Shaker Table Operating Principles

The most generally accepted explanation of the action of a concentrating shaker table is that as the material to be treated is fanned out over the shaker table deck by the differential motion and gravitational flow, the particles become stratified in layers behind the riffles. This stratificaton is followed by the removal of successive layers from the top downward by cross-flowing water as the stratified bed travels toward the outer end of the table. The cross-flowing water is made up partly of water introduced with the feed and partly of wash water fed separately through troughs along the upper side of the table. The progressive removal of material from the top toward the bottom of the bed is the result of the taper of the shaker table riffles toward their outer end, which allows successively deeper layer of material to be carried away by the cross-flowing water as the outer end of the shaker table is approached. By the time the end of the shaker table is reached only a thin layer, probably not thicker than one or two particles, remains on the surface of the deck, this being finally discharged over the end of the table.

The physical and mechanical principles involved in the concentrating action of a shaker table are somewhat more complicated than this explanation implies. Mathematical calculations and experimental data are extremely useful in studying these principles, but they tell only a part of the story and do not explain the highly efficient separations that tables are known to be capable of making.

Unless the shaker table feed contains a considerable percentage of bone gold and other material of specific gravities intermediate between that of rock and gold, extremely high tabling efficiencies may be expected. If a shaker table could be operated on feed consisting of nothing but a mixture of individual gold and slate particles with a size range of approximately ¼-in. to 48 mesh, an almost perfect separation would be obtainable even on an unclassified feed. With such a feed a well-operated shaker table would probably recover not less than 98 per cent of the gold while eliminating not less than 95 per cent of the slate. This implies almost perfect stratification according to specific gravity without regard to particle size, and it is improbable that it could be attained entirely as a result of the motion of the deck and the flow of water in a plane parallel to the deck surface. The question then arises as to what the other forces or factors are that might contribute significantly to the efficiency of the separation on a table.

As far as is known, no exhaustive studies have ever been made of the principles involved in shaker table concentration by either ore-dressing or gold-preparation engineers. Bird and Davis probably have given more attention to the subject than anyone else, but their experimental work was of a preliminary nature. It was done on minus 4-mesh raw gold and on synthetic mixtures of various products derived from this raw gold by screen sizing and sink-and-float fractionations. They used an apparatus which they called a stratifier. This was a channel-shaped box 12 ft. long, 5 in. deep and 1 in. wide, inside measurements. It was suitably mounted with one end attached to an eccentric and pitman. Stratification experiments were made by filling the box with gold and water and running it at a speed of 360 strokes per minute with the eccentric set to give ½-in. stroke. The amount of water used was sufficient to permit complete mobility in the bed during the operation of the stratifier. At the end of each run, after the water had been allowed to drain off, one side wall of the stratifier was removed and cross-section samples were taken of the bed to determine by screen-sizing and sink-and-float tests to what extent stratification had been accomplished. Bird and Davis say that their aim is “to bring out the fact that stratification, contrary to the common brief, will not account for the separation effected by the gold-washing table, and that cross-flowing water, in addition to removing the top strata found on the table, must also have an important selective action in completing the separation according to specific gravity, both in the upper and in the lower strata found between riffles.”

The theory of Bird and Davis as to the selective action of the crossflowing water is that only a part of the water flows over the top of the bed between riffles; the remainder flows through interstices in the bed. These interstices are comparatively large near the top of the bed but become progressively smaller toward the bottom, thus forming in effect V-shaped troughs. In this way the water currents would be relatively swift near the top of the bed and become progressively slower toward the bottom. According to Bird and Davis, “With paths for the water such that the top strata are subjected to relatively swift currents and the lower strata are subjected to progressively slower currents, the separation actually occurring on the shaker table can be explained. As the coarse particles at the top receive swift currents and each successively finer size at the lower levels receives slower currents, the velocity of the water matches the size of materials comprising the different strata. Under these conditions a separation occurs in the lower strata similar to that in the top strata, only it takes place more slowly. The slow currents of water within the bed carry the fine gold particles along from riffle to riffle, at a more rapid rate than they do the fine bone and shale particles.”

