to the design of ore pockets, storage plants,16 chemical reactors and adsorption columns. A comprehensive report of this work is given in a previous work." THEORY OF FLOW The most important question which a theory of flow has to answer is: what conditions need to be satisfied for a given solid to flow uniformly and without obstructions through a channel? First of all, the following observations are made:

Jan 1, 1967

By A. M. Gaudin, D. W. Fuerstenau

By streaming potential techniques, the zeta potential of quartz has been measured as a function of the concentration of dodecylammonium salts at different pH values. These experiments indicate that adsorption of dodecylammonium ions takes place as individual aminium ions until a certain critical concentration is reached in the double layer, at which point the adsorbed collector ions begin to associate into patches of ions, called hemi-micelles. Because of its effect on the surface potential of quartz, pH controls the adsorption of dodecylammonium ions. The value of cooperating agents and nonionized constituents of flotation systems is discussed. A PREVIOUS investigation' of the role of pH in the flotation of quartz with anionic collectors is here extended to an inquiry of how the cationic collector, dodecylammonium acetate, is adsorbed at the quartz-solution interface. The equilibrium amount of this collector adsorbed by quartz has already been measured as a function of concentration and pH.2,3 The weight of prior evidence has suggested that two mechanisms are involved in the adsorption of this collector by quartz, but with existing adsorption measurements alone it is not possible to determine what the mechanisms are. Gaudin and Bloecher2 postulated that a polymolec-ular adsorption coating begins to form before the first monolayer has been completed. Gaudin and Morrow4 indicated that a similar situation might exist in the film of dodecylammonium ions on hematite. A study of electrokinetic potentials at the quartz-solution interface provides some detail toward a fuller understanding of the adsorption mechanism. Experimental Results The zeta potential at the quartz-solution interface was calculated from streaming potential data measured with the apparatus described in detail in a thesis by Fuerstenau5 and briefly in a paper by Gaudin and Fuerstenau.1 The quartz used in these experiments was a 48/65-mesh fraction prepared in the manner described previously.' Zeta potential of quartz was obtained for various amounts of the amine salt at pH 4, 7, 10, 11. In

Jan 1, 1956

By C. S. Chang

Replacement of active groups in corn starch, as shown by results of the starch derivatives tested, impairs rather than improves the value of corn starch as a selective iron oxide depressant. Increased ability to depress iron oxides is generally accompanied by a parallel ability to depress quartz. PEARL starch, a corn starch prepared with a minimum of alteration, is a good selective depressant for iron oxide minerals.' From what is known of the complex chemistry of corn starch, there is evidence that it consists of a mixture of saturated giant polymeric molecules, both straight and branch-chained, the constituents consisting of glucose units with hydroxyl radicals as active or polar groups. To add to the existing knowledge of the chemistry of the flotation process, it is desirable to ascertain the effects on the corn starch as an iron oxide depres-

