Category THE DEVELOPMENT. OF PROPULSION. TECHNOLOGY. FOR U. S.. SPACE-LAUNCH. VEHICLES,. 1926-1991

Soon after the army deployed Corporal I, Aerojet had occasion to develop its storable-propellant technology further with stage two of the Vanguard launch vehicle for the navy. The firm’s Aerobee sounding rockets, building on the WAC Corporal engine technology, had led to the Aerobee-Hi sounding rocket that provided the basis for the projected stage two. As requirements for that stage became more stringent, though, Aerobee-Hi proved deficient, and Aerojet had to return to the drawing board. The firm charged with designing Vanguard, the Martin Company, contracted with Aerojet on Novem­ber 14, 1955, to develop the second-stage engine. Martin had deter­mined that the second stage needed a thrust of 7,500 pounds and a specific impulse at altitude of 278 lbf-sec/lbm to provide the required velocity to lift the estimated weight of the Vanguard satellite.20

The development of an engine to meet these specifications proved to be difficult. Martin’s calculated thrust and specific im­pulse would not meet the vehicle’s velocity requirements without severe weight limitations. Aerojet engineers selected unsymmetri­cal dimethyl hydrazine (UDMH) and inhibited white fuming ni­tric acid (IWFNA) as the propellants because they were hypergolic (eliminating problems with ignition), had a high loading density (reducing the size, hence weight, of propellant tanks), and delivered the requisite performance. Another advantage of hydrazine and acid was a comparative lack of problems with combustion instability in experimental research.21

The history of the evolution from the aniline-nitric acid propel­lants used in the WAC Corporal (specifically, red fuming nitric acid with 6.5 percent nitrogen dioxide plus aniline with the addition of 20 percent furfuryl alcohol) and in the first Aerobee sounding rocket (35 instead of 20 percent furfuryl alcohol) to the UDMH and IWFNA used in Vanguard is complicated. But it illustrates much about pro­pellant chemistry and the number of institutions contributing to it. The basic aniline-RFNA combination worked as a self-igniting pro­pellant combination. But it had numerous disadvantages. Aniline is highly toxic and rapidly absorbed via the skin. A person who came into contact with a significant amount of it was likely to die rapidly from cyanosis. Moreover, aniline has a high freezing point, so it can be used only in moderate temperatures. RFNA is highly corrosive to propellant tanks, so it has to be loaded into a missile or rocket just before firing, and when poured, it gives off dense concentra­tions of nitrogen dioxide, which is also poisonous. The acid itself burns the skin, as well. Two chemists at JPL had discovered as early as 1946 that white fuming nitric acid (WFNA) and furfuryl alcohol with aniline were just as poisonous and corrosive but did not pro­duce nitrogen dioxide.

But WFNA turned out to be inherently unstable over time. A complicated substance, it was hard for propellant chemists to ana­lyze in the early 1950s. By 1954, however, researchers at the Na – 154 val Ordnance Test Station and at JPL had thoroughly investigated Chapter 4 nitrogen tetroxide and nitric acid and come up with conclusions that were to be used in the Titan II. Meanwhile, chemists at the Naval Air Rocket Test Station, Lake Denmark, New Jersey; JPL; the NACA’s Lewis Flight Propulsion Laboratory; the air force’s Wright Air Development Center in Dayton, Ohio; and Ohio State Univer­sity, among other places, had reached a fundamental understanding of nitric acid by 1951 and published the information by 1955. In the process, the Naval Air Rocket Test Station was apparently the first to discover that small percentages of hydrofluoric acid both reduced the freezing point of RFNA/WFNA and inhibited corrosion with many metals. Thus were born inhibited RFNA and WFNA, for which the services and industrial representatives under air force sponsorship drew up military specifications in 1954. In this way, the services, the NACA, one university, and the competing indus­tries cooperated to solve a common problem.

During the same period, chemists sought either replacements for aniline or chemicals to mix with it and make it less problematic. Hydrazine seemed a promising candidate, and in 1951 the Rocket Branch of the navy’s Bureau of Aeronautics, issued contracts to

Metallectro Company and Aerojet to see if any hydrazine deriva­tives were suitable as rocket propellants. They found that UDMH rapidly self-ignited with nitric acid, leading to a military specifica­tion for UDMH in 1955.22

Despite the severe weight limitations on the second-stage en­gine, the Vanguard project engineers had decided to use a pressure – fed (rather than a pump-fed) propellant-delivery system. The pumps produced angular momentum as they rotated, and for stage two, this would be hard for the roll-control system to overcome, especially after engine cutoff. Concerns about reliability led to a decision to use heated helium gas as the pressurant in the feed system. Aerojet convinced the Martin Company and the navy to use stainless-steel instead of aluminum propellant tanks. Because steel had a better strength-to-weight ratio than aluminum, Aerojet argued that the lighter metal would, paradoxically, have had to weigh 30 pounds more than the steel to handle the pressure.

Moreover, a “unique design for the tankage" placed the sphere containing the helium pressure tank between the two propellant tanks, serving as a dividing bulkhead and saving the weight of a sepa­rate bulkhead. A solid-propellant gas generator augmented the pres­sure of the helium and added its own chemical energy to the system at a low cost in weight. Initially, Aerojet had built the combustion chamber of steel. It accumulated 600 seconds of burning without corrosion, but it was too heavy. So engineers developed a lightweight chamber made up of aluminum regenerative-cooling, spaghetti-type tubes wrapped in stainless steel. It weighed 20 pounds less than the steel version, apparently the first such chamber built of aluminum tubes for use with nitric acid and UDMH.23

During 1956 there were problems with welding the stainless – steel tanks despite Aerojet’s experience in this area. Martin recom­mended a different method of inspection and improvements in tool­ing, which resolved these problems. The California firm also had to try several types of injector before finding the right combination of features. One with 72 pairs of impinging jets did not deliver suffi­cient exhaust velocity, so Aerojet engineers added 24 nonimpinging orifices for fuel in the center portion of the injector. This raised the exhaust velocity above the specifications but suggested the empiri­cal nature of the design process, with engineers having to test one design before discovering that it would not deliver the desired per­formance. They then had to use their accumulated knowledge and insights to figure out what modification might work.24

The development of the combustion chamber and related equip­ment illustrated the same process. Despite the use of inhibited

white fuming nitric acid, the lightweight aluminum combustion chamber—which could be lifted with one hand—gradually eroded. It took engineers “weeks of experimenting" to find out that a coating of tungsten carbide substantially improved the life of the combus­tion chamber. There also were problems with the design of valves for flow control, requiring significant modifications.25

A final problem lay in testing an engine for start at altitude. At the beginning of the project, there was no vacuum chamber large enough to test the engine, but according to NRL propulsion engineer Kurt Stehling, “Several tests were [eventually?] made at Aerojet with engine starts in a vacuum chamber." In any event, to preclude prob­lems with near-vacuum pressure at altitude, the engineers sealed the chamber with a “nozzle closure" that kept pressure in the cham­ber until exhaust from ignition blew it out.26

The original Vanguard schedule as of November 1955 called for six test vehicles to be launched between September 1956 and Au­gust 1957, with the first satellite-launching vehicle to lift off in Oc­tober 1957.27 It was not until March 1958, however, that the second stage could be fired in an actual launch—that of Test Vehicle (TV) 4. TV-4 contained modifications introduced into the stage-one engine following the failure of TV-3 (when stage one exploded), but it did not yet incorporate the tungsten-carbide coating in the aluminum combustion chamber of the stage-two engine. And it was still a test 156 vehicle. On March 17, 1958, the slender Vanguard launch vehicle Chapter 4 lifted off. It performed well enough (despite a rough start) to place the small 3.4-pound Vanguard I satellite in an orbit originally esti­mated to last for 2,000 (but later revised to 240) years.28

On TV-5, launched April 28, 1958, the second stage provided less-than-normal thrust, but the first stage had performed better than normal, compensating in advance for the subpar second stage. Then an electrical problem prevented ignition of the third stage, pre­cluding orbit. On the first nontest Vanguard, Space Launch Vehicle (SLV) 1, apparent malfunction of a pressure switch also prevented orbiting a 21.5-pound satellite on May 27, 1958. Here, the second stage performed normally through cutoff of ignition. On SLV-2, June 26, 1958, the second-stage engine cut off after eight seconds, probably due to clogged filters in the inhibited white fuming nitric acid lines from corrosion of the oxidizer tank. The Vanguard team flushed the oxidizer tanks and launched SLV-3 on September 26, 1958, with a 23.3-pound satellite. Despite the flushing, second – stage performance was below normal, causing the satellite to miss orbital speed by a narrow margin. This time, the problem seemed to be a clogged fuel (rather than oxidizer) filter.

On February 17, 1959, however, all systems worked, and SLV-4 placed the 23.3-pound Vanguard II satellite in a precise orbit ex­pected to last for 200 years or more. This did not mean that Aerojet had gotten all of the kinks out of the troublesome second stage. SLV-5 on April 13, 1959, experienced a flame oscillation during sec­ond-stage ignition, apparently producing a violent yaw that caused the second and third stages with the satellite to tumble and fall into the ocean. Engineers made changes in the second-stage engine’s hy­draulic system and programmed an earlier separation of the stage, but on SLV-6 (June 22, 1959), a previously reliable regulating valve ceased to function after second-stage ignition. This caused helium pressure to mount (since it could not vent), resulting in an explo­sion that sent the vehicle into the Atlantic about 300 miles down – range. At least the problem-plagued Vanguard program ended on a happy note. On September 18, 1959, a test vehicle backup (TV-4BU) version of the launch vehicle placed a 52.25-pound X-ray and envi­ronmental satellite into orbit.29

Meanwhile, the navy’s Polaris missile had made more far-reaching contributions. Until Polaris A1 became operational in 1960, all U. S.

long-range missiles had used liquid propellants. These had obvious advantages in their performance, but their extensive plumbing and large propellant tanks made protecting them in silos difficult and costly. Such factors also made them impractical for use on ships. After the operational date of Minuteman I in 1962, the Department of Defense began phasing out liquid-propellant strategic missiles.34

Meanwhile, given the advantages that liquid propellants en­joyed in terms of performance, their head start within the defense establishment, and the disinclination of most defenders of liquids to entertain the possibility that solid propellants could satisfy the demanding requirements of the strategic mission, how did this solid-propellant breakthrough occur? The answer is complicated and technical. But fundamentally, it happened because a number of heterogeneous engineers promoted solids; a variety of partners in their development brought about significant technical innovations; and although interservice rivalries encouraged the three services to development separate missiles, interservice cooperation ironically helped them do so. Despite such cooperation and the accumulat­ing knowledge about rocket technology, however, missile designers still could not foresee all the problems that their vehicles would develop during ground and flight testing. Thus, when problems did occur, rocket engineers still had to gather information about what had caused problems and exercise their ingenuity to develop solu­tions that would cope with the unexpected.

By the time that Polaris got under way in 1956 and Minuteman in 1958, solid-propellant rocketry had already made the tremen­dous strides forward discussed previously. But there were still enor­mous technical hurdles to overcome. The problems remaining to be solved included higher performance; unstable combustion; the inadequate durability of existing nozzle materials under conditions of heat and exposure to corrosive chemicals from the exhaust of the burning propellants; a lack of materials and technology to provide light but large combustion chambers so the burning propellants had to overcome less mass during launch; and ways to terminate com­bustion of the propellant immediately after the desired velocity had been achieved (for purposes of accuracy) and to control the direction of the thrust (for steering).35

Once the navy had overcome the bureaucratic obstacles to devel­oping its own, solid-propellant missile, the Special Project Office (SPO) under Adm. William F. Raborn and Capt. Levering Smith achieved breakthroughs in a number of these technical areas. In early January 1956, the navy had sought the assistance of the Lock­heed Missile and Space Division and the Aerojet General Corpora-

tion in developing a solid-propellant ballistic missile. The initial missile the two contractors and the SPO conceived was the Jupiter S (for “solid"). It had enough thrust to carry an atomic warhead the re­quired distance, a feat it would achieve by clustering six solid rock­ets in a first stage and adding one for the second stage. The problem was that Jupiter S would be about 44 feet long and 10 feet in diam­eter. An 8,500-ton vessel could carry only 4 of them but could carry 16 of the later Polaris missiles. With Polaris not yet developed, the navy and contractors still were dissatisfied with Jupiter S and con­tinued to seek an alternative.36

One contribution to a better solution came from Atlantic Re­search Corporation (ARC). Keith Rumbel and Charles B. Hender­son, chemical engineers with degrees from MIT who were working 240 for ARC, had begun theoretical studies in 1954 of how to increase Chapter 6 solid-propellant performance. They learned that other engineers, in­cluding some from Aerojet, had calculated an increase in specific impulse from adding aluminum powder to existing ingredients. But these calculations had indicated that once aluminum exceeded 5 per­cent of propellant mass, performance would again decline. Hence, basing their calculations on contemporary theory and doing the cumbersome mathematics without the aid of computers, the other researchers abandoned aluminum as an additive except for damping combustion instability. Refusing to be deterred by theory, Rumbel and Henderson tested polyvinyl chloride with much more alumi­num in it. They found that with additional oxygen in the propellant and a flame temperature of at least 2,310 kelvin, a large percent­age of aluminum by weight yielded a specific impulse significantly higher than that of previous composite propellants.37

ARC’s polyvinyl chloride, however, did not serve as the binder for Polaris. Instead, the binder used was a polyurethane material developed by Aerojet in conjunction with a small nitropolymer pro­gram funded by the Office of Naval Research about 1947 to seek high-energy binders for solid propellants. A few Aerojet chemists synthesized a number of high-energy compounds, but the process required levels of heating that were unsafe with potentially explo­sive compounds. Then one of the chemists, Rodney Fischer, found “an obscure reference in a German patent" suggesting “that iron chelate compounds would catalyze the reaction of alcohols and iso­cyanates to make urethanes at essentially room temperature." This discovery started the development of polyurethane propellants in many places besides Aerojet.