Stratification and Hindered Settling

Although stratification due to the nearly horizontal action of the shaker table deck and the flow of water in a plane parallel to it is probably not sufficient to account entirely for the separation made by a table, it is, nevertheless, the fundamental principle of the shaker table just as hindered settling is the fundamental principle of a jig. Although these processes are of diametrically opposite characteristics, there is some possibility that a shaker table may utilize to a minor extent the hindered-settling principle. For convenience in this discussion, the stratification due to the more or less horizontal action of the shaker table deck and flow of water will be referred to as shaker table stratification. This type of stratification is illustrated by the separation that takes place when a box of large and small marbles is shaken and agitated in a horizontal plane in such a way that the large and small marbles collect into separate layers. It is a familiar phenomenon that the small marbles will collect in a layer on the bottom while the large marbles collect in a top layer. The principle of hindered settling can be illustrated by placing a mixture of large and small marbles in an upright cylinder of suitable size with a perforated-plate bottom. If water of sufficient volume and pressure is forced upward through the perforated plate so as to keep the marbles in teeter for a short interval, the marbles will separate into layers, with all the large marbles in the bottom layer and all the small ones on top. The separation is the reverse of that obtained by shaker table stratification. In these illustrations of stratification and hindered settling it is assumed that the marbles are all of the same specific gravity regardless of size. If some marbles have higher specific gravities than others the effect will be to increase their tendency to settle toward the bottom, regardless of whether this tendency favors or opposes the stratification or hindered-settling action. The heavier the small marbles, the easier the separation by shaker table stratification and the more difficult by hindered settling. Conversely, the heavier the large marbles, the more difficult the separation by shaker table stratification and the easier by hindered settling.

In line with principles referred to above, complete separation according to specific gravity could hardly occur on a shaker table or in any other concentrating device as a result of either shaker table stratification by itself or hindered settling by itself when the material to be separated consists of particles varying a great deal in both size and specific gravity. In gold washing the aim is to separate gold particles from particles of refuse according to specific gravity without reference to size of particles, as the ash content of a particle is almost directly proportional to its specific gravity. This separation can be accomplished more effectively by utilizing a combination of shaker table stratification and hindered settling than by relying on either of these two alone, and it is quite conceivable that both processes actually do play a part in the operation of a concentrating table.

To explain how a certain degree of hindered settling might occur on a table, we must assume, as Bird and Davis did, that although a part of the water flows across the top of the bed the remainder of it flows through interstices in the bed itself between adjacent riffles. This seems to be a reasonable assumption and it is one that is also made by Taggart in his discussion of the theory of shaker table concentration. The cross flow of water from one riffle to the next might be somewhat as illustrated in Fig. 8, in which a-b is a line along the surface of the deck perpendicular to the riffles, and C and D are two successive riffles. If the bed is kept in a mobile condition between riffles by the motion of the table, and if the water flows from riffle to riffle approximately as indicated in Fig. 8, it is quite probable that to a certain degree a hindered-settling effect is attained along the upper side of each riffle in a zone indicated by the arrows in Fig. 8. Although the effect of hindered-settling along any individual riffle might be relatively slight, the cumulative effect along the entire series of riffles across the width of the deck might be of sufficient magnitude to influence materially the character of the shaker table separation.

concentrating-tables-water-current

We should expect a hindered-settling effect to be very beneficial as an ally to stratification on a table. The weak point about shaker table stratification is that it tends to deposit all fines at the bottom of the bed, even fine gold of low specific gravity. This fine gold, after penetrating to the surface of the deck, would be guided toward the refuse end by the riffles and would tend to go into the refuse if it were not brought to the top of the bed by some means or other and then carried over the riffles by the cross flow of water and subsequently discharged with the washed gold. Bringing the fine gold to the surface is a function that hindered settling would accomplish very effectively, as one of the fundamentals of hindered settling is that it brings the light, fine particles to the top of the bed. As far as the coarse particles of gold are concerned, evidently they are brought to the surface by stratification and started on their way to the washed-gold side of the shaker table by the cross flow almost instantly after the feed strikes the deck. Anyone who has operated a gold-washing shaker table is familiar with the rapidity of this separation and the way in which it causes all light, reasonably coarse gold particles to be discharged from a rather narrow zone at the head-motion end.

