US 3908465 A
Method and apparatus are provided for a simple onstream particle size analysis. The device can be a single or multiple channel device. The percent solids is continuously measured both in a sample stream and in an internal reference stream, containing particles below a predetermined size. A semi-permeable hydraulic system is used such that the particle mass concentration in the internal reference stream can vary and remain proportional to the particle mass concentration of particles below the predetermined size in the sample stream.
Description (OCR text may contain errors)
United States Patent Bartlett 1 Sept. 30, 1975 PARTICLE SIZE ANALYZER  Inventor: Robert W. Bartlett, Palo Alto, Calif. f
' Assistant E.\'ammerStephen A. Kreitman  Assignee: Board of Trustee of Leland Stanford Jr. University, Stanford. Calif. 57] ABSTRACT  Filed: Feb. 25, 1974 Method and apparatus are provided for a simple onstream article size analysis. The device can be a sin- [211 445l97 gle or r riultiple channel device. The percent solids is continuously measured both in a sample stream and in  US. Cl 73/432 PS; 73/61 R an internal reference stream, containing particles  Int. Cl. ..G01N 15/02 below a predetermined size. A semi-permeable hy-  Field of Search 73/432 PS, 61 R, 53, 64.3. draulic system is used such that the particle mass con- 73/32 R, 28; 210/85. 96 centration in the internal reference stream can vary and remain proportional to the particle mass concen-  References Cited tration of particles below the predetermined size in UNITED STATES PATENTS the Sample Stream- 2,995,030 8/1961 Fcigley 73/61 R The apparatus provides a compartment having two 3,368,389 2/1968 Barnett 73/53 cells separated by a perforated barrier and means for 3.400.575 9/1 6 Madd n----- 3/ R measuring the pulp density of the slurries in each of 3,719,090 3/l973 Hathaway .4 73/6l R the cells 3,779.070 l2/l973 Cushman et al... 73/432 PS 3,797.31) 3/1974 Abe 73/432 PS 8 Claims, 3 ng ig US. Patent Sept. 30,1975
COMPUTER COMPUTER mm PARTICLE SIZE ANALYZER BACKGROUND OF THE INVENTION 1. Field of the Invention In many operations, where particles are being formed, either by the grinding of larger particles or building up from smaller particles, it is desirable to continuously monitor the size distribution. In ore grinding, the grinding operation adds a major cost to the refining of the ore. Therefore, during the grinding operation it is desirable to optimize the degree of grinding, so as to avoid an unnecessary amount of grinding to overly fine particles, while still providing an appropriate size distribution for further processing. Therefore, there is a need to analyze particle sizes obtained from the ore grinding operation as a slurry, and relate this information to the grinding operation, so as to maintain the optimum degree of grinding.
Continuous particle size analyzers fall into two broad categories. For bulk property analyzers an average slurry property is influenced by the particle size distribution. Sensing zone analyzers examine discrete particles sequentially. This information must be stored and the results of several particles examined to arrive at a statistically accurate characterization of the slurry.
An ultrasonic particle size analyzer uses the influence of particle size distribution on the attenuation of an ultrasonic beam. This device lacks sensitivity and is subject to large errors caused by bubbles, but it is the basis of a recently introduced, rather complicated commercial instrument.
Classification of the sample stream by sifting or hydraulically separating it into two streams, involving undersize and oversize particles, can be the basis of a determination. This is an old idea but complete classification is difficult to achieve in a continuous instrument. Although the present invention uses a sieve, it does not sift the sample slurry or classify it. Although an equilibrium exchange of small particles is involved, very few particles will enter or leave the sample slurry as it passes through the device.
The Royal School of Mines, Imperial College, London, has developed a device in which the slurry passes through a helix. Larger particles centrifuge to the outer zone of the slurry and cause differential density. These density profiles are measured with a nuclear density gauge(s). A U tube commercial device operates on a similar principle. However, results also depend on the particle mass concentration as well as the particle size distribution. This parameter must be measured and corrections must be made. In the present invention changes in mass concentration in the sample stream are followed by corresponding changes in the internal stream automatically by the exchange of undersize particles and an electronic logic circuit correction is not needed.