In the meantime, in 1949 Karl Klager, then working for the Of­fice of Naval Research in Pasadena, suggested to Aerojet’s parent

firm, General Tire, that it begin work on foamed polyurethane, leading to two patents held by Klager, along with Dick Geckler and R. Parette of Aerojet. In 1950, Klager began working for Aerojet. By 1954, he headed the rocket firm’s solid-propellant development group. Once the Polaris program began in December 1956, Klager’s group decided to reduce the percentage of solid oxidizer as a compo­nent of the propellant by including oxidizing capacity in the binder, using a nitromonomer as a reagent to produce the polyurethane plus some inert polynitro compounds as softening agents. In April 1955, the Aerojet group found out about the work of Rumbel and Hender­son. Overcoming explosions due to cracks in the grain and profiting from other developments from multiple contributors, they discov­ered successful propellants for both stages of Polaris A1.

These consisted of a cast, case-bonded polyurethane composition including different percentages of ammonium perchlorate and alu­minum for stages one and two, both of them featuring a six-point, internal-burning, star configuration. With four nozzles for each stage, this propellant yielded a specific impulse of almost 230 lbf – sec/lbm for stage one and nearly 255 lbf-sec/lbm for stage two. The latter specific impulse was higher in part because of the reduced atmospheric pressure at the high altitudes where it was fired, com­pared with stage one, which was fired at sea level.38

The addition of aluminum to Aerojet’s binder essentially solved the problem of performance for Polaris. Other innovations in the areas of warhead size plus guidance and control were necessary to make Polaris possible, but taken together with those for the propel­lants, they enabled Polaris A1 to be only 28.6 feet long and 4.5 feet in diameter (as compared with Jupiter S’s 44 feet and 10 feet, respec­tively). The weight reduction was from 162,000 pounds for Jupiter S to less than 29,000 pounds for Polaris. The cases for both stages of Polaris A1 consisted of rolled and welded steel. This had been modified according to Aerojet specifications resulting from exten­sive metallurgical investigations.

Each of the four nozzles for stage one (and evidently, stage two as well) consisted of a steel shell, a single-piece throat of molybdenum, and an exit-cone liner made of “molded silica phenolic between steel and molybdenum." A zirconium oxide coating protected the steel portion. The missile’s steering came from jetavators designed by Willy Fiedler of Lockheed, a German who had worked on the V-1 program during World War II and had developed the concept for the device while employed by the U. S. Navy at Point Mugu Naval Air Missile Test Center, California. He had patented the idea and then adapted it for Polaris. Jetavators for stage one were molybdenum

FIG. 6.4

An unidentified solid-rocket motor being tested in an altitude wind tunnel at NASA’s Lewis Research Center (later Glenn Research Center) in 1959, one kind of test done for Polaris. (Photo courtesy of NASA)

rings with spherical inside surfaces that rotated into the exhaust stream of the four nozzles and deflected the flow to provide pitch, yaw, and roll control. The jetavators for stage two were similar.39

Besides requiring steering, the missile needed precise thrust ter­mination when it reached the correct point in its trajectory. This could be achieved on liquid-propellant missiles simply by stopping the flow of propellants. For solids, the task was more difficult. The Polaris team used pyrotechnics to blow out plugs in six ports in front of the second stage at the proper moment in the trajectory. This per­mitted exhaust gases to escape and halt the acceleration so that the warhead would travel on a ballistic path to the target area.40

Flight testing of Polaris revealed, among other problems, a loss of control due to electrical-wiring failure at the base of stage one. This resulted from aerodynamic heating and a backflow of hot exhaust gases. To diagnose and solve the problem, engineers in the program obtained the help of “every laboratory and expert," using data from four flights, wind tunnels, sled tests, static firings, “and a tremen­dous analytic effort by numerous laboratories." The solution placed fiberglass flame shielding supplemented by silicone rubber over the affected area to shield it from hot gases and flame.41

Another problem for Polaris to overcome was combustion in­stability. Although this phenomenon is still not fully understood,
gradually it has yielded to research in a huge number of institutions, including universities and government labs, supported by funding by the three services, the Advanced Research Projects Agency, and NASA. Levering Smith credited Edward W. Price in the Research Department at NOTS with helping to understand the phenomenon. By this time, Price had earned a B. S. in physics and math at UCLA. In February 1960, he completed a major (then-classified) paper on combustion instability, which stated, “This phenomenon results from a self-amplifying oscillatory interaction between combustion of the propellant and disturbances of the gas flow in the combustion chamber." It could cause erratic performance, even destruction of motor components. Short of this, it could produce vibrations that would interfere with the guidance/control system. To date, “only marginal success" had been achieved in understanding the phenom­enon, and “trial-and-error development continues to be necessary." But empirical methods gradually were yielding information, for example, that energy fluxes could amplify pressure disturbances, which had caused them in the first place.42 The subsequent success of Polaris showed that enough progress had been made by this time that unstable combustion would not be a major problem for the missile.

Long before Polaris A1 was operational, in April 1958 the DoD had begun efforts to expand the missile’s range from the 1,200- nautical-mile reach of the actual A1 to the 1,500 nautical miles origi­nally planned for it. The longer-range missile, called Polaris A2, was originally slated to achieve the goal through higher-performance propellants and lighter cases and nozzles in both stages. But the navy Special Projects Office decided to confine these improvements to the second stage, where they would have greater effect. (With the second stage already at a high speed and altitude when it began fir­ing, it did not have to overcome the weight of the entire missile and the full effects of the Earth’s gravity at sea level.) Also, in the sec­ond stage, risk of detonation of a high-energy propellant after igni­tion would not endanger the submarine. Hence, the SPO invited the Hercules Powder Company to propose a higher-performance second stage.43

As a result, Aerojet provided the first stage for Polaris A2, and Her­cules, the second. Aerojet’s motor was 157 inches long (compared with 126 inches for Polaris A1; the additional length could be ac­commodated by the submarines’ launch tubes because the navy had them designed with room to spare). It contained basically the same propellant used in both stages of Polaris A1 with the same grain con­figuration. Hercules’ second stage had a filament-wound case and a

cast, double-base grain that contained ammonium perchlorate, ni­trocellulose, nitroglycerin, and aluminum, among other ingredients. The grain configuration consisted of a 12-point, internal-burning star. It yielded a specific impulse of more than 260 lbf-sec/lbm under firing conditions. The motor was 84 inches long and 54 inches in diameter, featuring four swiveling nozzles with exit cones made of steel, asbestos phenolic, and Teflon plus a graphite insert.44

This second-stage motor resulted from an innovation that in­creased performance by adding ammonium perchlorate to the cast, double-base process used in Hercules’ third stage for the Vanguard launch vehicle. Hercules’ ABL developed this new kind of propel­lant, known as composite-modified double base (CMDB), by 1958, evidently with the involvement of John Kincaid and Henry Shuey, 244 developers of the earlier cast, double-base process.45

Chapter 6 Even before 1958, however, Atlantic Research Corporation had developed a laboratory process for preparing CMDB. In its manu­factured state, nitrocellulose is fibrous and unsuitable for use as an unmodified additive to other ingredients being mixed to create a propellant. Arthur Sloan and D. Mann of ARC, however, developed a process that dissolved the nitrocellulose in nitrobenzene and then separated out the nitrocellulose by mixing it with water under high shear (a process known as elutriation). The result was a series of compact, spherical particles of nitrocellulose with small diameters (about 1 to 20 microns). Such particles combined readily with liq­uid nitroplasticizers and crystalline additives in propellant mixers. The result could be cast into cartridge-loaded grains or case-bonded rocket cases and then converted to a solid with the application of moderate heat. Sloan and Mann patented the process and assigned it to ARC. Then, in 1955, Keith Rumbel and Charles Henderson at ARC began scaling the process up to larger grain sizes and devel­oping propellants. They developed two CMDB formulations begin­ning in 1956. When ARC’s pilot plant became too small to support the firm’s needs, production shifted to Indian Head, Maryland. Be­cause the plastisol process they had developed was simpler, safer, and cheaper than other processes then in existence, Henderson said that Hercules and other producers of double-base propellants even­tually adopted his firm’s basic method of production.

Engineers did not use it for upper stages of missiles and launch vehicles until quite a bit later, however, and then only after chem­ists at several different laboratories had learned to make the pro­pellant more rubberlike by extending the chains and cross-linking the molecules to increase the elasticity. Hercules’ John Allabashi at ABL began in the early 1960s to work on chain extenders and

cross-linking, with Ronald L. Simmons at Hercules’ Kenvil, New Jersey, plant continuing this work. By about the late 1960s, chem­ists had mostly abandoned use of plastisol nitrocellulose in favor of dissolving nitrocellulose and polyglycol adipate together, followed by a cross-linking agent such as isocyanate. The result was the type of highly flexible CMDB propellant used on the Trident submarine- launched ballistics missiles beginning in the late 1970s.46

Meanwhile, the rotatable nozzles on the second stage of Polaris A2, which were hydraulically operated, were similar in design to those already being used on the air force’s Minuteman I, and award of the second-stage contract to Hercules reportedly resulted from the performance of the third-stage motor Hercules was developing for Minuteman I, once again illustrating technology transfer be­tween services. (Stage one of the A2 retained the jetavators from A1.) The A2 kept the same basic shape and guidance/control system as the A1, the principal change being more reliable electronics for the guidance/control system. By the time Polaris A2 became opera­tional in June 1962, now-Vice Admiral Raborn had become deputy chief of naval operations for research and development. In February

1962, Rear Adm. I. J. “Pete" Gallantin became director of the Spe­cial Projects Office with Rear Adm. Levering Smith remaining as technical director.47

As a follow-on to Polaris A2, in September 1960, Secretary of De­fense Robert McNamara approved development of a 2,500-nautical – mile version of Polaris that became the A3. To create a missile that would travel an additional 1,000 nautical miles while being launched from the same tubes on the submarine as the A1 required new propellants and a higher mass fraction. The new requirement also resulted in a change from the “bottle shape" of the A1 and A2 to a shape resembling a bullet. In the attempt to help increase the mass fraction, Aerojet, the first-stage manufacturer, acquired the Houze Glass Corporation at Point Marion, Pennsylvania, and moved that firm’s furnaces, patents, technical data, and personnel to the Aerojet Structural Materials Division in Azusa, California, in early

1963. This acquisition gave Aerojet the capability to make filament – wound cases like those used by Hercules on stage two of Polaris A2. The new propellant Aerojet used was a nitroplasticized polyurethane containing ammonium perchlorate and aluminum, configured with an internal-burning, six-point star. This combination raised the spe­cific impulse less than 10 lbf-sec/lbm.48

Unfortunately, the flame temperature of the new propellant was so high that it destroyed the nozzles on the first stage. Aerojet had to reduce the flame temperature from a reported 6,300°F to slightly

less than 6,000°F and to make the nozzles more substantial, using silver-infiltrated tungsten throat inserts to withstand the high tem­perature and chamber pressure. As a result, the weight saving from using the filament-wound case was lost in the additional weight of the nozzle. Hence, the mass fraction for the A3’s first stage was actually slightly lower than that for the A2, meaning that the same quantity of propellant in stage one of the A3 had to lift slightly more weight than did the lower-performing propellant for stage one of the A2.49

For stage two of the A3, Hercules used a propellant containing an energetic high explosive named HMX, a smaller amount of am­monium perchlorate, nitrocellulose, nitroglycerin, and aluminum, among other ingredients. It configured this propellant into an in – 246 ternal-burning cylindrical configuration with many major and mi – Chapter 6 nor slots in the aft end of the stage, creating a cross section that resembled a Christmas-tree ornament. This propellant offered a significantly higher specific impulse than stage two of Polaris A2. Also, the new stage two used a different method of achieving thrust vector control (steering). It injected Freon into the exhaust, creating a shock pattern to deflect the stream and achieve the same results as movable nozzles at a much smaller weight penalty.

A further advantage of this system was its lack of sensitivity to the temperature of the propellant flame. The Naval Ordnance Test Station performed the early experimental work on this use of a liq­uid for thrust vector control. Aerojet, Allegany Ballistics Laboratory, and Lockheed did analytical work, determined the ideal locations for the injectors, selected the fluid to be used, and developed the injectors as well as a system for expelling excess fluid. The Polaris A3 team first successfully tested the new technology on the sec­ond stage of flight A1X-50 on September 29, 1961. This and other changes increased the mass fraction for stage two of Polaris A3 to 0.935 (from 0.871 for stage two of the A2), a major improvement that together with the increased performance of the new propellant, contributed substantially to the greater range of the new missile.50

One major area of difference between missile and launch-vehicle de­velopment lay in the requirement for special safeguards on launch vehicles that propelled humans into space. Except for Juno I and Vanguard, which were short-lived, among the first U. S. space-launch vehicles were the Redstones and Atlases used in Project Mercury and the Atlases and Titan IIs used in Project Gemini to prepare for the Apollo Moon Program. Both Projects Mercury and Gemini re­quired a process called “man-rating" (at a time before there were women serving as astronauts). This process resulted in adaptations of the Redstone, Atlas, and Titan II missiles to make them safer for the human beings carried in Mercury and Gemini capsules.

