The Bombsight

Giving the precision and accuracy of artillery, our famed bombsight enables the AAF to search out and hit the enemy far behind the front lines.

Only yesterday, accurate bombing was considered impossible. Ballistic data for ground weapons had been adapted to bombs and bombing theories. Mechanisms generally used gave only approximate solutions of the bombing problem. Comparatively simple mechanisms guaranteed fair accuracy at lower altitudes but not at heights above the effective range of antiaircraft guns. As bombers flew faster and higher the problem became very difficult. Only in the United States did anyone believe enough in the possibility of a solution to seek the answer. And they found it.

The airplane carrying bombs might well be compared with a gun platform, but it is an unstable and fleeting platform. It never remains in one place long enough for normal calculations of firing data. It moves rapidly in three dimensions as well as about three axes. The propelling charge of the projectile is gravity and the speed of the plane itself.

The problem was not easy. If the bombardier would mount his sight on a steady platform which moved smoothly along a level, straight line, he could use any of a number of known devices for accurately determining the point of release. But his platform is far from steady. High speed does not aid level, straight flight through normally rough air.

During a bomb's final trip to earth, three forces work on it. The actions of these forces must be calculated in a few seconds before its release and then used in determining the exact instant of release,

While the bomb is hanging on the shackle of the bomb-rack one force affecting it is the forward motion of the plane. When the bomb is released it has the same horizontal speed as the plane. Its inertia will maintain that same horizontal speed unless it is acted upon by other forces. A small boy, jumping off a moving car, senses this when he lands on his feet prepared to tumble or to run along with the car for a few seconds.

In a vacuum the bomb would continue to move at the same speed and in the same direction as the airplane at the time of release. If the bomber maintained speed and direction after "bomb away," the plane would still be directly over the bomb at the moment of impact.

The second major force, gravity, affects the bomb the instant it is released from the shackle. Its effect {if we again ignore air resistance) is easily calculated in terms of speed or of distance fallen at any instant after release.

Gravity works straight down. The bomb's original velocity works horizontally. By combining the two, we get the fundamental problem of a bombsight: how far down and how far forward will the bomb be after it has fallen a certain amount of time?

The third force, which we cannot ignore, is air resistance. We can streamline the bomb and give it fins which will keep it true {though some foreign air forces do neither very well), but air resistance is everywhere. Air acts to resist both other forces. It retards the acceleration due to gravity. Even more strongly, air lessens the horizontal velocity.

A freely falling bomb, starting from rest, will fall in a given number of seconds (t), a distance in feet (h), their relation being expressed in the formula:

h = 16.1 t2

If the time of bomb fall is 37 seconds, the distance fallen without air resistance would be 16.1 x 37 x 37, or 16.1 x 1369 = 22,040.9 feet.

Actually, the best-designed bombs fall only a shade under 20,000 feet during that time. If the fins and body are true enough to give consistency we merely measure the time of fall from all altitudes in ordnance tests. Then we give the bombardier a table from which he gets the first setting on his bombsight — the actual time of fall from his altitude.

We noted that a bomb in a vacuum would stay directly below the airplane if the plane continued to fly straight and level at the same speed. If the plane were flying 200 mph at the time of release the bomb would move horizontally about 294 feet while falling only 16 feet in the first second. In the next second, the bomb would fall about 48 feet and would go forward about 294 feet again. The bomb follows a steadily steepening path downward. Air resistance retards horizontal velocity more than it slows the effect of gravity. Though the actual time of fall from 1,000 feet is eight seconds (nearly the same as in a vacuum) the bomb loses enough of its horizontal velocity to make it travel about 67 feet short of 2,350 feet which it would have gone at a constant speed of 200 mph.

This lag is called "trail."

We measure this lag for each speed and altitude and give it to the bombardier in a table just as with the other data setting on his bombsight. We need only look at the target through the telescope and "track" it for a few seconds. The bombsight will do the rest.

But other factors must be solved either by the bombardier or by the bombsight. The most obvious are wind and target motion.

Consider first only wind, and let the target be fixed. Suppose the wind is constant — 30 mph for example — all the way to the ground, and exactly "on the nose." If the bomber's airspeed is 200 mph it will move 2,350 feet through the air in the eight seconds that the bomb requires to fall 1,000 feet. In that same air "block," moving or still, the bomb will lag 67 feet behind the airplane in that time. But the air "block" is not still.