2. Description of the Prior Art Articles of interest concerning particle size analyzers include: Nakajima et al, I & EC Fundamentals 4 587 (1967); Holland-Batt, Sec. C, Trans. Inst. Min. Metall. 78, 185 (1969); Osborne, C.I.M. Bulletin 65, 97 (1972) (see also SME-AIME Preprint 73-8-80); and Bassarear et al., Min. Cong. J. 58, 36 (1972).
SUMMARY OF THE INVENTION Method and apparatus are provided for on-stream particle size analysis. A sample slurry stream and an internal reference slurry stream communicate through a barrier having openings of a predetermined size. The pulp density of the two streams is monitored. By means of a mathematical formula, the weight percent of particles at and below the predetermined size, as compared to total solids in the sample stream, can be determined as a function of the measured pulp densities.
The apparatus provides a compartment having two cells sample and reference cellswhich communicate solely through a perforated barrier, having openings of a predetermined size which is intermediate the size of particles present in the slurry stream to be measured. Agitation means is provided in the reference cell for maintaining the slurry in the cell as a relatively uniform composition. Monitoring means is provided for monitoring the pulp density of the surries in each of the cells. Optionally, electronic circuitry can be provided for directly relating the measured pulp densities of the slurries in the two cells to the weight percent of particles at and below the predetermined size, as compared to the total solids in the slurry in the sample cell.
BRIEF DESCRIPTION OF THE DRAWING FIG. 1 is a diagrammatic view of a single channel particle size analyzer;
FIG. 2 is an elevational view, partially broken away, of a single channel particle size analyzer and employing a single pulp density gauge; and
FIG. 3 is an elevational view, partially broken away, of a single channel particle size analyzer using two pulp density gauges.
DESCRIPTION OF THE SPECIFIC EMBODIMENT A method for analyzing particle size of a slurry having a range of particle sizes and an apparatus for measuring the particle size is provided. The method employs a reference slurry stream with the sample slurry stream, measuring the pulp density of each of the streams, either continuously or intermittently. The first slurry stream, the sample stream to be analyzed, may be the entire stream from the source or a diversionary stream from the source. The second stream, the internal reference stream, is a slurry of particles which are at and below a predetermined size, the size being intermediate the size of the particles present in the sample stream. The particles in the internal reference stream will be referred to hereafter as undersized particles or particles below a predetermined size.
The two streams continuously flow by a perforated barrier through which the streams communicate. The perforations of the barrier are of a size to allow exchange between the two streams of undersized particles. As the size distribution of particles varies in the sample stream, the population of undersized particles in the reference stream will also vary, so that the two streams remain substantially in equilibrium as to the undersized particles. In most systems of interest, the rate of change of distribution in the sample stream will be relatively slow as compared to the establishment of the equilibrium between the two streams.
The pulp density of the two streams can be measured by various devices, which produce an electrical signal related to the pulp density. The electrical signal can be recorded and calibrated as an absolute measure of pulp density or, preferably, the two signals can be fed to circuitry, which according to the following mathematical formula will provide the mass fraction of undersized particles within the external slurry or pulp. Alternatively, the mass fraction can be determined manually in accordance with the following formula:
m,, mass of undersized particles m,- mass of solid particles of all sizes p, density of the reference pulp p density of the fluid p, density of the sample pulp p specific gravityof the solid in any system, the density of the fluid (usually water and usually about 1) and the specific gravity of the solid will be constant and known. Where large fluctuations in temperature may be encountered, thermostatting can be employed. Therefore, the only variables will be the density of the sample pulp and the density of the reference pulp, which are the measured values. Thus, from these two density values, the mass fraction can be determined.
By employing the information obtained as to the variation in mass fraction, this information can be fed back to the grinding operation, so that the grinding operation can be modified to maintain the desired size distribution.
The second expression on the right-hand side of the formula is the correction factor for the volume or space occupied by the particles greater than the predetermined size in the sample stream. In most cases, this correction factor is a number slightly less than one. It will therefore be usually satisfactory, to leave out the correction factor, to simplify the circuitry or the manual calculation.
