In the summer of 1937 Thiel realized that the question of the length and shape of the engine nozzle needed to be reexamined too. He created a systematic test program that led directly to his fourth innovation. Although the compressed-air experiments at Heylandt in 1931–32 had shown that a long, very narrow nozzle was inefficient, von Braun’s group had not carried that fundamental research further. Practical experience and the existing engineering knowledge of nozzles seemed to confirm the accepted norm after 1932: a 10–12-degree angle of opening between the sides of the cone. Thiel was able to show in 1937–38 that an angle of 30 degrees reduced friction between the exhaust gases and the wall. Nozzles could also be shortened, because excessive length caused turbulence at the end of the expansion nozzle. The net effect was to increase exhaust velocity and shorten the length of the 25-ton-thrust engine further. By the end of 1938, Thiel’s team was close to achieving the target performance of 2,000 m/sec.8
The cumulative effect of those four innovations—improved injection, the “pre-chamber system,” a spherical combustion chamber, and a widened nozzle—was nothing short of revolutionary. It showed what could be accomplished when increased money and manpower were combined with a systematic attack on a technology ripe for transformation. Thiel’s penetrating mind, practical talent, and capacity for hard work had inspired the propulsion group and had driven it forward, notwithstanding his sometimes difficult personality. Von Braun’s leadership role was less important, since he had turned his attention to guidance-and-control problems once he saw that Thiel had engine development firmly in hand. Yet there is no doubt that von Braun’s charismatic and visionary personality, combined with his excellent management skills, further inspired Thiel and his group to push the frontiers of rocket propulsion.
The theoretical feasibility of building a powerful yet relatively compact rocket engine did not, however, mean that the practical problems were solved in 1938. Thiel’s group faced perplexing difficulties in two areas: scaling up the injection system to 25,000 kg of thrust and cooling an engine that produced considerably more heat per unit area because of its efficiency and reduced size.
Faced with increasing the injection of the propellants by a factor of seventeen, Thiel took the logical yet imaginative step of building a motor of intermediate size using a cluster of injection chambers or “pots.” Since Peenemünde’s Test Stand I would not be ready for 25-ton engine tests until April 1939, Thiel designed a motor of a size that could be accommodated in Kummersdorf: a 4.2-ton thrust one that grouped three injectors of 1,400-kg thrust each from the experiments of 1937–38. Those experiments showed that clustering raised combustion efficiency slightly by improving mixing. There was no straight line, however, from the three-pot, 4.2-ton engine to the eighteen-pot, 25-ton engine depicted in Figure 3.1. In 1939 Thiel and his associates favored the “star” configuration of six or eight larger injectors arrayed around the sides of the combustion chamber. Those configurations were actually tested in Peenemünde but must have failed because of cooling and burnthrough problems. By mid-1940 efforts concentrated on refining the eighteen-pot configuration of smaller injectors, which promised to deliver a working A-4 engine more quickly—something Army Ordnance demanded even more urgently from Thiel after World War II started in September 1939.9
The eighteen-pot concept may have been promising, but the engine failures did not cease. One difficulty had plagued liquid-fuel rocket engineering from the outset: The more efficient the engine, the more urgent the cooling problem. At 2,400° C, the temperature of the combustion chamber gases in the A-4 engine greatly exceeded the melting point of the metal shell that held them. Regenerative cooling (circulation of the fuel through the jacket around the engine) was only part of the answer. Improved alloys helped, but metallurgical research for the rocket program was limited before the war by secrecy restrictions. After the war started, an increasingly severe shortage of aluminum, magnesium, and other metals forced the Ordnance rocket program to go back to steel as much as possible, and not always steel of the best quality. That was ironic, considering the great amount of effort expended in the 1930s on using aluminum alloys to reduce weight.
The answer to the burnthrough problems proved to be another important innovation: film cooling. Diploma engineer Moritz Pöhlmann, who came to head the propulsion design office at Kummersdorf in August 1939, almost immediately suggested that a film of alcohol fuel along the wall of the combustion chamber and nozzle would provide an insulating layer against the massive heat flux from the burning gases. Tests with smaller engines immediately proved the validity of the concept. The final configuration is shown in Figure 3.1. Four rings of small holes seeped alcohol into the chamber, taking up 70 percent of the total heat flux. The fuel circulating through the cooling jacket absorbed the rest.10
It was by no means a simple task, however, to make this idea work, any more than choosing the form of the injection system ended all problems in that area. It actually required two years of repetitive trial-and-error experiments with the film cooling orifices and with the configuration of holes and nozzles in the injectors to make the eighteen-pot engine work consistently. Test stand failures often resulted in spectacular burnthroughs or explosions that ruined the engines. Metal fatigue would wear out even successful combustion chambers after limited use.11 Only large resources of material and manpower, plus extensive manufacturing and test facilities at Peenemünde and Kummersdorf, allowed the innovations of Thiel and his assistants to be quickly transformed into working technology.
In addition to having primary responsibility for developing the 25-ton engine, the propulsion group had to develop related devices in conjunction with Walter Riedel’s design bureau. Problems with the jet vanes were largely solved in mid-1938, when a new draftsman at Kummersdorf suggested that graphite replace the expensive and unsatisfactory tungsten-molybdenum alloys employed in the A-3 vanes. Thiel’s group also inherited the development of the turbopumps that would be needed to move the large volumes of propellants. Von Braun had initiated that project in mid-1935 at the southwest German firm of Klein, Schanzlin & Becker. Apparently the company had experience with large firefighting water pumps, but extreme temperatures, difficulties lubricating the liquid-oxygen side, and the need for lightness and compactness posed fundamental problems that would require years to overcome. The demands of liquid-fuel rocket engineering pushed turbopump development to the absolute limits of the technology.12
The Kummersdorf engineers also had to find a way to drive the pump’s central turbine that would provide the energy to suck liquid oxygen and alcohol from the tanks. Ordnance patented various schemes in the spring of 1936, including drawing hot gases from the combustion chamber to push the turbine blades. But the eventual solution actually derived from collaboration with Hellmuth Walter’s hydrogen peroxide rocket work. In March 1936 von Braun asked Walter to design what would soon be called the “steam generator.” When peroxide was mixed with a catalyst, it produced superheated steam that could be used to drive a turbine. The first design, based on a torpedo system, did not work well but promised a more practical solution than employing the fiery exhaust gases of the rocket engine. After 1937 the smaller turbopump/steam-generator system for the rocket aircraft projects provided further development experience (and funds from the Luftwaffe) to help solve the problem. The first turbopump-fed Ordnance rocket engines were actually the ones developed for the second version of the He 112 flown in 1939–40. By 1941 Thiel’s propulsion group, aided by Riedel’s design bureau, had the basic configuration of the A-4 turbopumps and steam generator in hand, but numerous development and manufacturing problems continued.13