Man-rating was but one of the ways missiles had to be modi­fied for use as launch vehicles, but the practice carried over to later launch vehicles initially designed as such (rather than as missiles). For Mercury-Redstone, Wernher von Braun’s Development Opera-

A Mercury – Redstone launching Freedom 7 with Astronaut Alan Shepard onboard,

May 5, 1961, from Pad 5 at Cape Canaveral. (Photo courtesy of NASA)

tions Division of the Army Ballistic Missile Agency was respon – 50 sible for the process. Von Braun established a Mercury-Redstone Chapter 2 Project Office to aid in redesigning the Jupiter C version of the Red­stone to satisfy the requirements of the Mercury project. To direct the effort, he chose Joachim P. Kuettner, a flight engineer and test pilot who had worked for Messerschmitt during the Nazi period in Germany.2

Kuettner’s group recognized that the Redstone missile could not satisfy the mission requirements for Project Mercury. These ne­cessitated sufficient performance and reliability to launch a two – ton payload with an astronaut aboard into a flight path reaching

an apogee of 100 nautical miles (115 statute miles). The Jupiter C, with its elongated propellant tanks and a lighter structure, had the required performance but not the safety features necessary for hu­man flight. To add these, Kuettner’s group reverted from the toxic hydyne to alcohol as a fuel. Other changes included an automatic, in-flight abort system with an escape rocket and parachutes to carry the astronauts to a safe landing. To ferret out potential sources of failure, Chrysler, as prime contractor, instituted a special test pro­gram to promote greater reliability. The overall process proved suc­cessful, resulting in the two flights of Alan Shepard and Virgil (Gus) Grissom in May and July 1961.3

Thereafter, Project Mercury switched to Atlas D missiles to pro­pel the astronauts and their capsules into orbit. For this function, the missile required strengthening in its upper section to handle the greater loads the capsule created. Following an explosion on Mer­cury-Atlas (MA) 1 (July 29, 1960), whose cause investigators could not determine, engineers developed an improved structure linking the booster and capsule, resulting in a successful flight of MA-2 on February 21, 1961.4 MA-2 also tested the Atlas abort sensing and implementation system (ASIS) and “escape tower" that were key features in man-rating the Atlas. Besides these two features, there had to be numerous other modifications to convert Atlas to its Mer­cury-Atlas configuration. For example, the Mercury capsule’s sepa­ration rockets potentially could damage the thin “steel-balloon" skin on the liquid-oxygen dome of the Atlas, so General Dynamics (formerly Convair) engineers had to add a fiberglass layer covering the dome. This and other changes, plus increased quality control, caused the Mercury-Atlas launch vehicle to cost 40 percent more than the Atlas missile. After a failure on MA-3 (due to guidance/ control problems), Atlas launch vehicles placed John Glenn, Scott Carpenter, Walter Schirra, and Gordon Cooper in orbit between Feb­ruary 20, 1962, and May 15, 1963.5

For Titan II-Gemini, there were major problems with longitudi­nal oscillations in the engines, known as pogo (from their resem­blance to the gyrations of the then-popular plaything, the pogo stick). These never occurred in flight but appeared in a severe form during static testing of second-stage engines. Surges in the oxidizer feed lines were causing the problem, which Martin engineers and others solved with suppression mechanisms. There was also the is­sue of combustion instability that occurred on only 2 percent of the ground tests of second-stage engines. But for man-rating, even this was too high. Aerojet (the Titan engine contractor) solved the problem with a new injector.6

For other aspects of man-rating Titan II for Gemini, procedures developed for Project Mercury offered a strong influence, especially 52 as many NASA and Aerospace Corporation engineers who had Chapter 2 worked on Mercury also worked on Gemini. Gemini engineers also benefited from Titan II test launches. As George E. Mueller, NASA’s associate administrator for manned spaceflight from 1963 to 1969, stated in February 1964, the 28 launches of Titan II missiles to that date “provide[d] invaluable launch operations experience and ac­tual space flight test data directly applicable to the Gemini launch vehicle which would [have] be[en] unobtainable otherwise,"7 one example of the symbiotic (though not homogeneous) relationship between missiles and launch vehicles.

In addition to a malfunction detection system, features added to the Titan II missile for astronaut safety included redundant com­ponents of the electrical systems. To help compensate for all the weight from the additional components, engineers also deleted ver­nier and retro-rockets, which were not necessary for the Gemini mission.8 From March 23, 1965, to November 11, 1966, Gemini 3 through 12 all carried two astronauts on the Gemini spacecraft. These missions had their problems as well as their triumphs. But with them, the United States finally assumed the lead in the space race with its cold-war rival, the Soviet Union. Despite lots of prob­lems, Gemini had prepared the way for the Apollo Moon landings and achieved its essential objectives.9

Following its successful Gemini missions, the Titan II did not serve again as a space-launch vehicle until the mid-1980s after

it was taken out of service as a missile. Meanwhile, subsequent launch vehicles that required [hu]man-rating, notably the Saturn launch vehicles and the shuttle, included equipment for accommo­dating humans in their original designs, capitalizing on the experi­ences with Mercury-Redstone, Mercury-Atlas, and Gemini-Titan, which NASA passed on to the subsequent programs.

Despite the problems with the second stage of Vanguard, the air force used a modified version on its Thor-Able launch vehicle, showing the transfer of technology from the navy to the air service. The Able was more successful than the Vanguard prototype for two reasons. One was special cleaning and handling techniques for the propel­lant tanks that came into being after Vanguard had taken delivery of many of its tanks. Also, Thor-Able did not need to extract maximum performance from the second stage as Vanguard did, so it did not have to burn the very last dregs of propellant in the tanks. The residue that the air force did not need to burn contained more of the scale from the tanks than did the rest of the propellant. Consequently, the valves could close before most of the scale entered the fuel lines, the evident cause of many of Vanguard’s problems.30

Designated AJ10-40 (in contrast to the Vanguard second stage’s AJ10-37), the Able was a modified Vanguard stage that still used a regeneratively cooled combustion chamber with aluminum-tubular construction. Able kept the propellants (IWFNA and UDMH), tanks, helium pressurization system, and propellant valves from the Van­guard. The engine produced a thrust of about 7,500 pounds for roughly 120 seconds.31

First used in 1958, the Able upper stage remained in operation until January 1960, when Aerojet’s much more capable Able-Star replaced it. The newer stage resulted from an Advanced Research

Projects Agency directive of July 1, 1959. Aerojet could develop the Able-Star engine (AJ10-104) in a matter of months because it was derived from the Able engines and because it was simple. Directed to make the system rugged, with only those subsystems and compo­nents needed to meet requirements for restart, attitude control dur­ing coasting periods, and longer burning time than the Able could provide, Aerojet engineers sought “to achieve maximum flight ca­pability through limited redesign, overall simplification and opti­mum utilization of flight-proven components."32

Aerojet designed and built the combustion chamber to be “prac­tically identical" to the one used on the Able upper stages, so it re­mained an aluminum, regeneratively cooled, pressure-fed device. For unstated reasons that may have involved the air force’s desire to have the same propellants for Agena and Able-Star, the latter stage switched from the IWFNA used in Able to IRFNA (inhibited red fum­ing nitric acid) as the oxidizer, keeping UDMH as the fuel. Helium under pressure continued to feed the propellants to the combustion chamber, where the injector had concentric rings of orifices that mixed the hypergolic IRFNA and UDMH in an impinging-stream pattern. There were three helium containers, made of titanium, to supply the pressurizing gas. Experience had suggested no need for a nozzle closure diaphragm previously used to ensure high-altitude starts. Another change from Able was an optional nozzle extension 158 that allowed an expansion ratio of either 20:1 without it or 40:1 Chapter 4 with it. Rated thrust rose from 7,575 to 7,890 pounds with the noz­zle extended; rated specific impulse climbed proportionally.33

Although Aerojet’s design and development of Able-Star were quick, they were not problem-free. The virtually identical com­bustion chambers for the Able engines required test firings of only 115 seconds in duration. Able-Star’s had to undergo five static test firings of 300 seconds in duration because of the longer tanks and the increased burning time of the newer upper stage. With a new design for the injector manifold apparently resulting from the con­version to IRFNA as the oxidizer and coolant, during November 1959 Aerojet experienced a burnthrough of the injector plate and the cooling tubes in its vicinity. In a further piece of apparent cut – and-try engineering, Aerojet made appropriate (but unspecified) ad­justments to the designs, and two combustion chambers operated successfully for the full 300 seconds later that month.34

The Transit 1B launch by a Thor/Able-Star on April 13, 1960, marked the first programmed restart of a rocket engine in flight. For Transit 2A—launched on June 22, 1960—there was a problem with sloshing of the propellants in the stage-two tanks, which produced

roll forces. This resulted in an imperfect but usable orbit. To limit the sloshing, engineers added anti-slosh baffles to both Able-Star propellant tanks.35

In a successful launch of the Courier 1B satellite (on October 4, 1960), the anti-slosh baffles apparently had worked. The purpose of the satellite was to test the ability of a spacecraft to relieve crowded communications lines via delayed relay of information. The experi­ment was successful, with large amounts of data transmitted be­tween Puerto Rico and New Jersey. However, the delay of up to two hours before the message was repeated (after the satellite completed its orbit) was unsatisfactory for military purposes and telephone transmission, so the future lay with much higher, geosynchronous orbits. There, the satellite remained in a fixed position relative to a given location on the rotating Earth, allowing nearly instantaneous relay of messages. Overall, there were 20 Thor/Able-Star launches, of which 5 were failures, for a 75 percent success rate.36

Development of the propellant for Minuteman began at Wright – Patterson AFB and continued after Lt. Col. Edward N. Hall trans­ferred to the Western Development Division as chief for propulsion development in the liquid-propellant Atlas, Titan, and Thor pro­grams. In December 1954, he invited major manufacturers in the solid-propellant industry (Aerojet, Thiokol, Atlantic Research, Phil-

lips Petroleum Company, Grand Central, and Hercules) to discuss prospects for solids. The result apparently was the Air Force Large [Solid] Rocket Feasibility Program (AFLRP), which involved a com­petition starting in September 1955 with specific companies looking at different technologies. It appears that the propellant for Polaris benefited from Aerojet’s participation in this air force program.51

The Thor missiles did not remain in operational use very long, but even before the air force retired them in 1963, it had begun to use the Thor’s airframe and propulsive elements (including its vernier engines) as the first stages of various launch vehicles. With a series of upratings and modifications, the Thor remained in use with such upper stages as the Able, Able-Star, Agena, Burner I, Burner II, and Burner IIA until 1980. In addition, NASA quickly chose the Thor as the first stage of what became its Thor-Delta (later, just Delta) launch-vehicle family, which has had an even longer history than the air force’s Thor series. The Delta launch vehicles initially drew upon Vanguard upper stages, as did the Thor-Able used by the air force.10

Throughout its history, the Delta evolved by uprating existing components or adopting newer ones that had proven themselves. It used a low-risk strategy to improve its payload capacity through the Delta II at the end of the period covered by this history. But it did not stop there, evolving through a Delta III, first launched (unsuc­cessfully) in 1998, and a Delta IV that finally had its successful first launch on November 20, 2002. (To be sure, the Delta IV used an entirely new first stage, making it in some senses a new launch ve­hicle, but the design emphasized reliability and low cost, hallmarks of the Delta program from the beginning.) The unsuccessful first (and second) launch(es) of the Delta III and numerous delays in the launch of Delta IV because of both software and hardware problems 54 suggested, however, that the design of new launch vehicles was still Chapter 2 not something engineers had “down to a science," even in the 21st century.11

For the Able upper stage, the air force and its contractors used many features of the Vanguard second stage but added a control compartment, skirts and structural elements to mate it with the Thor, a tank venting and pressurization safety system, new electri­cal components, and a roll-control system. Used for reentry test­ing, the first Thor-Able failed because of a faulty turbopump in the Thor, but the second launch on July 9, 1958, was successful.12

Succeeding versions of Thor-Able modified both second and third stages of Vanguard. The final Thor-Able launch on April 1, 1960, placed the Tiros 1 meteorological satellite in orbit. In the 16 Thor-Able launches, all of the stages worked satisfactorily on 10 of the missions, whereas at least one stage failed or was only partly successful on 6 flights. Although this was only a 62.5 percent suc­cess rate, it was sufficiently good for this early period that the air force could refer to Thor-Able as “an extremely capable and reliable vehicle combination."13

Long before the final launch of the Thor-Able, nevertheless, the Department of Defense’s Advanced Research Projects Agency had

issued an order on July 1, 1959, calling for the development of the Able-Star upper stage, derived from the Able but possessing two and one half times its total impulse plus the capability to shut down its propulsion in space, coast, and restart. This ability would permit a more precise selection of the orbit for a satellite than was possible before. Once Able-Star became operational in January 1960, it ef­fectively replaced the Able as an upper stage.14

The Able-Star engine was a derivative of the several Able engines except that it had the added restart capability plus the capacity to provide attitude control during coasting periods and to burn longer than the earlier engines. Following a rapid but not unproblematic development, an Able-Star upper stage on a Thor booster launched the Transit 1B navigation satellite on April 13, 1960, marking the first programmed restart of a rocket engine in flight. Although the coast attitude-control system worked, a malfunction in the Able – Star ground guidance system resulted in a still-useful elliptical rather than a circular orbit. Sloshing in stage-two propellant tanks for Transit 2A—launched on June 22, 1960—again produced an el­liptical orbit because it caused roll forces that the guidance/control system could not overcome. Again the orbit was useful. Following placement of anti-slosh baffles in both Able-Star propellant tanks, the Thor booster failed on the attempted launch of Transit 3A, No­vember 30, 1960. Then on February 21, 1961, a Thor/Able-Star failed to place Transit 3B in a usable orbit because a part malfunctioned in the programmer before it could signal the restart of stage two from its coasting orbit. Substantially the same launch vehicle as on Transit 3A successfully launched Transit 4A into a nearly circular orbit on June 28, 1961.15 Through August 13, 1965, including the launches just discussed, the Thor/Able-Star completed a total of 20 missions with 5 failures, for a success rate of 75 percent. Quite successful for an early launch vehicle, the Thor/Able-Star also marked a step forward in satellite-launching capability.16

Even before the first launch of Thor/Able-Star, the air force had 56 begun using an Agena upper stage with the Thor, and this combi – Chapter 2 nation became a preferred choice for a great many often-classified missions, including those to place a family of reconnaissance satel­lites in orbit under what began as the WS-117L program. Initially, the Agena upper stages flew on basic Thors, but in three versions from A to D (without a C), the Agena also operated with uprated Thors, Atlases, and Titan IIIs to orbit a great many military and NASA spacecraft until 1987.17

