The Estoppey D-series was a line of inter-war era bombsights developed by Georges Estoppey of the US Army Air Corps' McCook Field, starting with the D-1 of 1922. A key feature was the use of a pendulum to keep the bombsight correctly oriented towards the ground even as the aircraft maneuvered, and dashpots to keep it from swinging around in turbulence.
In testing in 1923, the D-1 proved more accurate than the US Navy's Mark III or the Army's Mk. I, and twice as accurate as the older Michelin bombsight designs. A series of minor updates led to the first widely used production model, 1925's D-4, of which several hundred were produced. The design was among the most advanced of its era but was still inaccurate to the point of being useless at altitudes over .
The D-7 of 1931 was similar to the D-4 but added illumination for night use, an automatic pilot direction indicator that signalled the pilot which direction to turn, and automatic bomb release to improve timing. While the D-7 was effective, the Army was planning to shift to the Sperry bombsight, which offered dramatically better accuracy. The Sperry proved difficult to produce in quantity, and in 1938 the Army ordered another production run of a much more basic update of the D-4, designated as the D-8. The D-8 was the most produced version in the D-series, with just over 10,000 built; most were delivered to the Soviet Union under the Lend-Lease program.
Estoppey entered a lengthy and acrimonious series of legal battles with the Army and Navy over royalties for his concepts. As most of the designs were produced while he worked for the Air Corps, he was not eligible for royalties. On leaving military employment in 1926, he immediately began demanding fees from both the Army and the Navy, eventually receiving several thousand dollars.
History Early bombsights Early bombsights were capable of calculating only the forward distance the bomb would move when dropped from a given altitude. To use these systems, the aircraft would first change its heading until it could see no further sideways drift of objects on the ground, indicating that the aircraft was flying in the same direction as the wind. The speed and current altitude were then entered into the bombsight, which would move a pointer fore and aft to indicate the correct spot to drop the bombs. The bomb drop would be made when the target passed through the indicator on the bombsight.
Even while approaching directly along the wind line, the speed of that wind will have an effect by speeding up or slowing down the speed compared to the ground. As the bomb moves forward by a fixed distance based on airspeed and altitude, any difference between airspeed and ground speed will result in the bomb falling long or short by that distance. For instance, if a particular bomb requires 20 seconds to reach the ground from a particular altitude, and the aircraft is flying at , the bomb would move forward before it hit. If the plane is flying in a wind that is on its nose, its ground speed is only 130 km/h, and in this case, the bombs will travel only , missing by over 100 m.
The solution used at the time was to use a stopwatch as the aircraft approaches the target. The bomb aimer would pick a suitable object on the ground along the line of approach, perhaps the target itself, and then time it as it passed through two markers on the sight set at a fixed angle. The ground speed could then be calculated by looking up the speed on a table of times and altitudes, although some systems offered some level of automation for this task. Taking this measurement was not a trivial process, especially as almost all of these systems had to be operated by the pilot while looking downward through the sights and trying to keep the plane level at the same time. If the plane's angle changed even a few degrees during the process, the measurements would become meaningless.
D-1 Georges Louis Estoppey immigrated to the United States in 1916 and soon began work on a new bombsight that would address these issues. Estoppey came up with a system that would keep the sight correctly perpendicular to the ground using a pendulum, and then stabilized its motion using dashpots so it did not swing about after moving. He filed for three patents on the concept during World War I, and formed the Musa-Estoppey Company in New York to build them. He received the patents in March 1919, but the company failed in 1921.
Estoppey then went to work for the Air Service Engineering Division at McCook Field, where he continued refining the design and introduced the first production-quality version as the D-1 in 1922. In tests at the Aberdeen Proving Ground the Army produced hit rates of 80% under standard conditions, concluding that it offered "much greater accuracy than with the old Navy Mark III or Army Mark IA sights" and that it was "twice as accurate as the Michelin". The Navy was also impressed, but only to the point of using a small number while they waited for newer developments of their own being carried out by Carl Norden to stabilize the Mark III. This led to an order for 16 D-1's, with two of those being sent to the Navy. Tests against the USS Virginia and USS New Jersey in 1923 led to a further order of 100, with 14 of those being sent to the Navy. Eventually, total production amounted to 114 from Pioneer Instrument Company for the Army and another 14 for the Navy from Eberhart Steel.