The subject method is particularly useful in systems, where there is interest in the weight of particles below a predetermined size as compared to the total weight of particles. The subject method, therefore, finds particular application in the grinding of various ores, such as iron ore, copper ore, and the like. The subject method can also be useful, where phosphate rock or other rock is employed to form a final product such as phosphoric acid. Where the rate of this solution of the rock is a function of the size distribution, one can relate the time necessary for the solution to the degree of rock grinding required.
The slurries which are needed need only be flowable slurries. Therefore, the weight percent of solids may vary widely, generally not exceeding about 70 weight percent and usually being at least about 1 weight percent, more usually ranging from about 10 to 60 weight percent. Any carrier fluid may be involved, but for the most part, large industrial processes employ water. The particular size distribution of the particles is not significant, since one can employ a perforated barrier having any convenient size opening. For the most part, the particles of interest will be 5 mesh or smaller, frequently 40 mesh or smaller.
The apparatus which is employed has a compartment with two cells which communicate solely through a perforated barrier or sieve. The open area of the perforated barrier should be maximized, frequently being at least about 60 percent of the surface area to maximize the contact of the two streams, while still providing sufficient structural strength for the barrier. The size of the perforations or openings is chosen so as to be intermediate the size of particles which is being assayed.
In one cell, which is referred to as the internal reference cell, a slurry is introduced having particles of the same composition as the particles in the sample slurry and the same fluid as the sample slurry. The size distribution of the particles is chosen so as to be below some predetermined size of the size range of the particles in the sample to be assayed. Means for agitating the slurry in the internal reference cell is provided to prevent any gradient of particle sizes to become established. The means can be a recirculating pump, a stirrer and baffle walls, or the like. The significant factor, is that the pop ulation of particles near the surface of the barrier is substantially the same as that throughout the internal reference cell slurry. The internal reference cell slurry is a closed system, except to the extent that it communicates with the sample stream.
The second cell is connected to the sample stream, either in its entirety or a diversionary stream. The cell can be simply a conduit, having an inlet and outlet, and communicating intermediate the inlet and outlet with the internal reference cell through the perforated barrier.
Means are provided for measuring the pulp density, either continuously or intermittently in the two cells. Various commercially available apparatuses may be employed for the pulp density determination. An accurate and convenient apparatus for pulp density measurement is a nuclear pulp density gauge, which is commercially available from Texas Nuclear Company, a division of G. D. Searle Company.
The size of the sample cell may be varied widely and is not a significant factor in the accuracy of the determination. The internal reference cell volume in relation to the area of the perforated barrier will affect the rapidity of response and sensitivity of the system to variations in the size distribution of particles in the sample stream. Therefore, the ratio of the internal reference cell volume to the surface area of the perforated barrier should be minimized. The greater the variation in the pulp density in the internal reference cell slurry with an absolute increase or decrease by weight of solids, the more rapid the response and the more sensitive the measurement.
With the nuclear density gauge, it is desirable to have a relatively large sample in the path of the gauge. This can be achieved for both cells in a variety of ways. A relatively wide narrow cell may be employed. Alternatively, a sigmoidal conduit can be employed with a relatively long central arm in the path of the gauge. Other techniques may also be employed.
Since pulp density is being measured, it is important to avoid the presence of air bubbles. Any means of agitation should substantially avoid the presence of air bubbles, or a sump or other means provided for the removal of air bubbles. The particular manner of agitation and removal and prevention of the formation of air bubbles is primarily one of expedience.
For further understanding of the subject invention, the drawings will now be considered. In FIG. 1, a diagrammatic view is provided of a single channel particle size analyzer 10 using two pulp density gauges. The sample cell or conduit 12 is connected to the sample stream by any convenient means at inlet 14 and discharges the stream at outlet 16. The sample cell 12 has port 20 having an annular ring 22.
Internal reference cell 24 is fitted with a stirrer 26 and baffle 30. In this way, the slurry in the internal reference cell is continuously circulated, so as to maintain a substantially homogeneous distribution of particles in the slurry. The arrows depict a particular mode of flow for the slurry. The internal reference cell 24 has an annular ring 32 which is joined to annular ring 22 and retains sieve 34 rigidly in position.