The air force began developing the Agena in July 1956. On Oc­tober 29, 1956, that service selected Lockheed Missile Systems

Division as the prime contractor for both the WS-117L reconnais­sance satellite system and an associated upper stage that became the Agena. The engine for the Lockheed upper stage was a modified version of the Hustler propulsion unit (model 117) that Bell Aero­space had developed for the B-58 bomber’s air-to-surface missile, designated the Powered Disposable Bomb Pod. The air force can­celed the missile, but Lockheed contracted with Bell in the fall of 1957 to develop the engine for Agena.18

One change from the Hustler engine was the addition of gim – balling. Another was a nozzle closure to ensure that the Agena started in space after cutoff of the first-stage engine. The Agena stage with this engine, known as the Bell 8001, flew only once, on February 28, 1959, for the launch of Discoverer 1 by a Thor-Agena A. (Discoverer was the name publicly released for the secret Corona reconnaissance satellites, which had separated from the WS-117L program by this time.) Accounts differ as to the outcome of this first launch into a polar orbit from Vandenberg AFB, California—some claiming the launch itself was successful, and others that it was not.19

The Agena nevertheless had an extensive career as an upper stage. The Agena A operated successfully on 78 percent of its 14 launches by September 13, 1960 (all by Thors; all but one with a new Bell model 8048 engine for the Agena), with 3 failures. A more capable Thor-Agena B appears to have had 39 successful performances on 48 launches from October 26, 1960, to May 15, 1966, an 81 percent success rate, mostly launching Corona satellites. With a thrust- augmented Thor, the Agena B could launch much heavier satellites, added thrust coming from solid-propellant strap-on boosters.20

Meanwhile, in the fall of 1959 Bell began designing the engine for the Agena D, which became the standard Agena propulsion unit. From June 28, 1962, to May 25, 1972, a large number of Thor-Agena D launches occurred, but because of the classified nature of many of the payloads, a reliable and precise tally is not available. During this period, the basic booster changed from the thrust-augmented to the long-tank, thrust-augmented Thor (called Thorad) with increased burning time and improved strap-on boosters.21

Another series of upper stages used with the Thor first stage in­cluded Burner I, Burner II, and Burner IIA. Burner I actually bore little relation to Burners II and IIA. Information about it is sparse, but sources refer to it as the Altair, a derivative of the Vanguard third stage developed by Hercules Powder Company at the Allegany Ballistics Laboratory. The first launch of the Thor-Burner I occurred on January 18, 1965, with the last one taking place on March 30, 1966. There apparently were only four such launches, all from Van-

An Agena upper stage, used also for many satellite launches, serving here as the Gemini 8 target vehicle for docking. (Photo courtesy of NASA)

denberg AFB into sun-synchronous orbits, one of them a mission failure. The spacecraft were classified at the time but appear to have been Block 4A Defense Satellite Applications Program weather sat­ellites used to inform the U. S. military of weather conditions for launching reconnaissance satellites and other defense purposes, such as mission planning during the conflict in Vietnam. In 1973, the program became the Defense Meteorological Satellite Program (DMSP) and was no longer classified.22

58 Burner I was little used because of the development of Burner Chapter 2 II. Conceiving a need for a guided upper stage that would be low in cost and usable with more than one first-stage vehicle, on Septem­ber 2, 1961, the air force’s Space Systems Division (SSD) awarded study contracts to the Boeing Company and Ling-Temco-Vought, Inc., pursuant to development of what became Burner II. As a result of its initial work, Boeing won a fixed-price contract on April 1, 1965, to provide one ground-test and three flight versions of the new upper stage. By September 15, Maj. Gen. Ben I. Funk, commander of SSD, could announce the development of the new stage, which

became the smallest maneuverable upper-stage vehicle in the air force inventory.23

The primary propulsion for Burner II came from the Thiokol TE- M-364-2 (Star 37B) motor, a spherical design promoted by NASA engineer Guy Thibodeaux. Between September 15, 1966, and Feb­ruary 17, 1971, Thor-Burner II vehicles launched four Block 4A, three Block 4B, and three Block 5A Defense Satellite Applications Program weather satellites from the Western Test Range. During this same general period, the Thor-Burner II also launched scientific satellites as part of the Department of Defense’s Space Experiments Support Program managed by SSD.24

The Block 5B versions of the Defense Satellite Applications Pro­gram weather satellites were about twice as heavy as the 5A ver­sions, necessitating increased thrust for Burner II. So the air force’s Space and Missile Systems Organization (created July 1, 1967, to bring the SSD and its sister Ballistics Systems Division into a single organization headquartered in Los Angeles at the former SSD loca­tion) contracted with Boeing for an uprated Burner II that became Burner IIA. Boeing did the uprating with a minimum of modifi­cations by adding a Thiokol TE-M-442-1 motor to form a second upper stage. With the Burner IIA, a Thor first stage launched five Block 5B and two Block 5C meteorological satellites (in what be­came the DMSP) from October 14, 1971, to May 24, 1975. A final Thor-Burner IIA launch on February 18, 1976, failed because the Thor prematurely ceased firing. This last use of the Burner IIA did not spell the end of the DMSP program, however, because a Thor coupled with a Thiokol TE-M-364-15 (Star 37S) motor that had a titanium case (rather than the steel used on the Star 37B) launched 4 improved Block 5D weather satellites between Sept ember 11, 1976, and June 6, 1979. Then Atlas Es and Titan IIs launched 10 more DMSP satellites by 1999.25

A final major Thor launch vehicle was the Thor-Delta. Whereas the other Thor-based launch vehicles were primarily air force as­sets sometimes used by NASA, Delta was a NASA-developed space – launch vehicle used on occasion by the air force until near the end of the period covered by this book, when the air force began to make extensive use of Delta IIs. Since it was conceived by NASA in 1959 as an interim vehicle to lift medium payloads by using existing technology, modified only as needed for specific missions, Delta has enjoyed a remarkably long career, attesting to its success.26

The initial idea for Thor-Delta apparently came from Milton Rosen. He was working at NASA Headquarters in the Office of Space Flight Development, headed by Abe Silverstein. His imme-

diate supervisor was Abraham Hyatt, who had become the assis­tant director for propulsion following a decade of work at the navy’s Bureau of Aeronautics. At Silverstein’s behest, Rosen worked with Douglas Aircraft Company to develop the vehicle. Using compo­nents already proven in flight, NASA and Douglas eliminated the need for developmental flights. Their contract set a very ambitious goal (for 1959) of an initial 50 percent reliability with a final rate of 90 percent.27

An important asset in Delta’s development consisted of the per­sonnel from the Vanguard program, including Rosen, who brought their experience to decision-making positions at the new Goddard Space Flight Center within NASA, as well as at NASA Headquarters. At Goddard, William R. Schindler, who had worked on Vanguard, headed a small technical group that provided direction and tech­nical monitoring for the Delta program, which initially borrowed technology from the Vanguard, Thor-Able, and Titan programs. On November 24, 1962, NASA converted this technical direction to formal project management for Delta.28

On May 13, 1960, an attempt to launch the spherical, passive reflector satellite Echo with the first Thor-Delta failed when the third-stage propellants did not ignite because a small chunk of sol­der in a transistor broke loose in flight and shorted out a semicon­ductor that had passed all of its qualification tests. A similar but less costly problem with another transistor on the third Delta launch led NASA to change its specifications and testing of such components. Meanwhile, on August 12, 1960, the second Delta launch success­fully placed Echo 1 into orbit. And the remainder of the original 12 Deltas all had successful launches of a variety of payloads from the Tiros 2 through 6 weather satellites to the Telstar 1 communica­tions satellite, the first commercial spacecraft launched by NASA (on July 10, 1962, the last of the original 12 launches by a Delta).29

From this beginning, the Delta went through a long and compli­cated series of modifications and upgrades. The initial Delta could 60 launch 100 pounds of payload to geostationary (also called geosyn – Chapter 2 chronous) transfer orbit. Starting in 1962, Delta evolved through a series of models with designations such as A, B, C, D, E, J, L, M, M-6, N, 900, 904, 2914, 3914, 3910/PAM (for Payload Assist Mod­ule), 3920/PAM, 6925 (Delta II), and 7925 (also a Delta II), the last of them introduced in 1990. The payload capabilities of these versions of the vehicle increased, at first gradually and then more rapidly, so that the 3914 introduced in 1975 could lift 2,100 pounds to geo­stationary transfer orbit and the 7925 could lift 4,010 pounds (40.1 times the original capability).30

FIG. 2.6

A Delta 1910 vehicle launching Orbiting Solar Observatory 8 on June 21, 1975, showing the Castor II strap-on boosters at the base of the vehicle to add to the thrust. (Photo courtesy of NASA)

as borrowers of new technology from other programs, they experi­enced fewer birth pangs than other missiles and rockets, showing the value of shared information. They nevertheless did experience some unexpected problems that required redesign. But by (mostly) using components already tested and proven, the Delta achieved a high reliability that made it an enduring member of the launch – vehicle family. From an interim launch vehicle in 1959, it became one of the few that lasted into the 21st century, a distinction shared with the Atlas family.32

Although there were other important upper-stage engines using storable propellants, such as the Bell Agena engine (starting with the 8048 model), the most significant engines with hypergolic pro­pellants were those used in the Titans II, III, and IV, designed and built by Aerojet. Here, the previous experience with Vanguard, Able, and Able-Star undoubtedly were extraordinarily valuable, as was Aerojet’s involvement with the development of UDMH. In the lat­ter development, Aerojet propellant chemist Karl Klager had been an important contributor. Klager held a Ph. D. in chemistry from the University of Vienna (1934) and had come to this country as part of Project Paperclip quite independently of the von Braun group. He worked for the Office of Naval Research in Pasadena during 1949 and started with Aerojet in 1950. The following year, Aerojet received a contract to develop an in-flight thrust-augmentation rocket for the F-86 fighter. The device never went into production, but in develop­ing it, Aerojet engineers conducted a literature search for candidate propellants, did theoretical performance calculations, and measured physical and chemical properties in the laboratory. Together with RFNA, UDMH seemed highly promising but was not available in sufficient quantities to be used. Klager devised (and patented) pro­duction processes that yielded large quantities at reasonable prices, but workers began to get violently ill from the toxic substance. All did recover, and Aerojet learned how to control exposure to the vapors.37

AGENA.. . versatile, upper-stage rocket vehicle employs a single rocket engine which provides 16,000 pounds of thrust. The engine can be shut down and re­started in flight through ground command signals.

Agena and its payload ride into space aboard a large booster rocket. Following staging, the Agena engine "first-burn" maneuvers the vehicle and its payload into an earth – oriented parking orbit. The Agena "second-burn" is geared to each particular mission – for example, an ellip­tical earth orbit or the ejection of a payload on a trajectory to the moon or planets.

This photographic exhibit presents the Agena missions. . . managed by the Lewis Research Center since January 1963.

National Aeronautics and Space Administration Lewis Research Center

When the time came in 1960 to begin designing engines for the Titan II, UDMH had a lower specific impulse than the new missile required. Hydrazine had better performance but could detonate if used as a regenerative coolant. Aerojet was the first firm to come up with an equal mixture of hydrazine and UDMH, Aerozine 50. This fuel combination ignited hypergolically with nitrogen tetrox – ide as the oxidizer. Neither was cryogenic. And both could be stored in propellant tanks for extended periods, offering a much quicker response time than Titan I’s 15 minutes for a propulsion system burning liquid oxygen and kerosene. Both Aerojet and Martin, the
overall contractor for Titan I, urged a switch to the new propellants, but it was apparently Robert Demaret, chief designer of the Titan, and others from Martin who proposed the idea to the air force’s Bal­listic Missile Division in early 1958.38

In May 1960, the air force signed a letter contract with the Mar­tin Company to develop, produce, and test the Titan II. On Octo­ber 9, 1959, Aerojet had already won approval to convert the Titan I engines to burn storable propellants. Research and development to that end began in January 1960. The Aerojet engineers also worked to achieve the improved performance called for in the April 30, 1960, plan for Titan II. Although the Titan II engines were based on those for Titan I, the new propellants and the requirements in the April 30 plan necessitated considerable redesign. Because the modi­fications did not always work as anticipated, the engineers had to resort to empirical solutions until they found the combinations that worked correctly and provided the necessary performance. Since the propellants were hypergolic, there was no need for an igniter in the Titan II engines (called XLR87-AJ-5 for the two engines in stage one and XLR91-AJ-5 for the single stage-two engine). The injectors for the Titan I engines had used alternating fuel and oxidizer passages with oxidizer impinging on oxidizer and fuel on fuel (called like-on – like) to obtain the necessary mixing of the two propellants. For the Titan II, Aerojet engineers tried fuel-on-oxidizer impingement. This evidently mixed the droplets of propellant better because higher performance resulted. But the improvement led to erosion of the injector face, necessitating a return to the like-on-like pattern.