Instead, during eight seconds at 30 mph, it moves about 350 feet in the opposite direction. Because the airplane and bomb were in the air "block," they were merely "carried along." Instead of traveling 2,350 feet over the ground, the airplane will move only 2,000 feet. The bomb, still falling 67 feet behind, moves 1,933 feet horizontally before hitting the ground. This horizontal distance moved by the bomb we call "range." It is one of the factors the bombsight must "calculate."

A 30 mph wind "on the tail" has exactly the opposite effect. The range is now 2,633 feet while the airplane moves 2,700 feet with respect to the ground. Data therefore is based upon true air speed and time of fall, while the bombsight must work on ground speed. Once we have set the trail and time of fall the bombsight must calculate ground speed and interpret it in terms of range.

That is the whole basic operation of a bombsight, which is fundamentally a ground speed meter.

Still more variations must be faced, however.

Before we consider more obstreperous winds — the kind that blow on almost every mission — let's bomb a target which moves in a straight line at constant speed of 30 mph in the direction opposite to that of the airplane.

During eight seconds the airplane will move 2,350 feet through the air which, we will now say, is not moving. But now the target is moving, with respect to the air, a distance of 350 feet in the same eight seconds. By combining the two distances, we see that the airplane "moves" 2,700 feet with respect to the target. The relative speed of the airplane and target is 230 mph, the same as when the target was fixed and when the wind was blowing on the tail at 30 mph. The bomb still lags behind the airplane 67 feet. Therefore, if we want the bomb and target to "connect," we must release the bomb on the same range, 2,633 feet, as for the tailwind conditions. Straight line target motion, we can thus see, has exactly the same effect as wind of the same velocity, but in the opposite direction.

Now notice what happens over the fixed target.

Consider a wind which is blowing at 30 mph directly across the ground track of the airplane. (See Figures 1 and 3). Obviously, in order to make good that track, the airplane must "crab." It does this by heading along the air path, PC. Actually, it makes good that path with respect to the air and the bomb still lags behind the airplane the same amount as in still air to the point S, for example. But during that time the entire block of air has moved to the right over the ground, carrying both the airplane and the bomb, so that (on Figure 3) the airplane ends up at Q and the bomb at T, still the same trail distance behind the airplane.

Here we seem to have hit a stumbling block: The bomb does not fall on the ground track of the airplane. The airplane was headed across its own ground track and the bomb lags behind on a line extending the axis of the plane. At no time after release is the bomb over the ground track of the airplane. We must find some means for determining the offset necessary.

In earlier American sights, and today in most foreign bomb-sights this offset, which we call "cross-trail," was handled by a special setting. Yet offset is merely a function of trail (which we have already set) and drift angle. Therefore in our latest sight we have used mechanical linkage between the trail mechanism and direction control. The cross-trail correction is entirely automatic.

Target motion again has the same effect as a wind of the same velocity in the opposite direction. All of the factors are the same and they will be unchanged in principle regardless of the direction of the wind or target motion. A combination of wind and target motion does not complicate the basic picture or require any new settings. After the settings for trail and time of fall, the bombsight is concerned only with the resultant "closing" speed and direction of the airplane with respect to the target. It need not be concerned with the causes of the relative motion.

The wind, unfortunately, never is constant "all the way down." Any change in wind direction or velocity during the bomb fall will tend to spoil the moving air "block" theory we have pictured. As long as the bomb is moving in air which was "carrying" the airplane at the time of release, the only "wind" acting on the bomb is that due to its own motion. However, if a lower layer of air is moving slower or faster, or in a different direction, it will act upon the bomb in an amount proportional to the change in direction or speed and to the length of time during which the bomb remains in the new strata.

This has long been known as "ballistic effect." Changing winds along the trajectory are termed "ballistic winds" and their effect is calculable if the amounts and positions of change are known.

Yes, we can calculate and allow for the ballistic effect if we know the exact wind changes below. Over enemy territory, of course, we can seldom count on having exact data but, fortunately, changing winds have slight effect on normal operations. Extreme changes in winds between low and high altitudes are sufficiently predictable at least to warn us, while under normal conditions no great wind change exists immediately below the airplane. Weather changes involve hours and days. A bomb falls in seconds.

Thus far we have analyzed the bomb trajectory to determine the basic problem but we have not touched one factor which has defeated many otherwise excellent bombsight designs.

An airplane is free to move forward, side to side, up and down. It also rotates about three axes — one vertical, one through the wing, one through the fuselage. To some extent, a plane moves and rotates regardless of its so-called inherent stability. or of the pilot's skill. Because this is true, any reference line — such as a line of sight to the target — which is attached rigidly to any part of the plane will bounce around with every motion of the aircraft. Slightly better is a sight reference not attached to the airplane but controlled by a pendulum. Yet this type also is thrown about by any acceleration.

