Eliminated as well were the small accelerometer wagons that measured horizontal movement away from the trajectory. The A-5 guidance system, which Kreiselgeräte soon called Sg 52, would concentrate only on maintaining the proper attitude of the rocket. To get more sensitive control and perhaps even stability, more inputs were needed to keep the rocket from yawing, pitching, and rolling and to return it to its desired position. In contrast to the A-3s, the gyros on the stabilized platform of the Sg 52 would feed in signals giving the attitude of the vehicle with respect to the platform. It would be necessary also to have a feedback of the positions of the jet vanes that deflected the engine’s thrust. Those signals would then be mixed with the ones from the three rate gyros, which were not on the platform and which sensed the rate (velocity) of the rocket’s turning. The mixed inputs would then be transmitted to the vanes via a mechanical system of rotating rods and gears.40
Most of the decisions about the new Kreiselgeräte system were made in short order, by the middle of January 1938. It took Dornberger and von Braun much longer to initiate a competing A-5 system with Siemens. The most relevant experience of Altvater’s group was with autopilots that kept airplanes traveling on specific headings at set altitudes. In such devices, a “course gyro” maintained the heading. In some versions, one or two other gyros, in conjunction with rate gyros for stability, also kept the aircraft from rolling around its longitudinal axis or pitching its nose up and down. Those autopilots had no stabilized platform. It was thus natural for Siemens to try also to use position gyros fixed to the body of the rocket, although a system of that type is inherently less accurate than a platform, because a movement of the vehicle in one axis shifts the orientation of the other axes. Only the short period of active guidance during engine firing—less than 45 seconds for the A-5—and a control system that effectively limited missile motion would make the errors tolerable. A rocket-fixed system would, however, have the virtue of greater simplicity.
Despite the similarities between such a system and autopilots, Altvater’s people had to start nearly from scratch in considering how to make it work on a rocket—and not just for the A-5, but above all for the A-4. Unlike an autopilot, a ballistic missile guidance system would have to work over a range of velocities from zero to about 5,400 kilometers per hour (3,350 miles per hour) and altitudes from sea level to 30 kilometers (100,000 feet), where the A-4 would burn out. Thus the external aerodynamic forces acting on the vehicle would be constantly changing, as would the rocket’s center of gravity and moments of inertia because of emptying fuel tanks; yet in 1938 very little was known about any of the crucial quantities that would be needed to design an A-4 system. The high vibration and acceleration of launch also imposed a much more demanding environment on the equipment.41
In response to those difficult requirements Karl Fieber, one of Altvater’s key subordinates, proposed a system using only two position gyros of a new specialized type. His idea would become the basic principle of the A-4 operational guidance system and thus was a crucial innovation for the missile. One gyro, the “Horizont,” controlled the pitch axis; its name derived from aircraft artificial horizon gyros. The other, which Fieber called the “Vertikant,” measured movement in both the roll and the yaw axes (like the Horizont, it was a two-degree-of-freedom gyro, that is, free to move in two axes). Siemens eventually came to apply the term “Vertikant guidance” to the whole system. According to Fieber, he filed a patent application for the idea at the end of 1938. When the patent was awarded, it was so secret that he was not even allowed to retain a copy.42
In Fieber’s system, the Horizont had a further purpose besides measuring the attitude of the vehicle in pitch. It also sent the signals that tilted the rocket over from the vertical to an angle of about 45 degrees. The guided phase of a simple ballistic missile trajectory has the same function as an artillery barrel. The projectile reaches a certain velocity at a certain angle of firing at the end of the barrel or at the burnout of the engine, after which it coasts under the influence of gravity to the target (see Figure 3.3). Velocity and angle of fire are the primary determinants of range for a given projectile. As Becker and the Ordnance rocket enthusiasts had recognized from the outset, the price for getting rid of the gun barrel’s limitations was the use of some complicated mechanism to stabilize and guide the rocket during burning. This mechanism must include some means to pitch the missile over as well, since a liquid-fuel rocket cannot easily be fired off an angled rail like a battlefield solid rocket. A missile using a liquid-fuel engine must take off vertically because its initial acceleration is low; if it were launched at an angle, the aerodynamic forces would be too weak to stabilize it against the disturbing forces of gravity and wind.
When Boykow was brought into the program, he outlined his scheme for carrying out the pitchover. After a period of vertical ascent, the missile would turn quickly to a 45-degree angle and stay rigidly on another straight trajectory. Immediately following the A-3 failures, the rocket group and Kreiselgeräte realized that the forces required from a control system by this scheme were highly unrealistic since the Sg 33 was too weak even hold the rocket on a vertical trajectory. During the ensuing discussions in early 1938 between Peenemünde and its two guidance contractors, von Braun and his assistants came to accept the need for a gradual pitching of the vehicle, at least for the A-5.43
The question was how to command such a maneuver. Fieber’s patent is unavailable, so it is unclear how he proposed to do it, but a solution was soon found that could be employed on any rocket-fixed system. By using a clockwork mechanism to rotate slowly the pickoff (sensing mechanism) of the pitch gyro, it would essentially be fooled into thinking that the missile was pitching in the opposite direction to the desired trajectory. The correction sent by the gyro to the control system would push the vehicle’s nose over. The stabilized platform of Kreiselgeräte’s Sg 52 and successor systems required a more complicated mechanism that rotated the platform in pitch to get the same effect.44
Fieber’s Vertikant guidance, particularly when it included this simple mechanism for carrying out the pitch program, was thus an imaginative and viable alternative to Kreiselgeräte’s platforms. Actually developing the system at Siemens proved impossible, however. One company manager—perhaps Altvater himself—called the project “charlatanry.” He was skeptical of the whole enterprise because of the rocket’s rugged environment, the specialized character of Fieber’s gyros, and Ordnance’s desire for an A-5 system in less than a year from early 1938, which reflected the Army’s belief in the urgency of the ballistic missile program. Additional opposition arose in Siemens because the aviation instruments division was strained to capacity producing for the Luftwaffe. Faced with resistance to the idea of producing even two dozen systems for the Army, Wernher von Braun seized on an “almost absurd” idea thrown out in desperation by Fieber. Von Braun asserted in a meeting that three ordinary aircraft “course gyros” could be used in the A-5 to indicate the position of the rocket, even though they lacked accuracy and were not normally exposed to such high accelerations. Afterward Fieber was furious, but von Braun slapped him on the shoulder and replied, “But, Dr. Fieber, boldly asserted is half proven!” It was one of his typically audacious moves to get around obstacles in his way.45