A pulp density gauge 36 is positioned on sample cell 12, so that the beam from the pulp density gauge monitors the density of the sample slurry passing through cell 12.
Pulp density gauge 40 is positioned on internal reference cell 24 so as to pass a beam through the slurry in the internal reference cell 24. Each of the pulp density gauges 36 and 40 will provide electrical signals which can be recorded and employed for monitoring the mass fraction of undersized particles in the sample stream. Alternatively, the signals can be fed to a relatively simple analogue or digital circuit, so that the mass fraction may be determined directly.
In FIG. 2, a single channel particle size analyzer 42 is depicted employing a single pulp density gauge. Because of employing only one pulp density gauge, the slurry streams from the sample cell and the internal reference cell are alternately measured by the pulp density gauge and logic circuitry'provided for retaining the information, so as to make the necessary calculation for the mass fraction.
The particle size analyzer 42 has a compartment 44 with a sample slurry cell 46 and an internal reference slurry cell 50. A well 52 is provided into which is introduced sieve 54, which separates the two cells 46 and 50. The sieve 54 is held in holder 56 which provides a seal with the well 52, so that no leakage can occur between the two cells, and communication is limited to the perforations in the sieve 54. An inlet port 60 connects the sample cell 46 to the sample stream. Conduit 62 connects the sample cell to density gauge exposure cell 64, which discharges through conduit 86. The internal reference cell 50 is substantially a closed circuit. Outlet port 66 'is connected by conduit 70 to sump 72. The slurry is fed into the sump 72, so as to remove any bubbles which may accumulate in the slurry stream. The slurry in the sump 72 is then led by conduit 74 to pump 76, which recirculates the slurry to conduit 80, through density gauge exposure cell 82 and conduit 84 back to cell 50.
The internal reference slurry stream is in continuous flow, with the pump operating constantly. Conduits 84 and 86 with cell 82 and conduits 62 and 80 with cell 64 have sigmoidal shapes in the beam of the density gauge 92, so as to increase the path of the density gauge beam through the slurry. The density gauge 92 can be modified to alternately read the pulp densities of cells 64 and 82.
The density gauge provides a signal for computer 100, which has appropriate logic circuitry to retain the information as to the density of one stream, so as to be able to calculate the mass fraction of the undersized particles in the sample stream. If desired, the computer can be arranged to be connected to the grinding operation, so as to continuously control the grinding operation to insure a substantially constant population profile of particles.
In F IG. 3, a particle size analyzer employing two density gauges is depicted, which will be discussed only briefly, since the device substantially parallels FIG. 1. The single channel particle size analyzer 102 has an internal reference cell 104 equipped with a stirring motor 106, with the stirrer extending into circular baffle 110. The sample stream cell 112 is bolted through annular ring 114 to the annular ring 116 of the internal reference cell 104. The two annular rings 114 and 116, re-
tain sieve 120 in a fixed position, the sieve 120 providing the sole communication between the two cells.
5 Density gauges 122 and 124 are electronically connected to computer 126 and signal the computer as to the pulp densities of the two slurry streams on a continuous basis.
While the figures have been concerned with a single channel particle size analyzer, it is evident that a number of different channels could be employed, having perforated barriers with different size openings. Thus, one could continually monitor different mass fractions of a slurry stream and determine, not only the mass fraction of all particles below a particular size, but also the mass fraction of particles within a certain range. Such determination could be carried out with a single density gauge, by employing a plurality of compartments with different sieves and appropriate plumbing and switching, but more conveniently, a plurality of density gauges would be employed. In the latter instances, the device would be a plurality of individual devices, with the signals from each of the density gauges being fed into an appropriate recorder or computer.
. In order to demonstrate the effectiveness of the subject device, an experimental demonstration apparatus was devised which consisted of two Cs nuclear density gauges, two recirculation pumps, and a cassette that held a standard 8 inchdiameter shallow laboratory sieve. O-rings and rubber gaskets were used to prevent oversized particles from entering the internal reference pulp. The cassette was designed to permit easy changing of sieves and sieves of any standard opening sizes could be used. For test purposes, the sample or external pulp was recirculated. However, changes in external pulp density and particle size distribution were made by adding dry solids to the sump of the external pulp recirculating pump.