This older arrangement caused combustion instability in the stage-two engines, and engineers tried several configurations of baf­fles to solve the problem before they came up with one that worked. One potential solution, uncooled stainless-steel baffles, did not last through a full-duration engine test. Copper baffles with both pro­pellants running through them for cooling resulted in corrosion of the copper from the nitrogen tetroxide. An eight-bladed configura­tion cooled by the oxidizer (evidently using another type of metal) yielded poor performance. The final configuration was a six-bladed, wagon-wheel design (with the baffles radiating outward from a cen­tral hub), again cooled by the oxidizer. This solved the problem, at least at for the time being.39

The turbopumps for Titan II were similar to those for Titan I, but differences in the densities of the propellants necessitated greater power and lower shaft speed for the Titan II pumps. This resulted in increased propellant flow rates. But Aerojet engineers had to rede­sign the gears for the turbopumps, making them wider and thus able

to withstand greater “tooth pressures" caused by the higher power. From two blades in the inducer for Titan I, the design went to three blades. Engineers also had to redesign the impeller and housing pas­sages to accept the higher flow rates, and there had to be new mate­rials that would not be degraded by the storable propellants.40

A significant innovation for the Titan II replaced the use of pressurized nitrogen (in stage one of Titan I) or helium (in stage two) to initiate propellant flow with a so-called autogenous (self­generating) system. Solid-propellant start cartridges initiated the process by spinning the turbines, whereupon gas generators kept the turbines spinning to pressurize the fuel tanks in both stages and to pump the Aerozine 50 into the thrust chamber. The second – stage oxidizer tank did not need pressurization because acceleration was sufficient to keep the nitrogen tetroxide flowing. In the first – stage propulsion system, though, oxidizer from the pump discharge served to pressurize the tank. The result was a simplified system saving the weight from the pressurized-gas storage tanks in Titan I and requiring no potentially unreliable pressure regulators. A simi­lar increase in reliability and saving in weight came from using the exhaust stream from the turbopump in stage two to provide roll control in place of an auxiliary power-drive assembly used in Titan I for the vernier thrusters. Gimbals (used in Titan I) continued to
provide pitch, roll, and yaw control in the first stage plus pitch and yaw control in the second stage.41

The Titan II propulsion system had significantly fewer parts than its Titan I predecessor. The number of active control components fell from 125 to 30; valves and regulators declined from 91 to 16. The engines had higher thrust and performance, as planned. The Titan II first-stage engines had a combined thrust of 430,000 pounds at sea level, compared to 300,000 for Titan I. Second-stage thrust rose from 80,000 pounds for Titan I to 100,850 pounds for Titan II. Specific impulses rose less dramatically, from slightly above 250 to almost 260 lbf-sec/lbm at sea level for stage one and remained slightly above 310 at altitude (vacuum) for stage two.42

A 1965 Aerojet news release on Titan II propulsion credited it to “the efforts of hundreds of men and women" but singled out four of them as leaders of the effort. Three of their biographies illustrate the way that engineers in aerospace migrated from one firm to an­other or from government work to the private sector, carrying their knowledge of various technologies with them. Robert B. Young was the overall manager of the design, development, and production ef­fort for both Titan I and Titan II. He was a chemical engineering graduate of Caltech, where Theodore von Karman had encouraged him to devote his knowledge to rocketry. He had worked for a year as director of industrial liaison on the Saturn program at NASA’s Marshall Space Flight Center, during part of the Titan years, and had risen within Aerojet’s own structure from a project engineer to a vice president and manager of the Sacramento, California, plant. Another leader was Ray C. Stiff Jr., who had discovered aniline as a hypergolic propellant while working at the navy Engineering Exper­iment Station in Annapolis. His role in the use of self-igniting pro­pellants in JATOs to assist a PBY into the air while carrying heavy loads proved a forerunner “of the storable, self-igniting propellants used to such advantage in the Titan II engine systems." Stiff served as Aerojet’s manager of Liquid Rocket Operations near Sacramento and was also a vice president.

A third manager Aerojet mentioned in the release was A. L. Feldman, a rocket engineer educated at Cornell University. While working at Convair, he had been “in on the initial design and devel­opment effort for the Atlas engines." After moving to Aerojet, he served as manager for both the Titan I and Titan II engines and as­sistant manager of Aerojet’s Liquid Rocket Operations. A fourth key manager was L. D. Wilson, who earned a degree in engineering at Kansas State University. “At 28, he managed the painstaking design, development, test and production of the ‘space start’ second-stage

engine for Titan I." Wilson then managed the entire propulsion sys­tem for the Titan II.43 Alone among the four managers, he appeared not to have worked for another rocket effort.

Between March 16, 1962, and April 9, 1963, there were 33 re- search-and-development test flights of Titan II missiles—23 from Cape Canaveral and 10 from Vandenberg AFB. Depending on who was counting, there were variously 8, 9, or 10 failures and partial successes (none of which occurred during the last 13 flights), for a success rate of anywhere from 70 to 76 percent. The problems ranged from failure of electrical umbilicals to disconnect properly on the first, otherwise-successful, Titan II silo launch (from Van­denberg) on February 16, 1963 (pulling missile-guidance cabling with them and causing an uncontrollable roll), to premature en­gine shutdown, an oxidizer leak, a fuel-valve failure, a leak in a fuel pump, and gas-generator failure. But the most serious problem was initially a mystery whose cause was unclear to the engineers in­volved. On the launch of the first Titan II on March 16, 1962, about a minute and a half after liftoff from Cape Canaveral, longitudinal oscillations occurred in the first-stage combustion chambers. They arose about 11 times per second for about 30 seconds. They did not prevent the missile (designated N-2) from traveling 5,000 nautical miles and impacting in the target area, but they were nonetheless disquieting.44

164 The reason for concern was that in late 1961 NASA had reached Chapter 4 an agreement with the Department of Defense to acquire Titan IIs for launching astronauts into space as part of what soon became Project Gemini. Its mission, as a follow-on to Project Mercury and a predecessor to Project Apollo’s Moon flights, was to determine if one spacecraft could rendezvous and dock with another, whether astronauts could work outside a spacecraft in the near weightless­ness of space, and what physiological effects humans would experi­ence during extended flight in space. Pogo (as the longitudinal oscil­lations came to be called ) was not particularly problematic for the missile, but it posed potentially significant problems for an astro­naut who was already experiencing acceleration of about 2.5 times the force of gravity from the launch vehicle. The pogo effect on the first Titan II launch added another ±2.5 Gs, which perhaps could have incapacitated the pilot of the spacecraft from responding to an emergency if one occurred.

Fixing the problem for Project Gemini, however, was compli­cated by an air force reorganization on April 1, 1961, creating Bal­listic Systems Division (BSD) and Space Systems Division (SSD). The problem for NASA lay in the fact that while BSD was intent

on developing Titan II as a missile, SSD would be responsible for simultaneously adapting Titan II as a launch vehicle for the astro­nauts.45 A conflict between the two air force interests would soon develop and be adjudicated by General Schriever, then commanding the parent Air Force Systems Command.

Meanwhile, there were further organizational complications for Project Gemini. SSD assigned development of the Gemini-Titan II launch vehicle to Martin’s plant in Baltimore, whereas the Denver plant was working on the Titan II missile. Assisting SSD in manag­ing its responsibilities was the nonprofit Aerospace Corporation. For Gemini, Aerospace assigned James A. Marsh as manager of its efforts to develop the Titan II as a launch vehicle. BSD established its own committee to investigate the pogo oscillations, headed by Abner Rasumoff of Space Technology Laboratories (STL). For the missile, it found a solution in higher pressure for the stage-one fuel tank, which reduced the oscillations and resultant gravitational forces in half on the fourth launch on July 25, 1962, without STL engineers’ understanding why. Martin engineers correctly thought the problem might lie in pressure oscillations in the propellant feed lines. They suggested installation of a standpipe to suppress surges in the oxidizer lines of future test missiles. BSD and NASA’s Manned Spacecraft Center, which was managing Gemini, agreed.46

Although the standpipe later proved to be part of the solution to the pogo problem, initially it seemed to make matters worse. Installed on missile N-11, flying on December 6, 1962, it failed to suppress severe oscillations that raised the gravitational effect from pogo alone to ±5 Gs. This lowered the chamber pressure to the point that instrumentation shut down the first-stage engine prema­turely. The following mission, N-13, on December 19, 1962, did not include the standpipe but did have increased pressure in the fuel tank, which had seemed to be effective against the pogo effect on earlier flights. Another new feature consisted of aluminum oxidizer feed lines in place of steel ones used previously. For reasons not fully understood, the pogo level dropped and the flight was successful.

Missile N-15, evidently with the same configuration as N-13, launched on January 10, 1963. The pogo level dropped to a new low, ±0.6 G, although problems with the gas generator in stage two se­verely restricted the vehicle’s range. This was still not a low enough oscillation level for NASA, which wanted it reduced to ±0.25 G, but it satisfied BSD, which “froze" the missile’s design with regard to further changes to reduce oscillations. Higher pressure in the first-stage fuel tanks plus use of aluminum for the oxidizer lines had reduced the pogo effect below specifications for the missile,

and BSD believed it could not afford the risks and costs of further experimentation to bring the pogo effect down to the level NASA wanted.47

NASA essentially appealed to Schriever, with NASA’s Brainerd Holmes, deputy associate administrator for manned space flight, complaining that no one understood what caused either the pogo oscillations or unstable combustion, another problem affecting Titan II. So Holmes said it was impossible to “man-rate" the mis­sile as a launch vehicle. The result, on April 1, 1963, was the forma­tion of a coordinating committee to address both problems, headed by BSD’s Titan II director and including people from Aerospace Cor­poration and STL.

After engineers from Aerojet, Martin, STL, and Aerospace stud­ied the problems, Sheldon Rubin of Aerospace looked at data from static tests and concluded that as the fuel pumped, a partial vacuum formed in the fuel line, causing resonance. This explained why the oxidizer standpipes had failed to suppress the pogo effect. The solu­tion was to restore the standpipe, keep the increase in tank pressure and the aluminum oxidizer feed lines, and add a fuel surge chamber (also called a piston accumulator) to the fuel lines. Nitrogen gas pressurized the standpipe after the nitrogen tetroxide had filled the oxidizer feed lines. An entrapped gas bubble at the end of the stand­pipe absorbed pressure pulsations in the oxidizer lines. The surge 166 chamber included a spring-loaded piston. Installed perpendicular to Chapter 4 the fuel feed line, it operated like the standpipe to absorb pressure pulses. Finally, on November 1, 1963, missile N-25 carried both of these devices. The successful flight recorded pogo levels of only ±0.11 G, well below NASA’s maximum of ±0.25 G. Subsequent tests on December 12, 1963, and January 15, 1964, included the sup­pression devices and met NASA standards, the January 15 mission doing so even with lower pressures in the fuel tank. This seemed to confirm that the two devices on the propellant lines had fixed the pogo problem.48

Meanwhile, the combustion instability Holmes had mentioned at the same time as the pogo problem, turned out to be another major issue. It never occurred in flight, but it appeared in a severe form during static testing of second-stage engines. Several engines experienced such “hard starts" that the combustion chambers fell from the injector domes as if somebody had cut them away with a laser beam. Engineers examined the test data and concluded that combustion instability at the frequency of 25,000 cycles per second had sliced through the upper combustion chamber near the face of the injector with a force of ± 200 pounds per square inch. This oc-

curred on only 2 percent of the ground tests of second-stage engines, but for Gemini, even this was too high.

Aerojet instituted the Gemini Stability Improvement Program (GEMSIP) in Sept ember 1963 to resolve the problem with com­bustion instability. Apparently, the instability occurred only when second-stage engines were tested in an Aerojet facility that simu­lated the air pressure at 70,000 feet, because there was no combus­tion instability in the first-stage engines at this stage of their devel­opment. The reason was that air pressure at sea level slowed the flow of propellants through the injectors. Aerojet engineers could solve the problem for the second stage by filling the regenerative – cooling tubes that constituted the wall of the combustion chamber with a very expensive fluid that afforded resistance to the rapid flow of propellants similar to that of air pressure at sea level.

Aerojet and the air force finally agreed on a more satisfactory solution, however. This involved a change in injector design with fewer but larger orifices and a modification of the six-bladed baffles radiating out from a hub that Aerojet thought had already solved the combustion-instability problem in the Titan II missile. Aerojet engineers increased the number of baffles to seven and removed the hub for the Gemini configuration. Testing in the altitude chamber at the air force’s Arnold Engineering Development Center proved that the new arrangement worked.49

A final problem that occurred in flight testing and required re­engineering involved the gas generator in the stage-two engine. This issue was a matter of concern to the Titan II Program Office at BSD, not just to SSD and NASA. Engineers and managers first became aware there was a problem on the second Titan II test flight (June 7, 1962) when telemetry showed that the second-stage engine had achieved only half of its normal thrust soon after engine start. When the tracking system lost its signal from the vehicle, the range safety officer caused cutoff of the fuel flow, making the reentry ve­hicle splash down well short of its target area. The data from telem­etry on this flight were inadequate for the engineers to diagnose the problem. It took two more occurrences of gas-generator problems to provide enough data to understand what was happening. Particles were partially clogging the small openings in the gas generator’s in­jectors. This restricted propellant flow and resulted in loss of thrust. Technicians very thoroughly removed all foreign matter from com­ponents of the generators in a clean room before assembly, sepa­rated the generators from the engines in transport, and subjected the assemblies to blowdown by nitrogen before each test flight to ensure no foreign particles were present.

These methods did not solve the problem but did narrow the list of sources for the particles. It became apparent that there was no problem with stage-one gas generators because sea-level air pressure did not allow particles from the solid-propellant start cartridges to reach the injector plates for the generators. Aerojet had tested the system for stage two in its altitude chamber, but it could simulate only 70,000 feet of altitude, which engineers assumed was high enough. There had been no problems with gas generators during the tests. It turned out, however, that even at 70,000 feet, enough atmo­sphere was present to cushion the injectors from the particles. At 250,000 feet, where the stage-two engine ignited, the atmosphere was so thin that particles from the start cartridge flowed into the gas generator, sometimes in sufficient quantity to cause difficulties. To rectify the problem, engineers added a rupture disk to the exhaust of the gas generator. This kept enough pressure in the system to cush­ion the flow of particles until ignition start, when the disk ruptured. Gas-generator failure was not a problem after this design change.50

The engines for Titan II went on to form the core of the Titan III and IV space-launch vehicles. Titan IIs also became launch vehicles in their own right, making their engines key propulsion elements in three different series of launch vehicles. Their use through the end of the period covered by this book suggested the importance of storable propellants for the history of space flight. But development 168 of hypergolic propulsion did not end with these rockets.