Jan 1, 1955

By E. J. Roberts

The theory is developed that the tons ground through a given mesh per day in a wet ball mill is proportional to the percent plus that mesh in contact with the balls and the net power applied to the balls at this point. A grindability test is described. DURING the course of a study of the fundamentals of classification in 1937, the need for a more basic understanding of the action of a ball mill became acute. Unless one knows how classification affects grinding, one cannot hope to effectively improve on classification. The methods of evaluating grinding efficiency that depend on surface developed were studied but soon discarded for two reasons: 1. There was no apparent method which could be generally used to give a reliable figure for the actual new surface developed as a result of grinding. Subsequent papers have not changed this conclusion. 2. The practical evaluation of grinding in the main ore dressing applications was in terms of the percentage retained on a screen which passes 90 to 99 pct of the material and not in terms of surface area. The Probability Theory With the background of our experience in the field of closed-circuit grinding, together with the papers of Lennox,1 Gow,2 Gaudin,8 Fahrenwald,4 Coghill, and others, the approach of the theoretical physicist was then tried. The thought was somewhat as follows: When one grinds in a ball mill, a given expenditure of power leads either to a certain number of point to point blows per hp-hr or to a certain distance of line contact per hp-hr, depending on whether the action of the balls is considered to be cascading or rolling. It is also assumed that the balls actually come together on each blow or during the roll. Then a volume of slurry will be covered per minute which is some function of the size of the particle being considered (see fig. 1). All particles coarser than this size will be reduced through this size. This volume of slurry contains a certain weight of ore, depending on the percent solids and the density of the solids. If we fix the percent solids and the density of the solids and let w be this certain weight of ore in the volume covered, then, in mathematical terms, what we have just postulated is, w —— 8 hp (a) dt If W is the total weight of ore present in the mill, then we can write. W w/8 hp (b) W dt and if C is the cumulative percent plus the size chosen at the start of the time interval dt, w w c/dt W 8 hp x c (c) wc But wc/100 is the weight plus the size chosen which at 100 wc the close of time dt is finer than that size, and W is the decrease in the percent plus of the whole mass of ore or —dC. Then, —W dC/dt 8 hp x C. (d) In other words, the mesh tons ground through a given size per unit of time is proportional to the hp and the percent plus the mesh. A crude analogy would be to picture a 1-ft-wide steam roller going down the road at 1 ft per sec. If we place one egg on the road per square foot, one egg will be smashed per second. If we place a dozen eggs per square foot, a dozen eggs will be crushed per second. Similarly, if all the particles in w are plus the mesh, i.e., C=100, we should have a maximum rate of reduction. If only 10 pct of them are plus the mesh (C=10), we would have only one tenth the maximum rate; if only 1 pct are plus the mesh, the balls have a hard time finding anything to work on. This is where the term "probability theory" comes from. The chances of the balls crushing a particle through a given mesh depends directly on the concentration of particles coarser than this mesh in the general pulp in the mill. Giving W the units of tons and dividing equation (d) through by W, we obtain -dC hp ----- = k---— C [1] dt ton where k is a constant for any one size of particle, density of solid and moisture content of pulp. Eq 1 is the rate equation for a first order reaction and says that the rate of decrease of the percent plus a given mesh with time is directly proportional to the hp per ton applied to the body of ore and to the percent plus the mesh in the ore mass as a whole. Since it is a differential equation, it only

Jan 1, 1951

By E. J. Roberts

The theory is developed that the tons ground through a given mesh per day in a wet ball mill is proportional to the percent plus that mesh in contact with the balls and the net power applied to the balls at this point. A grindability test is described. DURING the course of a study of the fundamentals of classification in 1937, the need for a more basic understanding of the action of a ball mill became acute. Unless one knows how classification affects grinding, one cannot hope to effectively improve on classification. The methods of evaluating grinding efficiency that depend on surface developed were studied but soon discarded for two reasons: 1. There was no apparent method which could be generally used to give a reliable figure for the actual new surface developed as a result of grinding. Subsequent papers have not changed this conclusion. 2. The practical evaluation of grinding in the main ore dressing applications was in terms of the percentage retained on a screen which passes 90 to 99 pct of the material and not in terms of surface area. The Probability Theory With the background of our experience in the field of closed-circuit grinding, together with the papers of Lennox,1 Gow,2 Gaudin,8 Fahrenwald,4 Coghill, and others, the approach of the theoretical physicist was then tried. The thought was somewhat as follows: When one grinds in a ball mill, a given expenditure of power leads either to a certain number of point to point blows per hp-hr or to a certain distance of line contact per hp-hr, depending on whether the action of the balls is considered to be cascading or rolling. It is also assumed that the balls actually come together on each blow or during the roll. Then a volume of slurry will be covered per minute which is some function of the size of the particle being considered (see fig. 1). All particles coarser than this size will be reduced through this size. This volume of slurry contains a certain weight of ore, depending on the percent solids and the density of the solids. If we fix the percent solids and the density of the solids and let w be this certain weight of ore in the volume covered, then, in mathematical terms, what we have just postulated is, w —— 8 hp (a) dt If W is the total weight of ore present in the mill, then we can write. W w/8 hp (b) W dt and if C is the cumulative percent plus the size chosen at the start of the time interval dt, w w c/dt W 8 hp x c (c) wc But wc/100 is the weight plus the size chosen which at 100 wc the close of time dt is finer than that size, and W is the decrease in the percent plus of the whole mass of ore or —dC. Then, —W dC/dt 8 hp x C. (d) In other words, the mesh tons ground through a given size per unit of time is proportional to the hp and the percent plus the mesh. A crude analogy would be to picture a 1-ft-wide steam roller going down the road at 1 ft per sec. If we place one egg on the road per square foot, one egg will be smashed per second. If we place a dozen eggs per square foot, a dozen eggs will be crushed per second. Similarly, if all the particles in w are plus the mesh, i.e., C=100, we should have a maximum rate of reduction. If only 10 pct of them are plus the mesh (C=10), we would have only one tenth the maximum rate; if only 1 pct are plus the mesh, the balls have a hard time finding anything to work on. This is where the term "probability theory" comes from. The chances of the balls crushing a particle through a given mesh depends directly on the concentration of particles coarser than this mesh in the general pulp in the mill. Giving W the units of tons and dividing equation (d) through by W, we obtain -dC hp ----- = k---— C [1] dt ton where k is a constant for any one size of particle, density of solid and moisture content of pulp. Eq 1 is the rate equation for a first order reaction and says that the rate of decrease of the percent plus a given mesh with time is directly proportional to the hp per ton applied to the body of ore and to the percent plus the mesh in the ore mass as a whole. Since it is a differential equation, it only

