The construction of this world-class aerodynamics institute at Peenemünde, with a staff of sixty in mid-1939 and two hundred in 1943, was another large stride on the road to massive in-house research and development capability. But the delay in putting the tunnels into operation meant that the critical aerodynamic innovations for the A-4 were largely obtained through educated guesses and improvised experiments. When the decision was made in January 1938 to build a new vehicle (the A-5) to test guidance systems, the question of the external form for a rocket again became pressing. Two problems in particular forced Hermann and his staff to improvise: fin design and the stability of rocket bodies as they passed through the sound barrier.
In designing the A-5, the body form was not the problem. Since the A-3 fuselage was considered adequate, and because extensive wind-tunnel testing was still unavailable, the body was made only slightly fatter on the A–5, which then became the model for the A-4. By contrast, no one knew how to design fins for supersonic flight. In 1937 Hermann had lured to Peenemünde as his chief assistant and head of research Dr. Hermann Kurzweg, an acquaintance from the University of Leipzig then working at the eminent optical company Zeiss. Kurzweg, an inactive but contributing member of the SS since 1934, tackled the A-5 fin problem in early 1938 with only the most limited information: a knowledge of supersonic aerodynamics, a rough estimate of the expected pressure distribution over the rocket body, and wind tunnel information on the characteristics of flat plates. (Another consideration must have been Dornberger’s injunction that the fins, when scaled to A-4 dimensions, be able to pass through a standard railroad tunnel.) To allow for the expansion of the exhaust jet at high altitude, where air pressure is low, Kurzweg made the rear of the A-5’s fins open at a much wider angle than those of the A-3. For good supersonic characteristics he swept the forward edge more strongly, kept the fins relatively thin, although thicker than on the A-3, and not quite as wide as they were long. The result was the first critical aerodynamic innovation for the A-4: a broad fin shape familiar from later pictures of that missile.29
Lacking any way to get quick answers about the appropriateness of his design, Kurzweg resorted to homemade improvisations. One weekend he carved a rocket body out of a Peenemünde pine branch, inserted weights into holes to get the proper balance, and made hard rubber fins in the proposed shape, but in three different sizes. To obtain low-speed stability information he tried throwing the model off the roof of his house. When that proved unsatisfactory, the model was mounted on a wire through its center of gravity, and Kurzweg drove down the Berlin-Anklam highway at a speed of 100 km/h (about 60 mph). The largest and second-largest versions of the fins seemed to be stable, but not the smallest. Since it now seemed that the aerodynamics group had a workable A-5 concept in hand, wind tunnel testing was done in 1938 and 1939 at Aachen and in the subsonic installation at the Zeppelin Airship Construction Company in Friedrichshafen on Lake Constance. Kurzweg’s fin design came to be embodied in the first unguided A-5s launched from the Greifswalder Oie in October 1938. With small modifications, this shape was also used on the A-4.30
But before the aerodynamics group could make a definite decision, a more systematic attack on fin form was necessary. In conjunction with the 1938 A-5 launches, which aimed to exercise the launch organization and demonstrate the aerodynamic stability of the rocket, Peenemünde-East sent aloft a number of models equipped with various fin shapes and propelled by small Walter hydrogen-peroxide motors. The exact purpose of those model experiments is unclear, but the most likely one was testing for an unguided anti-aircraft missile proposed by Wa Prüf 11. Inspired by that effort, the aerodynamics group then fired off forty subscale A-5 models using Walter motors in March 1939. Eight different fin shapes were given to the 1.6-meter-(5-foot)-high models, and the launches were systematically photographed. The results confirmed that, of the designs tested, Kurzweg’s original had the best subsonic stability characteristics. Later, extensive wind tunnel work at Peenemünde and Zeppelin substantiated this research for the whole velocity range and refined the shape for the A-4. It is a tribute to Kurzweg’s ability that he had so successfully defined the solution from the outset.31
A second critical innovation was finding a method to determine whether the A-5 and A-4 would be stable as they passed through the speed of sound. No wind tunnel before the late 1940s could successfully produce velocities between Mach 0.85 and 1.2, and theory regarding this “transonic” region was in an equally primitive state. An unreliable Mach 0.95 test run (at Aachen?) suggested to Hermann that the A-5’s center of pressure would move far in front of its center of gravity, making the rocket unstable—it would tumble out of control. To gain some data on the problem, the head of the aerodynamics group proposed in July 1938 that an iron model could be dropped from an airplane at an altitude of 7,000 meters (about 23,000 feet). It would pass through the speed of sound at 1,000 meters while being carefully photographed from the ground and observed from the air.32
Those tests were started no later than the autumn of 1939, using a Luftwaffe He 111 bomber from Peenemünde-West. A new recruit to the guidance group, Dr. Walter Haeussermann, witnessed one of the drops at the end of 1939 from a second plane piloted by Wernher von Braun. After the release from the He 111, von Braun dove his airplane after the model to observe its behavior and to radio to recovery crews where it hit the water. In the end, the drop tests laid aside concerns about serious transonic instability problems for the A-5/A-4 design but were not exacting enough to show that no problem existed. In fact, repeated launches of the A-5 from 1939 to 1942 demonstrated that the rocket was marginally unstable in the transonic region. Its nose would wobble in a circular motion of a few degrees radius, creating enough drag to prevent it from ever passing through Mach 1. Thus, when the first A-4 launches were attempted in 1942, the aerodynamics group, along with everyone else, had to cross their fingers and hope that, in the quick transition through the sound barrier, the control system would prevent any dangerous movement of the missile.33
It is thus clear that the Peenemünde wind tunnel was not essential to the fundamental shape of the A-4, but assembling a highly competent aerodynamics staff was. Even so, the systematic work that began in the tunnel in 1940, along with the subsonic experiments of Dr. Max Schirmer at Zeppelin, was highly important for the refinement of the rocket’s design and the elimination of uncertainties in many areas. When the rocket group began working on the A-3 and the A-5, virtually nothing was known about the heating caused by friction with the air. For the A-4 the problem was especially urgent because of much higher velocities and the need to survive reentry into the earth’s atmosphere after passing the 80-kilometer (50-mile) peak of the trajectory. Peenemünde aerodynamicists undertook fundamental heat-transfer research in their tunnels using simple shapes and rocket models equipped with temperature sensors. Their data helped verify theoretical calculations and provided guidance to the designers as to the steels that would have to be used. Elaborate pressure measurements were also taken on the surface of wind tunnel models cut in half lengthwise, allowing many more sensors to be installed. That procedure helped verify the predicted loads on the vehicle—information crucial to the structural designers. Another area where Peenemünde broke new ground was in assessing the impact of engine exhaust jets on the aerodynamic characteristics of missiles. Using compressed air jets in the tails of models, the group found that drag increased in the subsonic range but decreased in the supersonic. Without that kind of information, launching the A-4 would have been very much a shot in the dark.34