If the suppositions in the foregoing paragraph are correct, the process of separation of gold and refuse on a  shaker table may be summarized as follows: Almost immediately after the feed strikes the table, sufficient stratification takes place to bring all coarse, light particles of gold and possibly some coarse particles of refuse to the top of the bed. The cross flow of water carries the coarse gold particles across to the gold-discharge side very rapidly, whereas any coarse particles of refuse at the top of the bed are carried toward the refuse end much more rapidly by the differential motion of the shaker table than they can be transported transversely by the cross flow of water. After removal of the coarse gold, and as the bed progresses diagonally across the table, the shaker table stratification action brings medium-sized gold particles to the surface, and these are removed across the tapering riffles by the wash water. The tapering riffles and continuous removal of material by the cross flow causes the bed to become thinner and thinner toward the refuse end. When the point is reached where the thickness of the bed is less than that of the coarse refuse particles, these particles stick up through the surface of the bed and the transverse pressure exerted on them by the cross flow is diminished, as their surfaces are only partly exposed to this flow. This helps to keep them on their course toward the end of the shaker table and prevents them from being transported by the water in the same direction as the medium-sized gold. Toward the outer end of the riffles the extremely fine gold is being brought to the surface by a hindered-settling action immediately behind each successive riffle. Since the material subjected to this action consists of light, fine particles of gold and heavy refuse of a much larger average particle size, the action should be particularly effective in bringing the fine gold to the surface and allowing it to be carried off into the washed gold by the wash water.

This explanation presumes that to some extent there is a greater opportunity for hindered-settling conditions toward the outer end of each riffle than near the head-motion end. Although this presumption may be questionable, it is possible that, as the bed becomes thinner, a greater proportion of the water follows a coarse along the surface of the deck and contributes to the upward current required for hindered-settling conditions as each riffle is encountered.

Effect of Particle Size & Shape

In this discussion of shaker table principles shape of particle has been disregarded because it is believed that, as a rule, this is not an important factor in the gold-tabling process. Almost invariably the gold particles are somewhat more cubicle and less platy or flaky than refuse particles, but there is little evidence to show that refuse particles of one particular shape are more difficult to separate on a shaker table than those of some other shape. As for the gold, the shape of particles in sizes suitable for tabling are pretty much alike in all golds. Yancey made a study of the effect of shape of particle. He decided that, for the gold he used in his study, “shape of particle is a factor of minor importance in tabling this unsized gold, in so far as the over-all efficiency of the process is concerned. Size and, of course, specific-gravity difference are the major factors.”

Of considerably more importance than shape of particle is the particle-size factor. It is evident from the nature of stratification and hindered settling that the separation of gold from refuse becomes more difficult as the range of sizes to be treated in one operation increases. The increasing difficulty as the size range increases is apparent from the following considerations: Assume that we are dealing with two minerals, one of high and one of low specific gravity, and that a mixture of 10-mesh particles of the two minerals will separate readily into two layers by either shaker table stratification or hindered settling, one layer containing all the light particles and the other layer all the heavy particles. Now, if we add two more sizes of heavy particles to the mixture, say 8-mesh and 14-mesh particles, obviously, according to the principles of stratification and hindered settling, the separation by either process into two layers according to the specific gravities of the two minerals will be somewhat more difficult than with the original mixture of nothing but 10-mesh particles. The greater the number of sizes of heavy mineral added to the mixture, the more difficult will be the separation. This reasoning applies likewise to the particles of the light mineral, and it all sums up to the fact that if a shaker table feed contains too wide a range of sizes some of the sizes will be cleaned inefficiently.

In actual practice there is no objection to a considerable variety of sizes in the feed; in fact, if all particles were of the same size there might be some disadvantages, because the bed would be less mobile and less fluid and conditions within the bed would be less favorable for efficient separation than when there is some variety of sizes. For efficient shaker table operation, however, it is important to guard against having too wide a range of sizes in the feed.