Chapter 4

Despite this preparatory work for Minuteman, the missile did not begin formal development until the air force secured final DoD ap­proval in February 1958. Meanwhile, civilian engineers employed by the air force at the Non-Rotating Engine Branch at Wright-Patterson AFB had continued efforts to develop large, solid-propellant motors. Perhaps from sources at Aerojet, they learned about adding large quantities of aluminum to a solid propellant to increase its perfor­mance. They combined this with other information they had been gathering on solid propellants while Hall was still with the branch. Bill Fagan from the branch carried this information to Hall at the Ballistic Missile Division (as the WDD had become).52

The specifics of Minuteman technology continued to evolve, but its basic concept took advantage of the reduced complexity of solids over liquids. This cut the number of people needed to launch it. Each missile could be remotely launched from a control center us­ing communications cables, without a crew of missileers in atten­dance. The air force initially considered using mobile Minutemen but ultimately decided to launch them from silos. Because solids were more compact than liquids, Minuteman IA (by one account) was only 53.8 feet long to Atlas’s 82.5, Titan I’s 98, and Titan II’s 108 feet in length. Its diameter was slightly over half that of the other three missiles (5.5 feet to the others’ 10), and its weight was only 65,000 pounds to Atlas’s 267,136, Titan I’s 220,000, and Titan II’s 330,000 pounds. All of this made the costs of silos much lower and substantially reduced the thrust needed to launch the missile. Consequently, the initial costs of Minuteman were a fifth and the annual maintenance costs a tenth those of Titan I. Moreover, a crew of two could launch 10 Minutemen, whereas it took six people to launch each Titan I—a 30-fold advantage in favor of Minuteman.53

Minuteman development used multiple approaches to arrive at individual technologies, as had been true with Atlas. Ballistic Mis­sile Division (BMD) contracted separately with Aerojet and Thiokol to work on all three stages of the missile. A later contract with Her­cules assigned it to work on the third stage, too. As firms developed technologies, parallel development gave way to specific responsi-

bilities, with Thiokol building the first stage, Aerojet the second, and Hercules the third. Space Technology Laboratories retained its role in systems engineering and technical direction at BMD. These and other companies and organizations all sent representatives to frequent program-review meetings and quarterly gatherings of top officials. They examined progress and identified problems requiring solution. Then, the relevant organizations found solutions to keep the program on track.54

One major technical challenge involved materials for the nozzle throats and exit cones. The addition of aluminum to the propellant provided a high enough specific impulse to make Minuteman fea­sible, as it had done for Polaris, and its combustion produced alu­minum oxide particles that damped instabilities in the combustion 248 chamber. But the hot product flow degraded the nozzle throats and Chapter 6 other exposed structures. It seemed that it might not be possible to design a vectorable nozzle that would last the 60 seconds needed for the missile to reach its ballistic trajectory so it could arrive accu­rately on target. Solving this problem required “many months and many dollars. . . spent in a frustrating cycle of design, test, failure, redesign, retest, and failure." The Minuteman team used many dif­ferent grades and exotic compounds of graphite, which seemed the most capable material, but all of them experienced blowouts or per­formance-degrading erosion. One solution was a tungsten throat in­sert, a compromise in view of its high weight and cost. For the exit cones, the team tried Fiberite molded at high pressure and loaded with silica and graphite cloth. It provided better resistance to ero­sion than graphite but still experienced random failures.55

Another significant problem concerned the vectorable feature of the nozzles. Polaris had solved its steering problem with jetavators, but flight-control studies for Minuteman showed a need for the stage-one nozzles to vector the thrust eight degrees, more than Min – uteman engineers thought jetavators could deliver. Thiokol’s test of the motor in 1959—the largest solid-propellant powerplant yet built—resulted in the ejection of all four nozzles after 30 millisec­onds of firing, well before the full stage had ignited. Five successive explosions of the motors and their test stands occurred in October 1959, each having a different failure mode. BMD halted first-stage testing in January 1960. Discussion among BMD, STL, and Thiokol personnel revealed two problem areas, internal insulation and the nozzles themselves, as masking other potential problems.

There followed two concurrent programs of testing. Firings with battleship-steel cases tested movable nozzles, while partici­pants used flight-weight cases to test a single, fixed nozzle massive

enough to sustain a full-duration firing. Thiokol solved the problem with insulation by summer but did not resolve the nozzle issue un­til fall. This was close to the date set for all-up testing of the entire missile. However, General Phillips had ordered Thiokol to begin manufacturing the first stage except the nozzles, permitting instal­lation of the nozzles as soon as their problem was solved.56

Among many other difficulties, a major concern was launching from a silo. The first successful launch of the missile occurred in early February 1961, which Phillips referred to as “December 63rd" since the planned date had been in December 1960. It and two suc­ceeding launches took place from a surface pad and were so suc­cessful that the team advanced the first silo launch to August 1961. The missile blew up in the silo, giving credence to critics in STL who had argued that firing a missile from a silo was impossible. As Phillips said, the missile “really came out of there like a Roman candle."57

Fortunately, team members recovered enough of the guidance system from the wreckage to find that the problem was not the silo launch itself but quality control. Solder tabs containing con­nections had vibrated together, causing all of the stages to ignite simultaneously. Knowing this, the team was able to prepare the fifth missile for flight with the problem solved by mid-November, when it had a successful flight from the silo. Previous testing of silo launches aided this quick recovery. In early 1958, BMD and STL engineers had arranged for development of underground silos. They divided the effort into three phases, with the first two using only subscale models of Minuteman. The third tested full-scale models. Most testing occurred at the air force’s rocket site on Edwards AFB in the remote Mojave Desert, but Boeing (contractor for missile as­sembly and test) tested subscale models in Seattle.

At the rocket site on Leuhman Ridge at Edwards, subscale silos investigated heat transfer and turbulence in some 56 tests by No­vember 1958. Meanwhile, Boeing modeled the pressures that the rocket’s exhaust gases imparted to the missile and silo. It also ex­amined acoustic effects of the noise levels generated by the rocket motor in the silo on delicate systems such as guidance. Armed with such data, engineers at Edwards began one-third-scale tests in Feb­ruary 1959. Full-scale tests initially used a mock-up made of steel plate with ballast to match the weight and shape of the actual mis­sile and only enough propellant (in the first-stage motor) to provide about three seconds of full thrust—enough to move the missile, on a tether, out of the silo and to check the effects of the thrust on the silo and missile. They configured the tether so that the missile

would not drop back on the silo and damage it. These tests ensured that the second silo launch at Cape Canaveral on November 17, 1961, was fully successful.58

The all-up testing on Minuteman was itself a significant inno­vation later used on other programs, including the Saturn launch vehicle for the Apollo program. It differed from the usual practice for liquid-propellant missiles—gradually testing a missile’s differ­ent capabilities over a series of ranges and tests (for first-stage or booster propulsion, for second-stage or sustainer-engine propulsion, for guidance/control, and so forth). For the first time, it tested all of the missile’s functions at once over the full operational range. Al­though the practice accorded well with general procedures at BMD, such as concurrency, its use came about in an odd way. According 250 to Otto Glasser, he was briefing Secretary of the Air Force James H.

Chapter 6 Douglas on Minuteman, with Gen. Curtis LeMay, vice chief of staff of the air force, sitting next to Douglas. Douglas insisted Glasser had moved the first flight of the missile to a year later than the original schedule. Glasser protested (to no avail) that this was not the case, and the only way he could conceive to cut a year out of the develop­ment process was all-up testing. “Boy, the Ramo-Wooldridge crowd came right out of the chair on that," Glasser said. They protested “a test program. . . with that sort of lack of attention to all normal, sensible standards." But all-up testing “worked all the way."59

Overcoming these and other problems, BMD delivered the first Minuteman I to the Strategic Air Command in October 1962, al­most exactly four years after the first contracts had been signed with contractors to begin the missile’s development. This was a year earlier than initially planned because of the speeded-up sched­ule. The missile in question was the A model of Minuteman I, later to be succeeded by a B model. The former was 53.7 feet long and consisted of three stages plus the reentry vehicle. Thiokol’s first stage included a new propellant binder developed by the company’s chemists from 1952 to 1954. Thiokol first tried a binder called poly­butadiene-acrylic acid, or PBAA, an elastomeric (rubberlike) copo­lymer of butadiene and acrylic acid that allowed higher concentra­tions of solid ingredients and greater fuel content than previous propellants. It had a higher hydrogen content than earlier Thiokol polysulfide polymers. With PBAA, a favorable reaction of oxygen with the aluminum generated significant amounts of hydrogen in the exhaust gases, reducing the average molecular weights of the combustion products (since hydrogen is the lightest of elements). This added to the performance, with Minuteman being the first rocket to use the new binder.60

But testing showed PBAA had a lower tear strength than poly­sulfide, so Thiokol added 10 percent acrylonitrile, creating poly­butadiene-acrylic acid-acrylonitrile (PBAN). The binder and curing agent constituted only 14 percent of the propellant, with ammonium perchlorate (oxidizer) and aluminum (fuel) the two other major in­gredients. The combination yielded a theoretical specific impulse of more than 260 lb-sec/lb, with the actual specific impulse at sea level at 70°F somewhat lower than 230.61

For stage two of Minuteman I, Aerojet used the polyurethane binder employed in Polaris, with ammonium perchlorate as the oxidizer and aluminum powder the major fuel. It used two slightly different propellant grains, with a faster-burning inner grain and a slower-burning outer one. The combination resulted in a conversion of the four-point, star-shaped, internal-burning cavity to a cylindri­cal one as the propellant burned, avoiding slivers of propellant that did not burn. The propellant yielded a vacuum specific impulse of nearly 275 lbf-sec/lbm at temperatures ranging from 60°F to 80°F.62

For stage three, the Hercules Powder Company used a glass- filament-wound case instead of the steel employed on stages one and two, plus a very different propellant than for the first two stages. The third stage featured four phenolic-coated aluminum tubes for thrust termination and a grain consisting of two separate composi­tions. The one used for the largest percentage of the grain included the high-explosive HMX, combined with ammonium perchlorate, nitroglycerin, nitrocellulose, aluminum, a plasticizer, and a stabi­lizer. The second composition had the same basic ingredients mi­nus the HMX and formed a horizontal segment at the front of the motor. A hollow core ran from the back of the motor almost to the segment containing the non-HMX composition. It was roughly cone shaped before tapering off to a cylinder. This motor yielded a specific impulse of more than 275 lbf-sec/lbm at temperatures ranging from 60°F to 80°F. The four nozzles for stage three of Min – uteman I rotated in pairs up to four degrees in one plane to provide pitch, yaw, and roll control.63 Minuteman I, Wing I became opera­tional at Malmstrom AFB, Montana, in October 1962.64

It would be tedious to follow the evolution of Minuteman through all the improvements in its later versions, but some discussion of the major changes is appropriate. Wings II through V of Minuteman I (each located at a different base) featured several changes to increase the missile’s range. This had been shorter than initially planned be­cause of the acceleration of the Minuteman I schedule. The shorter range was not a problem at Malmstrom because it was so far north (hence closer to the Soviet Union), but range became a problem

starting with Wing II. Consequently, for it and subsequent missiles, more propellant was added to the aft dome of stage one, and the exit cone included contouring that made the nozzle more efficient. In stage two, the material for the motor case was changed from steel to titanium. Titanium is considerably lighter than steel but more expensive. Since each pound of reduced weight yielded an extra mile of range, use of titanium seemed worth the extra cost. The nozzles also were lighter. Overall, the reduction in weight totaled slightly less than 300 pounds despite an increase in propellant weight. The increase in propellant mass plus the decrease in weight yielded a range increase of 315 miles to a figure usually given as 6,300 nauti­cal miles. There were no significant changes to stage three.65

MINUTEMAN II

For Minuteman II, the major improvements occurred in Aerojet’s stage two. There had been problems with cracking and ejection of graphite from the nozzles and aft closure of stage one. An air force reliability improvement program solved these difficulties. There had also been problems with insulation burning through in the aft dome area of stage three. Unspecified design changes inhibited the flow of hot gases in that region. Stage two, however, featured an entirely new rocket motor with a new propellant, a slightly greater length, a substantially larger diameter, and a single fixed nozzle that used a liquid-injection thrust-vector-control system for directional control.66

The new propellant was carboxy-terminated polybutadiene (CTPB), which propellant companies other than Aerojet had devel­oped. Some accounts attribute its development to Thiokol, which first made the propellant in the late 1950s and converted it into a useful propellant in the early 1960s. Initially, Thiokol chemists used an imine known as MAPO and an epoxide in curing the CTPB. It turned out that the phosphorous-nitrogen bond in the imine was susceptible to hydrolysis, causing degradation and softening of the propellant. According to Thiokol historian E. S. Sutton, “The post­curing problem was finally solved by the discovery that a small amount of chromium octoate (0.02%) could be used to catalyze the epoxide-carboxyl reaction and eliminate this change in properties with time." A history of Atlantic Research Corporation agrees that Thiokol produced the CTPB but attributes the solution of the curing problem to ARC, which is not incompatible with Sutton’s account. According to the ARC history, “ARC used a complex chromium compound, which would accelerate the polymer/epoxy reaction,

paving the way for an all epoxy cure system for CTPB polymer." The result was “an extremely stable binder system."67

It frequently happens in the history of technology that innova­tions occur to different people at about the same time. This ap­pears to have been the case with CTPB, which Aerojet historians attribute to Phillips Petroleum and Rocketdyne without providing details. These two companies may have been the source for infor­mation about the CTPB that Aerojet used in Minuteman II, stage two. Like Thiokol, in any event, Aerojet proposed to use MAPO as a cross-linking agent. TRW historians state that their firm’s labo­ratory investigations revealed the hydrolysis problem. They state that “working with Aerojet’s research and development staff," they developed “a formulation that eliminated MAPO. . . ." The CTPB that resulted from what apparently was a multicompany develop­ment effort had better fuel values than previous propellants, good mechanical properties such as the long shelf life required for silo – based missiles, and a higher solids content than previous binders. The propellant consisted primarily of CTPB, ammonium perchlo­rate, and aluminum. It yielded a vacuum specific impulse more than 15 lbf-sec/lbm higher than the propellant used in stage two of Minuteman I, Wing II.68

Although CTPB marked a significant step forward in binder tech­nology, it was not as widely used as it might have been because of its higher cost compared with PBAN. Another factor was the emer­gence in the late 1960s of an even better polymer with lower viscos­ity and lower cost, hydroxy-terminated polybutadiene (HTPB). It became the industry standard for newer tactical rockets. HTPB had many uses as an adhesive, sealant, and coating, but to employ it in a propellant required, among other things, the development of suit­able bonding agents. These tightly linked the polymer to such solid ingredients as ammonium perchlorate and aluminum. Without such links, the propellant could not withstand the temperature cy­cling, ignition pressure, and other forces that could cause the solid particles to separate from the binder network. This would produce voids in the grain that could result in cracks and structural failure.