Jan 1, 1951

By W. H. Yarroll

This paper presents a general discussion of the utility of the statistical technique known as multiple correlation, and gives three specific examples of its application. The first demonstrates the most simple form, or straight-line, multiple correlation. The other two demonstrate multiple correlation analysis in best-fitting third-degree parabolas. With the increasing availability of digital computers, the method becomes entirely feasible where heretofore its use was impractical. The advantages of the method are discussed, and curves are presented which were developed by computerizing the regression equations developed in the examples. The method described in this paper is for use with complex systems, and the examples given are complex metallurgical systems. This subject falls under the general classification of cybernetics, which is defined' as "the science of control and communication, in the animal and the machine." Machine, in this case, can include chemical and physical processes which take place within machines and processing equipment of all kinds. The classical approach to scientific investigation has been the construction of simple models and the changing of variables one at a time. This method is wholly inadequate for obtaining useful information from the complex systems of modern technology including many of our flotation and hydrometallurgical processes. Cybernetics offers the method by which complex systems which have been implemented as an art in the past can now be analyzed with all the discipline of a science. Mathematical techniques for achieving multiple correlation analysis have decided advantages over some of the other statistical methods.2-4 Multiple correlation does not require a series of experiments that have been rigidly planned as do many of the other statistical methods. For this reason, it can be used effectively for interpretation of information from op- erating plants, where planned experiments are impractical, as well as for laboratory experiments. First-degree multiple correlation has been used for many years, and was useful, but limited to linear or nearly linear relationships. The technique described in this paper provides a method for analyzing systems which may be non-linear. This particular technique is multiple correlation analysis in best-fitting third-degree parabolas. It is quite possible that lack of reference to the technique has been due to the overwhelming mass of computations it requires even for relatively simple problems. With the advent of computers, however, its use has become entirely feasible. For the benefit of those who may not be familiar with the subject, it might be well to pause at this point to discuss multiple correlation analysis — what it is and what it achieves. To go back to the old familiar fitting of a set of points with a best-fitting straight line by the least squares procedure, it is conceded that this a good method where there is only one independent variable influencing the dependent variable under consideration, and providing that the relationship is linear. In most fields of investigation, however, there are usually a number of factors exerting important influence, and they frequently bear relationships with the dependent variable other than straight line. Under these conditions, it is easy to see how misleading a simple correlation study might be. Multiple correlation can be used for any number of variables. Within the experience of the writer, this technique has been used to establish regression equations for determination of carbon rate, burden permeability, and production rate on iron blast furnaces, production rate on open hearth steel furnaces, life of molds for steel ingots, rolled steel tensile strength and various defects in rolled products, the performance of ore dressing equipment, and optimization of extractive metallurgical process variables. The discovery of mathematical correlation does not prove, per se, true cause-and-effect relationship, but simply establishes a hypothesis. If the regression equation can be used to make successful predictions of what will happen with future arrays of independent variables, then one can feel reasonably assured that true cause-and-effect relationships have been discovered.

Jan 1, 1968