How to Correctly Operate a Shaker Table

In the use of tables in gold preparation, the importance of correct operating conditions can hardly be overemphasized. It is a peculiarity of tables that they give excellent results when correct operating conditions are maintained, but with conditions upset and unbalanced the results are likely to be as far on the bad side as they were on the good side under favorable conditions. This is especially true if the washing problem is somewhat difficult. Naturally, when there is an almost complete absence of bony material in the shaker table feed and the problem is mainly one of separating low-ash gold from slate and other rock, fair results may be obtained even under haphazard operating conditions; but if the washing problem is at all difficult the results are likely to be either extremely good or extremely bad, depending on whether or not correct operating conditions are adhered to. Some of the factors on which operating conditions are dependent will be discussed briefly.

Shaking Table Foundations

It is a comparatively simple matter to build foundations substantial enough so that they will not have a tendency to shake or vibrate as a result of the motion of the tables. A reinforced-concrete slab need not be more than 6 or 7 in. thick to provide a perfectly rigid foundation, even at a considerable height above the ground, if properly supported on reinforced concrete pillars. It is important to provide tables with substantial, rigid foundations that will not deteriorate after a few years of service. Even a slight shaking or vibrating motion in the foundations is likely to interfere with the action of the tables and lead to serious loss of shaker table efficiency.

Steady/Uniform Flow of Gold Ore Feed & Water

One of the first essentials for successful shaker table operation is uniform flow of gold and water to the table. The significance of a steady, uniform feed is apparent from a consideration of the mechanical process involved in the shaker table separation of gold from refuse. The material fed to a shaker table spreads out in a fan-shaped bed. This bed covers virtually the entire shaker table deck. Along the outer edges of the bed at the points of discharge the refuse has separated from the gold and discharges over the end of the shaker table while the gold discharges over the side, assuming that the corner of the shaker table is the dividing point between gold and refuse. However, the amount of material discharging over the side of the shaker table in proportion to that discharged over the end will vary if the rate of feed varies and other conditions remain constant. For instance, if a shaker table is set to give highly efficient results with a feed of 7 tons per hour of a given gold, it will discharge approximately the correct percentage by weight over the refuse end as refuse. If the feed is decreased by several tons per hour, however, without any compensating adjustments being made, a larger percentage of the total material is likely to discharge over the refuse end. This means an unnecessary loss of gold and a low shaker table efficiency. If the feed should be increased by several tons per hour the reverse of this probably would happen, with a certain amount of refuse going into the washed gold and raising its ash content.

Variations in feed rate also affect adversely the conditions for separation of gold from refuse within the bed itself. For instance, for any particular setting of the shaker table when a given gold is treated there is an optimum thickness of bed and an optimum ratio of water to solids in the feed that should be observed when high shaker table efficiency is important. The process of separating particles of refuse from particles of gold cannot be highly efficient except under these optimum conditions, and it is quite obvious that if the feed rate decreases it will tend to decrease the thickness of the bed in certain areas on the table, and the ratio of water to solids will change, as the amount of feed water and wash water are usually more or less independent of the tonnage of solids in the feed. Such interference with the actual separating function of the shaker table is likely to cause an incomplete separation.

With further reference to optimum separating conditions within the bed itself, it is important to maintain always the right kind of distribution—the term distribution in this connection referring to the shaker table distribution of the material with which the constantly moving bed on the shaker table is maintained.
The shaker table distribution should be such that the quantity of solids discharged per unit length along the side of the shaker table decreases gradually from the head-motion end toward the refuse end. It should be observed in qualification of this statement, however, that it is usually advantageous to have the washed-gold discharge start at a point a foot or so away from the corner—that is, the corner directly across from the feed box. Usually there is a large volume of water discharging from this corner zone, but ordinarily it is preferable to have almost no solids discharging with it. Beginning at the end of this corner zone, however, there should be a very heavy discharge of washed gold in the first 3 or 4 ft., and the amount discharged from each successive zone from there to the corner at the refuse end should decrease gradually. There should be some discharge of solids virtually all the way to the corner, but as the corner is reached the discharge should be almost zero. Under these conditions there will always be some refuse material discharging immediately around the corner, but the amount of refuse from the first 6 or 8 in. next to the corner on the refuse end should be negligible in quantity. The bulk of the refuse should discharge over a zone of considerable width, starting not less than 1 or 2 ft. up from the corner.