A key figure in the development of HTPB for use as a binder was Robert C. Corley, who served as a research chemist and project manager at the Air Force Rocket Propulsion Laboratory at Edwards AFB from 1966 to 1978 and rose through other positions to become the lab’s chief scientist from 1991 to 1997. But many other people from Thiokol, Aerojet, the army at Redstone Arsenal, Atlantic Re­search, Hercules, and the navy were also involved. Even HTPB did

not replace PBAN for all uses, including the Titan III, Titan IVA, and Space Shuttle solid-rocket motors, because PBAN could be pro­duced for the comparatively low cost of $2.50 per pound at a rate in the 1980s of 4 million pounds per year, much higher than for any other propellant.69

To return to Minuteman II, however, the second major change in the stage-two motor was the shift to a single nozzle with liquid thrust vector control replacing movable nozzles for control in pitch and yaw. Static firings had shown that the same propellants pro­duced seven to eight points less specific impulse when fired from four nozzles than from a single one. With the four nozzles, liquid particles agglomerated in their approach sections and produced exit-cone erosion, changing the configuration of the exit cone in 254 an unfavorable way. The solution was not only a single nozzle on Chapter 6 Minuteman II’s second stage but also the change in thrust vector control. The navy had begun testing a Freon system for thrust vec­tor control in the second stage of Polaris A3 in September 1961, well before the Minuteman II, stage-two program began in Febru­ary 1962. The system was low in weight, was insensitive to pro­pellant flame temperature, and posed negligible constraints on the design of the nozzle. The Minuteman engineers adopted it—but one more example of borrowings back and forth between the Polaris and Minuteman projects despite the air force’s view of Polaris as a threat to its roles and missions.

Despite this pioneering work by the navy and its contractors, ac­cording to TRW historians, their firm still had to determine how much “vector capability" stage two of Minuteman II would require. TRW analyzed the amount of injectant that could be used before sloshing in the tank permitted the ingestion of air, and it determined the system performance requirements. Since Aerojet was involved in the development of the system for Polaris, probably its participa­tion in this process was also important. In any event, the Minute – man team, like the navy, used Freon as the injectant, confining it in a rubber bladder inside a metal pressure vessel. Both TRW and Aerojet studied the propensity of the Freon to “migrate" through the bladder wall and become unavailable for its intended purpose. They found that only 25 of 262 pounds of Freon would escape, leav­ing enough to provide the necessary control in pitch and yaw. A separate solid-propellant gas generator provided roll control. In addi­tion to these changes, stage two of Minuteman II increased in length from 159.2 inches for Minuteman I to 162.32 inches. The diameter increased from 44.3 to 52.17 inches, resulting in an overall weight increase from 11,558.9 to 15,506 pounds. Some 3,382.2 of this ad-

ditional 3,947.1 pounds consisted of propellant weight. Even so, the propellant mass fraction decreased slightly from 0.897 to 0.887.70

The Strategic Air Command put the first Minuteman II squadron on operational alert in May 1966, with initial operational capability declared as of December 1966. In the next few years, the air force began replacing Minuteman Is with Minuteman IIs.71

MINUTEMAN III

Minuteman III featured multiple, independently targetable reentry vehicles with a liquid fourth stage for deployment of this payload. This last feature was not particularly relevant to launch-vehicle de­velopment except that the added weight required for it necessitated higher booster performance. Stages one and two did not change from Minuteman II, but stage three became larger. Hercules lost the contract for the larger motor to Aerojet. Subsequently, Thiokol and a new organization, the Chemical Systems Division of United Technologies Corporation, won contracts to build replacement mo­tors. Stage three featured a fiberglass motor case, the same basic propellant Aerojet had used in stage two only in slightly different proportions, a single nozzle that was fixed in place and partially submerged into the case, a liquid-injection thrust-vector-control system for control in pitch and yaw, a separate roll-control system, and a thrust-termination system.

Aerojet had moved its filament-wound case production to Sacra­mento. It produced most of the Minuteman fiberglass combustion chambers there but ceased winding filament in 1965. Meanwhile, Young had licensed his Spiralloy technology to Black, Sivalls, and Bryson in Oklahoma City, which became a second source for the Minuteman third-stage motor case. This instance and the three firms involved in producing the third stage illustrate the extent to which technology transferred among the contractors and subcon­tractors for government missiles and rockets.

The issue of technology transfer among competing contractors and the armed services, which were also competing over funds and missions, is a complex one about which a whole chapter—even a book—could be written. To address the subject briefly, there had been a degree of effort to exchange knowledge about rocket propul­sion technology beginning in 1946 when the navy provided fund­ing for a Rocket Propellant Information Agency (RPIA) within the Johns Hopkins University’s Applied Physics Laboratory. The army added support in 1948, and the RPIA became the Solid Propellant Information Agency (SPIA). The air force joined the other services in 1951. After the Sputnik launch, the newly created NASA be-

gan participating in SPIA activities in 1959. Meanwhile, the navy created the Liquid Propellant Information Agency (LIPA) in 1958. The SPIA and LPIA combined on December 1, 1962, to create the Chemical Propulsion Information Agency (CPIA).

With the further development of rockets and missiles, the need had become obvious by 1962 for a better exchange of information. So the DoD created an Interagency Chemical Rocket Propulsion Group in November 1962, the name later changing to the Joint Army/Navy/NASA/Air Force (JANNAF) Interagency Propulsion Committee. Together with CPIA, JANNAF effectively promoted sharing of technology. In addition, “joint-venture" contracts, pio­neered by Levering Smith of the navy, often mandated the sharing of manufacturing technology among companies. These contracts 256 served to eliminate the services’ dependence for a given technology Chapter 6 on sole sources that could be destroyed by fire or possible enemy targeting. It also provided for competitive bidding on future con­tracts. The air force had a similar policy.72

Meanwhile, the propellant for Aerojet’s third stage of Minuteman had less CTPB and more aluminum than the second stage. The grain configuration consisted of an internal-burning cylindrical bore with six “fins" radiating out in the forward end. The igniter used black – powder squibs to start some of the CTPB propellant, which in turn spread the burning to the grain itself. The 50 percent submerged nozzle had a graphite phenolic entrance section, a forged tungsten throat insert, and a carbon-phenolic exit cone. As compared with the 85.25-inch-long, 37.88-inch-diameter third stage of Minuteman II, that for Minuteman III was 91.4 inches long and 52 inches in di­ameter. The mass fraction improved from 0.864 to 0.910, and with a nearly 10-lbf-sec/lbm greater specific impulse, the new third stage had more than twice the total impulse of its predecessor—2,074,774 as compared with 1,006,000 pounds force per second.73

The thrust-vector-control system for the new stage three was similar to that for stage two except that strontium perchlorate was used instead of Freon as the injectant into the thrust stream to pro­vide control in the pitch and yaw axes. Helium gas provided the pressure to insert the strontium perchlorate instead of the solid – propellant gas generator used in the second stage. Roll control again came from a gas generator supplying gas to nozzles pointing in opposite directions. When both were operating, there was neutral torque in the roll axis. When roll torque was required, the flight – control system closed a flapper on one of the nozzles, providing unbalanced thrust to stop any incipient roll.

To ensure accuracy for the delivery of the warheads, Minute – man had always required precise thrust termination for stage three, determined by the flight-control computer. On Minuteman I, the thrust-termination system consisted of four thick carbon-phenolic tubes integrally wound in the sidewall of the third-stage case and sealed with snap-ring closures to form side ports. Detonation of explosive ordnance released a frangible section of the snap ring, thereby venting the combustion chamber and causing a momentary negative thrust that resulted in the third stage dropping away from the postboost vehicle.

The system for Minuteman III involved six circular-shaped charges on the forward dome. Using data from high-speed films and strain gauges, the Minuteman team learned that this arrange­ment worked within 20 microseconds, cutting holes that resulted in a rupture of the pressure vessel within 2 additional milliseconds. But the case developed cracks radiating from the edge of the holes. TRW used a NASTRAN computer code to define propagation of the cracks. It then determined the dome thickness needed to eliminate the failure of the fiberglass. Aerojet wound “doilies" integrally into the dome of the motor case under each of the circular charges. This eliminated the rupturing, allowing the system to vent the pressure in the chamber and produce momentary negative thrust.74

Minuteman Ills achieved their initial operational capability in June 1970, the first squadron of the upgraded missiles turned over to an operational wing at Minot AFB, North Dakota, in January 1971. By July 1975, there were 450 Minuteman Ils and 550 Minute – man Ills deployed at Strategic Air Command bases.75

Even before it began service as a missile, the Atlas had started to function as a launch vehicle. In December 1958, an entire Atlas (less two jettisoned booster engines) went into orbit carrying a re­peater satellite in Project Score. Then, simultaneously with their role in Project Mercury, modified Atlas missiles served as space – launch vehicles for both the air force and NASA in a variety of missions. The basic Atlas was standardized, uprated, lengthened, and otherwise modified in a variety of configurations, often indi­vidually tailored for specific missions. Engineers mated the vehicle with a number of different upper stages, of which the Agena and Centaur were the best known and most important. In these various configurations, Atlas space boosters launched satellites and space­craft for such programs as Samos, Midas, Ranger, Mariner, Pioneer, International Telecommunications Satellite Consortium (Intelsat), the Fleet Satellite Communications System (FLTSATCOM), the Defense Meteorological Satellite Program, and the Navstar Global Positioning System. Following the end of the period covered in this book, some Atlases even used strap-on solid motors to supplement their thrust at liftoff.33

After initial failures of three Atlas-Ables in 1959-60, Atlas-Agena 62 had a number of problems but became a successful launch combi – Chapter 2 nation. From February 26, 1960, until June 27, 1978, Atlas-Agenas flew approximately 110 missions, many of them classified. Mean­while, in 1962 NASA urged the air force to upgrade the Atlas D basic launch vehicle to a standardized launch configuration known as Space Launch Vehicle 3 (SLV-3), which was much more reliable than the Atlas D (96 versus 81 percent successful). A further up­grade after 1965 known as SLV-3A featured longer tanks, allowing heavier payloads in conjunction with other modifications. Because of the classified nature of many Agena missions, precise and reliable

statistics are not available, but by May 1979, on Thor, Atlas, and Titan boosters, Agena had proved itself to be a workhorse of space, achieving a reported success rate of higher than 93 percent.34

With the exception of Agena, most of the upper stages used with Atlas were derivatives from other programs. The Centaur, however, was a derivative, in a sense, of Atlas in that it used the steel-balloon tank structure envisioned by Charlie Bossart and developed for the Atlas missile. Adapting that structure to the liquid-hydrogen fuel used on the Centaur proved to be a major challenge, however. It required a lot of engineering changes when problems occurred, a major reorganization of the way Centaur was managed, and a great deal of testing. But after initial delays, it worked well.35

If Agena was the workhorse of space, Centaur was the Clydes­dale. Its powerful engines enabled it to carry heavier payloads into orbit or farther into space than Agena could manage. The Centaur could do this because it burned liquid hydrogen as well as liquid oxygen. Hydrogen offered more thrust per pound of fuel burned per second than any other chemical propellant then available—about 35 to 40 percent more than RP-1 (kerosene) when burned with liq­uid oxygen.36

This added performance allowed various versions of Atlas – Centaur to support such NASA missions as landing on the lunar surface in the Surveyor project and orbiting High-Energy Astron­omy Observatories, as well as placing 35 communications satel­lites into orbit through 1989. As with other upper stages flying on Atlas vehicles, not all of the Centaur missions were successful, but most were.37

The intellectual push for Centaur came from Convair Division of General Dynamics engineer Krafft Ehricke, who had worked for von Braun at Peenemunde and Huntsville and for Bell Aircraft be­fore moving to Convair. When General Dynamics managers asked him to design an upper stage for Atlas, he and some other engineers, including Bossart, decided that liquid hydrogen and liquid oxygen were the propellants they needed. Aware to some degree that liquid hydrogen’s very low density, extremely cold boiling point (-423°F), low surface tension, and wide range of flammability made it unusu­ally difficult to work with, Ehricke faced funding limitations under an air force contract that precluded performing as many tests as the propellant required—an important restriction on normal rocket­engineering practice.38

This, among other issues, prevented Convair engineers from dis­covering problems occasioned by liquid hydrogen’s unique properties as early as they otherwise might have done, necessitating redesign.

