I need to further comment on the influence of surge storage upon plant operation. The provision of adequate surge storage at the right places in the flow line is of sufficient importance to warrant a little further consideration.
It is desirable, in the crushing plant, as it is in any production process, to achieve a smooth and uninterrupted flow through the plant so that the reduction crushers and sizing screens will be operating at all times under uniform load conditions, and at full capacity. It is not always possible to feed the plant in this manner; a smooth flow of feed to the primary crusher is the exception, rather than the rule; therefore, if we are to smooth out the flow in the plant itself we must have elasticity in our flow line, and the logical way to obtain it is to provide surge storage ahead of each crushing stage. We can, of course, get around this by providing enough crushing capacity to absorb the peaks, but this method is expensive p. also it throws a variable load on-the screens; and screens do not perform at maximum efficiency under such a load.
While the case for an ample surge storage is clear cut enough, the question of just what constitutes an adequate storage is not always an easy one to answer. It is difficult to visualize beforehand just how smoothly a projected scheme of quarry or pit operation is going to function, and impossible to forecast accidental delays in loading, transportation, and feeding. About all we can do is to rough out our plan of operation, and allow a comfortable margin for contingencies.
For example, if it is calculated that one trainload if rock will be de live red to the plant each one-half hour, on the average, and our primary crusher has been chosen with a view to getting this tram through in ten minutes, then we must, enviously. either provide stage crushing to match the high capacity of the primary or, if we want to get the most work out of our investment dollar, set up a surge capacity in the plant to absorb the peak load and feed it through the plant steadily over the entire 30 minute cycle. That is the essence of the problem: to keen as many as possible of the plant units rummy under uniform mad and at full capacity.
As a a example of one extreme in the need for surge capacity we have those plants whose primary and often secondary units operate only one shift of 8 or 10 hours/day, whereas the re-crushing and screening units run full time, or nearly so. The problem of calculating surge storage for such an operation is a simple one if operating conditions, such as character of feed and product size, remain unchanged, and if the feed rate to the primary unit is fairly constant from shift to shift. Where all of these conditions a re sub yet to significant variation the calculation is a more complicated one: we must calculate the need for each combination of conditions, and provide a surge storage that will take care of the maximum.
Regardless of how extreme the requirement for surge capacity may be it is usually unnecessary to provide an extensive storage ahead of more than one crushing stage. If the plant units are property balanced the :low rate can be smoothed out by regulating the feed from this one point. This does not necessarily mean that mechanical feeders need be used to regulate the flow; the crushers, will do this for the succeeding stages of the plant if the stage at which the regulating is done is the “bottle-neck” in the how line, i.e., the ‘t.age of least capacity. Naturally, this regulating stage should be as near the head end of the How line as possible, and, if it were possible, or practicable, to achieve an “idea!” plant layout, each successive stage would have a capacity slightly in excess of the preceding stage. 
Any multi-stage crushing plant will have a “bottle-neck” stage, and it is this stage which establishes the production rate of the entire plant; therefore, it should be kept busy, and the only way this can be done, if the feed to the plant is a fluctuating one, is to provide sufficient surge storage ahead of it to absorb such fluctuations, at least, such as may be exported under normal operating conditions.
We have mentioned elsewhere that some operators of gravel plants have realized the desirability of incorporating in their flow line a surge storage of sufficient proportions to compensate for delays of considerable duration, such as might be caused by a major breakdown of the loading equipment. There is no reason why this same expedient cannot be applied in a quarry operation; following the primary, or secondary, crushing stage. Aside from the production insurance it affords, such a stockpile is an excellent point from which to regulate the feed to the reduction crushing stages and screens, by means of one or more mechanical feeders. With such regulation the need for surge storage at other points in the how line is minimized.
Already, I have discussed only that type of surge storage which lies directly in the flow line, a “series” storage, as if were. There is another type which can be utilized to advantage in certain operations, particularly in those gravel-pit operations where the gradation of the pit-run material is subject to considerable variation. This type of storage might be called the ”bleeder” system, inasmuch as it consists of drawing off an intermediate size, or sizes, of product the amount of which, at any time, runs in excess of the capacity of the reduction crushers. Then, during periods when the pit-run material is running to tines, this storage is drawn upon to keep the reduction crushers busy. This is an excellent system for smoothing out plant production, especially in those plants which intermittently run heavily to the production of fine sizes of crushed rock. It is difficult to forecast the requirements for such a system: the application can best be made in a going plant where the operator ‘mows his pit and his requirements for certain sizes of product.