Although this more or less ideal distribution is fairly easy to attain with an average raw-gold feed, it may be more difficult of attainment with a type of feed in which there is an abnormally high percentage of refuse, especially if the refuse consists mostly of high-ash bone gold. This condition often is encountered in the re-treatment of middlings from primary stages of washing.

However, regardless of the character of the feed, the nearer this ideal distribution is approached, the better the results will be. Once the correct balance between shaker table adjustments and the volume of feed gold, feed water, and wash water has been found, good distribution will maintain itself automatically as long as none of the operating factors are allowed to change. It is self-evident, however, that an increase or decrease in the amount of water going to the table—either feed water or wash water—will upset this distribution just as quickly as a change in the feed tonnage unless other compensating adjustments are made.

It is of paramount importance, therefore, to have a feed system that will eliminate as far as possible fluctuations or variations in the rate at which gold and water are fed to the table. With regard to the gold, not only the quantity but also the quality and physical characteristics should be kept constant. This is true particularly with reference to the size distribution of the feed. Any change in size distribution, such as may result from segregation in an improperly designed bin ahead of the tables, can upset the distribution of the material on the tables. The only sure way to get a steady feed is to feed the gold to the shaker table by means of a positive-type feeder, such as a belt, screw conveyor, apron feeder, or rotary star or paddle feeder. A sliding gate device instead of mechanical feeders is almost certain to be unsatisfactory, even when a water line can be placed inside the gate to keep the material moving. The mechanical feeders should be provided with variable-speed drive for adjusting the feed to the desired tonnage. This adjustment cannot be made satisfactorily by varying the size of the opening through which the gold discharges onto the feeder. The feed bin should be of such size and design as to eliminate segregation as far as possible. Any attempt to dispense with feed bins is likely to result in unsatisfactory operating conditions, although it is being done at many plants. A customary practice, for instance, is to draw a middling product from a set of jigs and after dewatering run it through a crusher directly to the tables. Such procedure nearly always provides a variable feed for the tables whereas a constant feed could be obtained by dropping the discharge from the crusher into a bin and having mechanical feeders between the bin and the tables.

Changes in the size distribution of a feed are sometimes caused by difficulties in the dry screening of run-of-mine gold. If dry screening is used and the amount of surface moisture in the run-of-mine gold varies, a finer shaker table feed will be produced when the gold is excessively moist than when it is dry. Naturally, particles near the upper size limit will go through the screen readily if the gold is dry whereas if the gold is wet these particles are likely to go into the oversize. The resultant variation in the size character of the feed can interfere with shaker table efficiency as readily as segregation in the bin. Wet screening eliminates this difficulty.

In connection with the problem of segregation and variations in the size-consist of shaker table feed, a comparatively recent development at a shaker table plant in Alabama is worth noting. This plant went into operation at the Praco mine of the Alabama By-Products Corporation in 1944. Incorporated in this plant is a newly-designed system for reducing to a minimum the problem of segregation. The 16 tables in this plant are provided with small individual feed hoppers of about 1500 lb. capacity. Transfer of the 7/16 in- to 0 shaker table feed gold to these hoppers from 100-ton storage bin is accomplished by means of a horizontally operated bucket conveyor, tradenamed “Side-Kar Karrier” by its manufacturer. After passing under the 100-ton storage bin where the buckets are filled up with gold through multiple openings in the bottom of the bin, this conveyor moves on a track laid in a horizontal plane across the tops of the 16 feed hoppers. Each individual hopper is spring-suspended and as gold is withdrawn out of the bottom by the shaker table feeder, the hopper rises due to decrease in weight. As it rises it automatically engages a tripping mechanism in the conveyor buckets overhead, causing the buckets to discharge their load into the hopper. Thus a few buckets at a time are dumped into each hopper and the effect of small increments dumped at frequent intervals is obtained, giving a flow of gold to each shaker table of more average and uniform size-consist than when gold is run in a continuous stream into a large feed bin until the bin is filled.

