In what way must development tend in order to increase the number of hits at battle ranges?
One frequently hears after a practice: "We straddled the target and would have made a number of hits had the dispersion been less." From others is heard: "Our salvos were beautifully grouped, but they were not on the target. More hits could have been made if dispersion had been greater."
In any analysis of gun fire it is evident that the hitting of a target depends on two factors:
(a) Range error and lateral error;
(b) Dispersion, both vertically and laterally.
For convenience they may be referred to as fire-control errors and dispersion errors.
At target practice an attempt is made to place the mean point of impact of each salvo exactly in the center of the target, but this is impracticable of attainment for obvious reasons, and it is found that the mean point of impact is either short of the target or beyond it. Similarly, the mean point of impact is either to the right or the left of the centre of the target.
On each salvo, therefore, there is a definite range error, and a definite lateral error. The average of these errors for a target practice gives an average range error and an average lateral error for each firing ship.
It is evident, too, that each salvo will be dispersed both vertically and laterally.
These conditions must be fully appreciated; the two kinds of errors (fire-control errors and dispersion errors) occur on every salvo fired, and they embrace every error that exists in the firing of guns.
These two errors vary on each salvo, and are closely associated with each other.
Let us consider first (a) the range error and the lateral error.
If the correct sight-bar range and the correct deflection were known for each salvo, it is evident that hitting would be a maximum when there was no dispersion, vertically or laterally. That is, with known sight-bar range, and known deflection, all dispersion should be wholly eliminated.
The range and deflection are not accurately known, however; they vary from salvo to salvo. The spotter, with the aid of fire-control instruments, attempts to keep the range and deflection accurately, but errors exist on each salvo.
Admitting then that there must be a range error and a lateral error on each salvo, it is important to observe what effect on hitting is caused by dispersion.
(Dispersion is a very loosely used word. So far as concerns gunnery it should be used to indicate the average distance of shots from the mean point of impact; it is the mean vertical dispersion, or the mean lateral dispersion. If it is used to indicate the distance between the outer shots of a salvo, it becomes confusing; furthermore, the distance between the outer shots of a salvo is of little consequence, and has not the mathematical significance conveyed by the term "mean dispersion." Dispersion will therefore be used to indicate mean vertical dispersion, or mean lateral dispersion.)
It will be found that a flat trajectory gun has large horizontal dispersion, but small vertical dispersion. At longer ranges the horizontal dispersion decreases, but the vertical dispersion increases. We are concerned principally with vertical errors, and it avoids confusion and makes the subject much clearer to consider all errors as either vertical or lateral. All horizontal, or range errors are therefore reduced to vertical errors to simplify the problem.
If it were known that on any salvo there would be a definite range error and a definite lateral error, it can be shown mathematically that, neglecting the smaller terms, the greatest possible number of hits would be made when the dispersion of the salvo is:
Dv = ½ √8.75 Ev2 + s2 /?
Dl = ½ √8.75 El2 + w2 /?
(1) Ev = mean vertical deviation of mean point of impact of salvos (range error) in feet.
(2) El = mean lateral deviation of mean point of impact of salvos (lateral error) in feet.
(3) Dv = mean vert. deviation of gun (dispersion) in feet.
(4) Dl = mean lateral deviation of gun (dispersion) in feet.
(5) s = height of target, in feet.
(6) w = width of target, in feet.
For practical purposes the dispersion that produces the greatest number of hits may be taken as 80 per cent of the fire-control errors.
If the dispersion of the salvo is greater or less than the values given, the hits will decrease.
On every salvo there is a range error and a lateral error, and consequently there is a definite dispersion for each salvo that will give the greatest number of hits.
It is evident that the range error and the dispersion are closely associated with each other. Hitting cannot be increased by reducing the dispersion, or increasing it, unless the effect on the range error is considered—but, on the other hand hitting can always be increased by reducing the fire-control error, no matter what the dispersion is. The question is frequently asked, should dispersion be increased or decreased to increase the hitting?
Even supposing that all dispersion could be eliminated, and that all the shots of a salvo fell in a point, it is evident that range errors and lateral errors of considerable magnitude would still exist. We should then be confronted with the fact that having eliminated all dispersion the hitting could actually be increased by introducing dispersion. We may accept as a settled fact, then, that in gunnery a certain amount of dispersion is necessary to produce the greatest number of hits.
There is no doubt that spotting is facilitated if the dispersion is small, and the fire-control error should thereby be reduced: This is a very important point. Were it not for this consideration, it would not be necessary to dwell on dispersion until the practical limits of the fire-control error had been established. Then it could be ascertained whether the existing dispersion should be increased or decreased.
An examination of firings of 8-inch and 12-inch guns of the Atlantic fleet, in the practices of the fall of 1911, and the spring of 1912, shows that, maintaining the same fire-control errors, the hitting would have been increased by having more dispersion. At a range of 10,000 yards the vertical dispersion should have been increased 15 to 30 per cent, and the lateral dispersion as much in one case as 40 per cent to ensure the maximum hitting with the existing fire-control errors.
It is important to understand this clearly. If the fire-control error had remained unchanged, the dispersion could have been nearly trebled in some cases without decreasing the hits, and an increase of dispersion of 15 to 30 per cent would have actually increased the hitting.
On the other hand, no matter whether the dispersion were large or small, every decrease in the fire-control error would have resulted in a direct increase of hits.
There is no doubt that decreased dispersion will assist the spotter to reduce the fire-control error, but as considerable dispersion will always be present, the probable reduction in the fire-control error due to decreased dispersion cannot be stated. Furthermore, it is important to keep in mind that the spotter is one element only in fire control, and the excellent work of a spotter may be vitiated by the other factors of the fire control. In some systems, too, more dependence is placed on range finders than on spotters, and dispersion would therefore not influence the fire-control errors.
In contemplating the subject one fact stands out apparent— all energy must be devoted to reducing in every way possible the fire-control errors, for no matter what the dispersion, any decrease in range and deflection errors at once increases the hitting. The crying need is for more accurate tracking and better spotting—in short, better fire control.
In order to assist the spotter there should be the least possible dispersion.
When everything possible has been done to reduce fire-control errors, and dispersion has been kept as small as possible, target practice firings will show what range and deflection errors are obtained. These firings will also give the dispersion of salvos, and it will be apparent at once whether the dispersion is too small or too great to ensure the greatest number of hits.