Changes in direction of such reference lines are normally as much as three or four degrees. Oscillations up to as much as 10 or 15° must be expected. These oscillations would make high altitude bombing completely futile if we desire to hit anything smaller than a whole city.

One degree equals approximately 17 mils, and one mil covers a distance of 25 feet on the ground when sighting from 25,000 feet. An oscillation of only three degrees while bombing from 25,000 feet would cause an error of 1,275 feet. Similarly, an error of three degrees in direction, maintained for 40 seconds in a 250 mph bombing run, would produce an error of 750 feet in location of the bomb release point. On the ground 25,000 feet below, this direction error would cause the bomb to strike over 1,500 feet to one side of the proper point.

Bombsight inventors have tried two general schemes to stabilize their line of sight. The free pendulum idea was a slight help and the installation of dashpots (to dampen oscillations) further improved results.

The gyroscope naturally appealed to inventors' imaginations. Its chief deficiency is its tendency to change direction if an unbalanced force is applied to it. As long as a "perfectly" balanced rotor untouched by outside forces rotates at high speed it will maintain its axis for all practical use in a fixed direction in space.

How could air engineers apply the force needed to keep the rotor turning? Electric motors, with the armature being a part of the rotor, seemed to be the answer but early airplanes did not have entirely adequate electrical systems. It was easier to build and balance solid air-driven rotors, though not as easy to apply the air without causing torque. Such gyroscopes were the first used in bombsights, sometimes combined with the pendulum-and-dashpot feature. Now, however, precision has dictated the use of electrical rotors.

Next we had to control the stabilized line of sight without applying a force to the gyro. Obviously the line of sight must be controlled and this cannot be done without applying a force to the stabilized system. Yet such force exerted by normal means will ruin stability. It must suffice here to say that this problem has been completely solved in some sights.

Other engineers have believed the objective so impossible that even today they have not solved it. This is why certain bombsights are almost unbelievably accurate, while others, having only the partial solution, are satisfactory for little better than "scatter" bombing of areas.

Air pioneers solved a further problem. Stabilization must be provided against oscillations around three axes, yet no one gyroscope can be used for stabilization with reference to more than two perpendicular axes. In the old, slow bombers, early bombsights that tried gyro-stabilization used only one gyro and that one had its spin axis vertical. But modern, high-speed bombing necessitates directional stabilization. Today the best sights add a second gyroscope with horizontal axis.

Except for the stabilization problem a bombsight is just another fire control instrument.

No one marvels very much about a cannon shooting accurately a distance of 15 or 20 miles, yet its shell travels under the same free flight condition as the bomb after release. In fact, a bomb released from an airplane traveling 200 mph moves along the same trajectory as it would if fired horizontally at that speed from a gun mounted on the top of a mountain. Its inertia moves it forward, gravity pulls it down, and air resistance retards both actions.

An instrument used to determine the correct release point must first be given the length of time required for the bomb to fall to the target. It then determines the horizontal speed with respect to the target and multiplies that speed by the time of fall. Result is the distance which the bomb would travel horizontally (with no retardation) during that interval. After mechanically subtracting from that total distance the trail (lag) which also was set in from a chart, the sight produces a range angle (Figure 2). This is the angle formed between the line of sight to the target and a vertical line from the airplane at the proper instant of release.

That, in essence, is the operation of a bombsight. It is merely an automatic speed and distance calculator which interprets its findings in the form of an angle.

Various forms of sights have been developed but they must all perform those functions unless, as in the cruder sighting mechanisms, the bombardier is expected to do some of the calculations. Disregarding these crude types, some of which are entirely satisfactory for minimum altitude (below 1,000 feet) bombing, two general types of mechanisms are used in bomb-sights for making the necessary calculations. These two types are known as "timing" and "synchronous." Of the two, the timing has proved less satisfactory.

The essential simplicity of a bombsight rests upon one theorem of plane geometry: In two similar triangles the ratio of the length of a side of one triangle to the length of the corresponding side of the other triangle is equal to the ratio of another side of the first triangle to the corresponding other side of the second triangle.

Consider, for example, the right triangle in Figure 2 formed by the vertical line PC from the airplane to the ground at the time of release, the line of sight PT, and the horizontal range CT. Now, still using the sighting angle at P as the apex, consider another very tiny right triangle formed at P by a horizontal wire (or bar) intersecting the vertical line and the line of sight. The ratio of the total altitude PC to the vertical distance between P and the wire equals the ratio of the range distance CT to the length of wire intercepted by the other two lines. That, in one form or another, is in every computing bombsight, although some are more accurate than others.