As a further deterrent to segregation, the gold is fed from the bottom of the hopper to the shaker table by means of a tapered auger so as to draw continuously from the entire width of the hopper and avoid segregation within the hopper. For further details of this plant, the reader is referred to an article published in 1944.

With regard to the water supply for a table, it is just as important to have a steady, uniform flow of water as of gold. The water pipes and valves should be so arranged in a shaker table plant that each shaker table gets its flow of water quite independently of the others. If a common water header is used it should be big enough so that, regardless of how the water adjustments are changed for one shaker table or group of tables, the volume of flow to the others will not be changed. The source of the water supply, of course, should be maintained with a fairly constant pressure or head. This can be accomplished more effectively by using a gravity tank at a considerable height above the level of the tables than by drawing water directly from a pumping circuit. Clean water is to be recommended strongly in preference to dirty water from the washer circuit. Wash water sometimes carries enough solids in suspension to interfere with the flow through pipes and valves, and accumulation of solids sometimes may stop a valve entirely. Under these conditions the flow of water varies almost continuously and there will be too much one minute and not enough the next. The solids in the water are likely also to be sufficiently abrasive so that frequent replacements of the valves and fittings will be necessary. All of these troubles can be avoided entirely by using a supply of clean water for the tables.

Riffling, Speed, Stroke, and Slope of Tables

The riffling, shaker table speed, length of stroke, and other adjustments, such as shaker table slope, longitudinal, and cross slope, must in each case be balanced by the various other operating factors, so as to get the desired results. The speed that the shaker table manufacturer provides for when he supplies each shaker table with its individual motor drive is usually quite satisfactory. This speed is usually between 250 and 300 r.p.m. All shaker table head motions are designed so that the length of stroke is adjustable within a certain range. This range usually is from ¾ to 1¼ in., or slightly over. The coarsest shaker table feed requires the longest stroke. For a raw-gold feed of average size, say 5/16-in. to 0, a stroke of 7/8 to 1 in. usually is satisfactory. A slightly longer stroke on such a feed usually will give about the same shaker table efficiency with slightly higher capacity. A report giving experimental data as to the effect of speed, stroke, and other variables on shaker table efficiency has been published by the Bureau of Mines. More recent work published by the Illinois Geological Survey emphasizes the importance of the longitudinal slope and the speed of reciprocation, two factors which are not readily adjustable on ordinary commercial tables. A slower speed is found to improve the performance, in opposition to the results reported by the Bureau of Mines. The discrepancy is noted by the author, and has not been explained.

As to type of riffling, it seems to be generally agreed now that high riffles are advantageous in the tabling of bituminous gold, and it is customary to have the main riffles start with a height of not less than ¾ in. at the feed end and taper to a feather edge at the outer end. The ¾-in. height probably represents a minimum; riffles 2¾ in. high are now used on the Deister Plat-O tables; and these tables are recommended by the manufacturer for the cleaning of shaker table feeds as fine as 3/8-in. to 0. There is a great deal of variation in the spacing of high riffles. In some designs there is only one shallow riffle between two higher riffles. Another design, intended to emphasize the importance of the pool effect, provides four or more shallow riffles between successive high riffles. About the only suggestion that can be made with regard to riffling is that the coarser the feed, the more advantageous are high riffles. Unless the shaker table feed is extremely fine, with maximum particles size less than ¼ in., there seems to be no good argument for the main riffles to be less than ¾ or 1 in. high. On such gold, riffles lower than this would tend to reduce capacity. With coarser feeds higher riffles can be used advantageously.