In use, the D-1 differed little from earlier bombsight designs like the Michelin. The bombardier would first watch the target move in relation to a vertical metal wire, and direct the pilot left or right until any sideways drift was seen to stop. At that point, he would watch the motion of the target along the wire until it crossed the line defined by two horizontal wires, the "top wire" and "timing wire". He would then begin turning a crank with his left hand which caused the timing wire to begin moving rearward, attempting to move it so the wire remained on top of the target as the aircraft moved forwards. The crank also moved the rear sighting wire moved forward at the same speed. When the watch indicated 15 seconds, he stopped cranking, thereby providing a measurement of the ground speed. This measurement left the upper wire positioned such that the ground speed had been accounted for. They then continued watching until the top wire aligned with the lower "range wire", and dropped at that instant.
Basic updates, D-4 Timing the drop using the stopwatch proved to be a practical problem, and Estoppey began work on an automatic timer for the D-2. The Engineering Division purchased three prototypes from Pioneer in 1925. The physical layout of the D-2 was significantly different than the D-1, with the sights in the center of a large U-shaped calculator. The two D-3 examples replaced the pendulum system of the D-2 with a gyroscopic platform from Sperry Gyroscope. Neither design proved to offer enough of an advance on the D-1 to warrant full-scale production.
Meanwhile, in service, the D-1 proved to be too fragile for the rough handling of the crew, who were found to often use the bombsight as a handle while getting in and out of the aircraft. This led to the D-4, which was more robust and also included the automatic timing system. Because it was so similar to the earlier models, three prototypes were ordered and rapidly delivered, but the Army concluded they too represented too small an improvement to consider production, especially as they were now awaiting much more advanced designs. Those advanced designs failed to emerge, and over the next five years, the Army ordered several batches of the D-4, amounting to 230 examples, with the Navy adding another 40 for low-altitude work. The Army put it into operational use in 1926.
The timing system on the D-4 greatly simplified operation. Instead of timing the motion for a fixed time, and then using that to calculate the needed setup, the system performed this automatically. The bomb aimer simply rotated a crank as the timer wound down, and when it ran out the crank was automatically disconnected with the sights set to the right angle.
New concepts In addition to the minor updates to the original concept, Estoppey also worked on entirely new designs. The first and most developed of these was the D-5, which used the new "synchronous" or "tachometric" concept to time the drop. In contrast to the previous designs where the speed over the ground was measured directly with the stopwatch, in these designs the speed was measured by a ball-and-disk integrator system. The bombardier would first enter an estimate of the wind speed and direction based on previous measurements, generally by the navigator. This would cause the sights, a low-power telescope, to begin moving so that it would track an object on the ground. The bombardier would then adjust the initial estimate by turning dials to change the direction or speed until objects could be seen unmoving in the sight. At this point, the system held a highly accurate measurement of the ground speed and course. While this offered unparalleled accuracy in theory, in practice the D-5 proved less accurate than the D-4, and the Air Corps recommended that "no further experimental work be done on the D-5 sight."
The next known version is the D-7, which used the same general mechanism as the D-4, but added several updates. Among these was an illuminated crosshair sighting system that was suitable for use at night, a system that automatically released the bombs at the right time to eliminate the delays in the bombardier pressing the release button, and a pilot direction indicator that was operated simply by turning the sight to point at the target. Although these were worthy updates, the expected imminent arrival of the more advanced designs from Sperry meant there was no reason to buy the D-7 in the meantime.
D-8 The final development of the line was the D-8, ordered due to ongoing production delays with the Sperry S-1 and Norden Mk. XV. The D-8 was a series of minor upgrades to the D-4 designed by its original manufacturer, Gaertner Scientific of Chicago. These allowed it to operate at higher speeds and altitudes of the new aircraft being introduced. After Gaertner supplied 80 hand-assembled units, the Army sent a production contract to National Cash Register (NCR) for 15,000 units immediately after the Attack on Pearl Harbor, at $200 a unit. In testing against the Norden, the D-8 managed 3 hits for 50 bombs dropped, compared to 48 out of 50 for the Norden. Aware of its limits, the Army nevertheless sent it to "airplanes for which no other sight is available."
The Army continued to press for faster deliveries of the Norden, but the Navy vacillated on allowing production to begin at other plants, fearing doing so would upset production at the main Norden plant and delay their own deliveries. At a meeting on 11 November 1942, the Army complained that while they were taking every Norden they could and immediately placing them in an airplane, the Navy had more sights than planes and were simply stockpiling their Nordens in warehouses. Meanwhile, the Navy would promise to deliver a certain quantity of sights for a given month, so the Army would adjust their aircraft orders to match, only to receive a different number and either have missed the opportunity to produce more than a month, or have to deliver them with what one Air Corps member called "this damn D-8 sight that I would just as soon have a couple of nails and a wire."