In the so-called timing type of bombsight, an object on the ground (generally the target itself along before release time) is "tracked" by moving a cross-wire back along the wire for a number of seconds corresponding to the time of fall. The cross-wire is kept in line between P and the object and, at the same time, a second cross-wire is moving forward from the vertical under P. The speed of the first wire corresponds to the ground speed. The distance which it moves along the horizontal sighting wire corresponds to the distance on the ground over which the airplane travels during the time of tracking.

If the rear wire travels forward at the same speed as the front wire moves to the rear, and if the tracking time is equal to the time of fall, it is apparent that the rear wire will move ahead of the vertical a distance corresponding to the actual distance CB which the airplane will travel during the time of fall.

But that is not the correct range because trail has not been subtracted. To accomplish this, we make an adjustment. The rear wire is moved to the rear of the vertical a small amount corresponding to the trail. It starts forward with a handicap which causes it to go beyond the vertical a distance corresponding to the actual range. Now the line from P through the rear cross-wire is the line of sight which should pass through the target at the instant of release. We direct the airplane along the proper course as the target and the rear cross-wire approach each other. Just as they come in line we drop the bomb.

For technical reasons the timing method is not entirely satisfactory. It has been described in detail because the basic ideas of similar triangles and of tracking are the same in both methods. The synchronous method, though better in practice, would merely confuse the basic idea.

One basic difference is the fact that the line of sight in the synchronous sight is always power-operated by a variable speed drive. This speed is varied until the line of sight (generally a telescope) moves automatically to keep lined on the target. The mechanism sets up within itself a release-line-of-sight corresponding to that generated by the rear crosswire in the timing sight. An electrical contact is positioned by this operation so that, when the telescope reaches the generated angle, the bombs are released automatically.

If he wishes, the bombardier can release his bombs manually even while using the synchronous bombsight. Manual release nullifies a considerable part of the advantage of the precision bombsight. There is a significant time interval consumed by seeing, registering the picture in the brain and then reacting appropriately, If this interval is only one-fifth of a second while bombing at 200 mph, the delay will cause the bomb to go 60 feet beyond the point which it would have hit on automatic release. A bombardier's physical and nervous fatigue on a mission may delay as much as one second — an error measured in hundreds of feet.

These advantages of manual operation apply equally to the pilot's job on the bombing run. For the 20 to 40 seconds during which the bombardier is tracking his target, air speed and altitude must be held exactly constant. Responses to direction indications from the bombsight must be precise and almost instantaneous. In fact, flying the bombing run properly is the most exacting kind of instrument flying. It is not improved by the long trip to the target nor by the distractions of enemy fighters and anti-aircraft fire.

Probably the most important single element in precise bombing is accurate flying. Yet even alert pilots have difficulty holding a course within 2° or 3° while at the same time maintaining exactly constant air speed and altitude.

To exploit the capabilities of the bombsight to the full it is essential that flying on the bombing run be as automatic and precise as is the operation of the bombsight. All of our heavy bombers have been equipped with new types of automatic pilots connected directly to the bombsight. The bombardier needs only to track his target by holding the telescope on it.

Instead of turn indications being given to the pilot by a needle on his instrument board, the airplane controls respond instantly. When no turn is needed, the airplane is held rigidly on the reading indicated by the bombsight directional gyroscope. During the approach to the bombing run the same mechanism can be used to make the airplane take violent evasive action to confuse ground gunners and enemy fighters.

These innovations are common to precision bombsights. They make it possible for our Air Forces to put into action the only concept of air power which is not utterly destructive and inhuman. Key points, not areas, in enemy war production are all that need be destroyed to cause the enemy to collapse gradually. These points can be most economically destroyed if they are singled out and attacked specifically.

Reducing bombing errors by 50 per cent has the result of multiplying the actual effectiveness of a given number of airplanes by over four times. It enables the job to be done with less than one-fourth of the formerly required number of bombs. Such a real saving in gasoline, planes and lives demands that our unrivaled bombsights be exploited to the limit.

This article was originally published in the October, 1943, "Special Issue US Air Forces At War" issue of Flying magazine, vol 33, no 4, pp 103-107, 342-344.
The PDF of this article includes a photo of the bombsight installation in a B-25, the three diagrams above illustrating the various factors involved in bombsight operation, a series of photos of the sinking of the Trieste by B-17s, and a time-of-fall graph.
Photos credited to Army Air Forces.