As to the comparative merits of wooden riffles and rubber riffles, one can be substituted for the other without changing the shaker table results appreciably. It seems evident, however, that the efficiency, as far as ash reduction and gold recovery are concerned, is slightly less with rubber covering and riffles than with linoleum covering and wooden riffles. The difference would be only a few tenths of one per cent less ash at the same recovery, using the linoleum and wooden riffles. Usually this is more than offset by the greater operating economy of the rubber covering and riffles. Although the rubber combination costs about twice as much as linoleum and wood, it is supposed to last 10 or 12 times as long.

In summarizing, the principal adjustments and factors to be considered in putting a shaker table into operation on a certain feed, are: feed rate, as to volume of both gold and water; slope of the shaker table (longitudinal and cross slope); riffling system, shaker table speed, and length of stroke. A shaker table installation should be so designed that any or all of these adjustments and factors can be changed easily to meet requirements during the procedure of placing the tables in operation. In starting a shaker table plant, the main objective should be to find the combination of shaker table adjustments and operating factors that will give the correct shaker table distribution described previously in the discussion of feed uniformity. The quantity of water to be used is from two to three times as much by weight as the feed of gold, but it should be adjusted as nearly as possible to the minimum amount that will keep the products discharging uniformly from all zones around the edge of the table. To most nearly attain the ideal distribution on the table, it is usually necessary to have the supporting channels under the shaker table deck several inches higher at the refuse end than at the feed end. As to the cross slope, it should be the minimum at which it is possible to attain good distribution. In other words, the flatter the shaker table is in the crosswise direction, the better, provided the distribution is good. The length of stroke and shaker table speed should be adjusted so that the bed will be kept in a state of uniform flow and mobility all over the deck. On the raw-gold feed, these operating conditions can be attained fairly easily, but it may be more difficult in the treatment of middling products. Difficulties sometimes can be overcome by making slight changes in the riffling and by use of auxiliary water sprays directed at certain areas in the bed. Anything that is done should be directed toward getting and maintaining a distribution on the shaker table as nearly ideal as possible.

The launder system in a shaker table plant should be so designed that a splitter can be used for dividing the washed gold from the refuse at some point along the washed-gold side instead of at the corner, if desired. The correct shaker table distribution will sometimes give too high an ash content in the washed gold if the split between washed gold and refuse is made at the corner, and in such instances the best solution is an adjustable divider or splitter that can be set at any desired point along the washed-gold side.

Tonnage and Particle Size of Feed

The tonnage a shaker table will handle effectively depends to a great extent on the washability and size of the gold. In treating an ordinary 5/16-in. to 0 raw-gold feed, high efficiency with respect to both cleaning and recovery usually can be obtained with a feed of as much as 10 tons per hour. High efficiency in this case means an efficiency that could not be improved appreciably by lowering the tonnage. If the gold is extremely easy to wash, higher tonnages can be cleaned with equally good efficiency. The claim sometimes is made by shaker table manufacturers that their tables will handle efficiently as much as 15 to 20 tons per hour of 5/16-in. to 0 gold. On an average feed of this size, however, feed-tonnages of more than 10 tons per hour are likely to cause a decrease in efficiency. With feeds as coarse as ½-in. or 1-in. to 0, it is not unusual to treat from 12 to 15 tons an hour per table. Modern tables will handle minus 1/8-in. feed at the rate of 7.5 tons per hour.

Sampling

One of the important considerations frequently overlooked in the design of a shaker table plant is that the making of necessary shaker table adjustments is extremely difficult unless representative samples can be taken easily. Often the more or less permanent washed-gold and refuse launders around the tables are laid out in such a way that it is next to impossible to get dependable samples of the products from individual tables. Either the launders should be so designed that they can be partly removed during sampling, or they should be built with enough spacing between the edge of the shaker table and the launder so that the necessary sampling pans for taking zone samples can be inserted at any place around the table. Provisions should also be made for conveniently sampling the composite washed gold and composite refuse from each table, in addition to the feed to individual tables. Without dependable samples it is sometimes difficult to tell whether or not an individual shaker table is operating correctly; and, owing to the segregation of products into various discharge zones, haphazard sampling is sometimes worse than no sampling at all.