Tests were conducted on a, copper mill tailing as received (60% -lOO mesh) and on 1OO mesh tailing (p,
'= 2.6). Different particle size distributions were obtained by adding reground tailings to the primary tailing. For most of the tests, a 100 mesh sieve was used. The sieve area was 300cm and the volume of the internal reference pulp was 21.6 liters. This yields a large volume/area ratio of 72 centimeters, which is undesirable for obtaining a fast instrument response time.
The kinetics of undersized particle transfer through the sieve and the response time required to reach equilibrium after a disturbance in the external pulp was determined by a discrete addition of undersized particles (100 mesh) to the external pulp sump while operating. In this experiment, 2.0 kilograms of previously sieved 100 mesh tailings were added to the sump of the external pulp. This amount of solids should increase the pulp density on both sides of the sieve by 5 weight percent at equilibrium. There is an initial delay inthe measured increase in the external pulp density, because mixing in the external pulp is not instantaneous. This delay would not be encountered in normal use. In addition, because of the transfer of particles through the sieve to the internal pulp, the measured increase in the internal pulp density encounters a longer delay. The experimentally determined response sign was found to be adequate for mill grinding circuit control purposes, particularly in view of the considerable amount of stored solids and inertia in the grinding circuit. The response time could be reduced even further, by decreasing the sieve-area/internal-pulp-volume ratio compared with that used in the test apparatus.
The subject invention provides a simple bulk sensing method of continuous (on-stream) measurement of particle size pulps. The method is simple and direct and has several advantages over known bulk sensing methods. There are no moving parts, except for agitators or pumps to maintain particle suspension. The basic measuring equipment, conveniently nuclear density gauges, are commercially available with a high degree of accuracy and reliability.
The equipment is rugged, and except for the replaceable sieve component, abrasion resistant. Particle classification or separation by size is not involved and sieve blinding does not occur. The instrument output is a direct algebraic result of the pulp density measurements, that requires no empirical calibration, such as is required by ultrasonic attenuation devices. The method is not upset by changes in pulp density. The dynamic response time is adequate. The method can be applied directly to the entire flow of the measured stream, so that representative sampling problems are avoided.
Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be obvious that certain changes and modifications may be practiced within the scope of the appended claims.
What is claimed is: i
l. A method for analyzing a slurry stream, having particles of varying size, for particles below a predetermined size which comprises:
directing a first sample slurry stream having particles in a range of sizes past and in contact with a perforated barrier having openings of a size intermediate the size range of said particles in said first slurry stream;
directing a second reference slurry stream past and in contact with the opposite side of said perforated barrier having particles of a size at and below the openings in said perforated barrier;
measuring the pulp density of said first and second slurry streams; and
determining the mass fraction of particles at and below said predetermined size in said first slurry stream from the pulp density determinations of said first and second slurry streams. v
2. A method according to claim 1, wherein said first stream has less than about 60 weight percent solids.
3. A method according to claim 1, wherein said first stream is derived from the grinding of ore.
4. A method according to claim 1. wherein said openings in said perforated barrier are not greater than about 40 mesh.
5. A particle analyzer comprising:
a compartment having first and second cells for containing a sample slurry and a reference slurry respectively, and a perforated barrier having openings of predetermined size between said cells through which said cells communicate;
inlet and discharge ports in said first cell;
means for agitation of said slurry in cooperative engagement with said second cell; and
means for measuring the pulp density of said slurries in said first and second cells.
6. A particle analyzer according to claim 5, wherein said measuring means are first and second nuclear density gauges.
7. A particle analyzer according to claim 6, wherein said first and second nuclear density gauges are mounted on said first and second cells respectively.
8. A particle analyzer according to claim 5, having in addition:
a first conduit in a fluid receiving relationship with said discharge port of said first cell and having a sigmoidal region;
inlet and discharge ports in said second cell;
a second conduit connecting said inlet and discharge ports of said second cell having a sigmoidal region;
pump means for circulating said reference slurry second conduits.