Other problems arose with Centaur engines, designed by Pratt & Whitney Division of United Aircraft Corporation. The extreme cold of liquid hydrogen required completely new design features, includ­ing the use of aluminum coated with Teflon in place of rubber gas­kets to seal pipe joints. Despite such problems plus burnthroughs of the combustion chamber that necessitated redesigns, Pratt & Whit­ney engineers conducted a successful engine run in September 1959, less than a year from the date of the initial contracts with their com­pany and Convair.39

However, explosions in engines in late 1960-early 1961 revealed other problems. One of these required an adjustment to the method of feeding the hydrogen to the combustion chamber. Because of such difficulties and resultant delays, an Atlas-Centaur did not launch on a test flight until May 8, 1962, 15 months later than planned. At the point of maximum dynamic pressure, 54.7 seconds into the launch, an explosion occurred as the liquid-hydrogen tank split open. Engineers did not discover the real cause of the problem until five years later, but meanwhile the delays and problems resulted in a complete reorganization of the Centaur program to provide better control and coordination. Funding also improved.40

Solutions to further problems and programmatic changes fol­lowed, but finally, on May 30, 1966, an Atlas-Centaur successfully launched Surveyor 1 to the Moon on the first operational Atlas – Centaur flight. Atlas-Centaur performed satisfactorily on all of the Surveyor launches, although two of the spacecraft had problems. But five of the seven missions were successful, providing more than 87,000 photographs and much scientific information valuable both for Apollo landings and for lunar studies. On Surveyors 5-7 the At­lases used longer tanks with greater propellant volumes and pay­load capacity than the earlier versions. With the longer tanks, the weight of payload that the Atlas-Centaur combination could place in 300-nautical-mile orbit rose from 8,500 pounds on the shorter version to 9,100 pounds.41

64 The longer-tank Atlas (SLV-3C) and the original Centaur (known Chapter 2 as Centaur D) launched on March 2, 1972, with a Delta third-stage solid-propellant motor, the Thiokol TE-M-364-4 (Star 37E), on the spectacular Pioneer 10 mission that was NASA’s first to the outer planets and the first to reach escape velocity from the solar system. Well before this launch, NASA, which had taken over the program from the air force, had decided to upgrade the Centaur with an im­proved guidance/control computer. The new computer allowed General Dynamics to simplify the Atlas to the SLV-3D configura­tion by removing the autopilot, programming, and telemetry units

from the earlier, long-tank SLV-3C and having the Centaur perform those functions. The new Centaur had two configurations, the D-1A for use with Atlas and the D-1T for use with Titan space-launch vehicles. The differences between the two configurations involved details of external insulation, payload-fairing diameter, battery ca­pacity, and the like.42

The first use of the Centaur D-1A and SLV-3D was on the launch of Pioneer 11, which had the same mission as Pioneer 10 plus making detailed observations of Saturn and its rings. As on Pioneer 10, the mission also employed the Star 37 motor in a third stage. Launched on April 5, 1973, Pioneer 11 returned much data about Saturn, in­cluding discoveries of Saturn’s 11th moon and two new rings. Be­tween 1973 and May 19, 1983, 32 SLV-3Ds launched with Centaur

FIG. 2.8

Launch of a Titan-Centaur vehicle from Cape Canaveral Air Force Station, Febru­ary 11, 1974. The two solid – rocket motors and the core stages of the Titan appear below the Centaur upper stage. (Photo courtesy of NASA)

D-1A upper stages. With the first launch of the Intelsat У with more relay capacity (and weight) on December 6, 1980, the Centaur be­gan to use engines that were adjusted to increase their thrust. Of the 66 total 32 SLV-3D/D-1A (and the slightly modified D-1AR) launches, Chapter 2 only 2 failed. This marked a 93.75 percent success rate, with no fail­ures caused by the Centaur stage.43

During the early 1980s, General Dynamics and Pratt & Whit­ney converted to new versions of Atlas and Centaur. The Atlas G was 81 inches longer than the SLV-3D because of additions to the lengths of the propellant tanks. It developed 438,000 pounds of thrust. Pratt & Whitney made several changes to the Centaur. The first Atlas G-Centaur launched on June 9, 1984, attempting to place an Intelsat У into orbit. It did so, but the orbit was not the intended

one and was unusable for communications purposes. After a modi­fication to fix the problem, there were four successes and one failure (caused by a lightning strike). Then, on September 25, 1989, an At­las G-Centaur launched the 5,100-pound FLTSATCOM F-8 satellite into geosynchronous transfer orbit. This was the last in a series of such navy ultra-high-frequency satellites, part of a worldwide com­munications system for the DoD.44

Meanwhile, forces had been building for commercializing launch- vehicle services. The air force had become unhappy with the idea, promoted by NASA, that all DoD payloads should be transported on the Space Shuttles instead of expendable launch vehicles. There was already competition from the Ariane launch vehicle in Europe, with prospects that other countries would sell launch-vehicle ser­vices to communications-satellite purveyors and other users. On January 28, 1986, the explosion of the shuttle Challenger grounded the remaining shuttles for more than two years. Early in 1987, General Dynamics announced that it would sell Atlas-Centaur as a commercial launch vehicle. NASA then signed a commercial con­tract with the company. General Dynamics decided to designate the commercial vehicles with Roman numerals, the first being Atlas I. All would have Centaur upper stages. On July 25, 1990, the first Atlas I successfully launched the joint NASA/Air Force Combined Release and Radiation Effects Satellite into a highly elliptical geo­synchronous transfer orbit.45

Through this launch, the Centaur had had a 95 percent success rate on 76 flights. This included 42 successes in a row for Centaur D-1 and D-1A between 1971 and 1984. The Centaur had led to the use of liquid-hydrogen technology on both upper stages of the Saturn launch vehicle and in the space shuttle main engines (SSMEs). It had thus made major contributions to U. S. launch-vehicle technology.46

Partway through the history of Centaur and the various Atlas models used to launch it, the air force contracted with General Dy­namics, beginning on February 14, 1966, to modify Atlas Es and Fs that had been in storage since their decommissioning as missiles in 1965. The process began with the newer F models. Rocketdyne in­spected each of the MA-3 engines and fixed or replaced any part that failed to meet specifications. In 1969, the rocket division started a more extensive program of refurbishment to ensure that the engines in storage would work when called upon. After two launch failures in 1980-81, Rocketdyne rebuilt the engines at its plant, performing static tests before installing them on a launch vehicle.47

Six Atlas Ds and four Fs joined forces in launching Orbiting Vehicle One (OV-1) spacecraft, beginning with a failed launch on

January 21, 1965, by a D model and ending on August 6, 1971, with the successful launch of OV-1 20 and OV-1 21 by an F model. A number of the Atlas launch vehicles carried multiple OV-1 satel­lites, each of which included an FW-4S solid-propellant rocket motor built by the Chemical Systems Division (CSD) of United Technologies Corporation, the organization that also provided the solid-rocket motors (SRMs) for the Titan III. Although the satellite failed to orbit for a variety of causes on four of the OV-1 launches, the air force’s Aerospace Research Support Program placed 117 space experiments in orbit to study a variety of phenomena.48

An Atlas F successfully launched a radar calibration target and a radiation research payload for the air force’s Space Test Program on October 2, 1972, using the Burner II solid-propellant upper stage that usually paired with the Thor booster. Another solid-propellant upper stage that operated only once with an Atlas E or F was the Pay­load Transfer System (PTS), which used the same basic TE-M-364-4 Thiokol motor as the Stage Vehicle System (SVS), employed mul­tiple times (as a different upper-stage system from PTS) with Atlas Fs and Es. On July 13, 1974, an Atlas F and the PTS successfully launched Navigation Technology Satellite (NTS) 1 to test the first atomic clocks placed in space to confirm their design and opera­tion and provide information about signal propagation to confirm predictions for the Navstar Global Positioning System (GPS). GPS was then in development and destined to become a vital naviga­tional aid, far more sophisticated and accurate than anything that preceded it.49

SVS, built by Fairchild Space and Electronics Company of Ger­mantown, Maryland, used two TE-M-364-4 motors in two upper stages to place NTS-2 and six Navigation Development System (NDS) spacecraft into orbit between June 23, 1977, and April 26, 1980. The NDS-7 launch failed on December 18, 1981, when the Atlas E launch vehicle went out of control. The other seven satel­lites all supported the development of GPS.50

68 The air force used a different upper stage, known as SGS-II, to – Chapter 2 gether with the Atlas E to launch NDS-8 through NDS-11 between July 14, 1983, and October 8, 1985, all four launches being success­ful. McDonnell Douglas Astronautics Company made the upper stage, using two Thiokol TE-M-711-8 (Star 48) motors, also featured on the Payload Assist Module (PAM), which the Space Shuttle and Delta launch vehicle had employed since 1980. Thiokol began de­veloping the motor in 1976. It used the same hydroxy-terminated polybutadiene (HTPB)-based propellant as Thiokol’s Antares III

rocket motor, a third-generation, third-stage propulsion unit for the Scout launch vehicle.51

The Atlas Es and Fs used other upper stages to launch satellites, including one Agena D. On June 26, 1978, an Atlas F—modified to mate with the Agena and to carry the Seasat-A oceanographic satellite—placed its payload into orbit. The other major upper stage used by the Atlas Es and Fs was the Integrated Spacecraft System (ISS), with a Thiokol TE-M-364-15 motor (Star 37S). In 1977-78, this was the latest in the Star 37 series of motors, also used as an upper stage on the Thor for launching weather satellites. Beginning with a launch of Tiros N from an Atlas F on October 13, 1978, the ISS served as an upper stage for launching the NOAA-6 through NOAA-11 polar orbiting meteorological satellites plus a number of DMSP satellites. The only failure in the series was NOAA-B on May 29, 1980.52

In February 1983, the air force began operating a derivative of the SLV-3D known as the Atlas H. It used most of the basic systems on the SLV but employed GE radio-inertial guidance. The particular solid-propellant upper stage used with the Atlas H and previous At­las Es and Fs to launch the White Cloud Naval Ocean Surveillance System (NOSS) satellites was classified. The White Cloud NOSS satellites provided the DoD (primarily the navy) with ocean surveil­lance. Overall, the Atlas E and F launch vehicles had only 4 failures in 41 launches by the end of 1990, yielding a success rate of more than 90 percent. All 5 launches with the Atlas H were successful.53

Conceived as a missile, the Atlas became a successful and ver­satile launch vehicle, mated with a great variety of upper stages. Featuring a controversial but “brilliant, innovative, and yet simple" concept (the steel-balloon tank design), both the Atlas and the Cen­taur proved to be flexible and effective. With commercialization, the Atlas and the Centaur continued to provide launch-vehicle ser­vices beyond the period of this book and into the 21st century. The Centaur proved to be especially difficult to develop because of the peculiar properties of liquid hydrogen. But it was also hampered by initial funding arrangements and other avoidable problems. As with many rocket programs, engineers found that the existing fund of knowledge was inadequate to predict all of the problems that would occur in developing and launching an extraordinarily com­plex machine. Unforeseen problems continued into the 1990s, and engineers had to relearn the lesson that continual and sophisticated testing was the price of success, even if it did not always preclude unanticipated failures.54

An important step forward occurred with the third liquid-propellant stage for the Titan III, known as Transtage, for which the air force decided on a pressure-fed engine that would use the same nitrogen tetroxide as oxidizer and Aerozine 50 for fuel as stages one and two. As planned, it would have two gimballed thrust chambers, each pro­ducing 8,000 pounds of thrust, and a capability of up to three starts over a six-hour period. Aerojet won this contract, with a Phase I agreement signed in early 1962 and a Phase II (development) award issued on January 14, 1963.51

Aerojet designed the Transtage engine (designated AJ10-138) at about the same time as a larger propulsion unit for the Apollo service module. The two engines used basically the same design, featuring the same propellants, ablatively cooled thrust chambers, and a radiatively cooled nozzle assembly. Since the Apollo service module’s engine bore the designation AJ10-137, its development ap­parently began earlier in 1962, but it also lasted longer. Although

Aerojet designed and built them both, and more information is available about the development of the spacecraft engine, it is not clear that any of the latter’s problems and solutions are relevant to the Transtage engine, which was less than half as powerful and roughly half the length and diameter of its sibling.52

Apparently, these two engines were not the only ones with ab­latively cooled combustion chambers in this period, because an important NASA publication on liquid-propellant rocket engines issued in 1967 stated that such “thrust chambers have many ad­vantages for upper-stage applications. They are designed to meet accumulated duration requirements varying from a few seconds to many minutes." Although construction could vary, in one ex­ample (unspecified), the ablative liner used a high-silica fabric im­pregnated with phenolic resin and then tape-wrapped on a mandrel. Asbestos impregnated with phenolic served as an insulator on the outer surface of the liner. A strong outer shell consisted of layers of one-directional glass cloth to provide longitudinal strength. Cir­cumferential glass filaments “bonded with epoxy resin" provided “hoop strength."53 This appears to have described the Transtage combustion chamber (as well as others?).54

On July 23, 1963, Aerojet had successfully operated a Transtage engine for 4 minutes, 44 seconds, considered “a long duration fir­ing." During that static test, the engine started and stopped three times, demonstrating the restart capability. However, a more criti­cal test of this crucial capability (which would allow it to place multiple satellites into different orbits on a single launch or to po­sition a single satellite in a final orbit, such as a geostationary or­bit, without a need for a separate kick motor) would occur in the simulated-altitude test chamber at Arnold Engineering Develop­ment Center in Tullahoma, Tennessee. In August 1963, tests at that center confirmed suspicions from the July 23 test that the combus­tion chamber would burn through before completing a full-duration firing (undefined). In addition, gimballing of the engine in a cold environment revealed a malfunction of a bipropellant valve (that fed propellants to the combustion chamber) and a weakness in the nozzle extension, made of aluminide-coated columbium and radia­tion cooled with an expansion ratio of 40:1. Information about how Aerojet solved these problems is not available in any of the sources for this book, with the official history of the Titan III merely stating that “by the close of 1963, an extensive redesign and testing pro­gram was underway to eliminate these difficulties so the contractor could make his first delivery of flight engine hardware—due in mid – December 1963." 55

One Aerojet source does not comment on these particular diffi­culties but does refer to “the error of trying to develop in a produc­tion atmosphere." The source explained in this connection that de­velopment of this small engine occurred while Titan I was starting into production, causing management and engineers/technicians to pay less attention to it. But presumably, the speed required in Transtage’s development was also a factor in these particular prob­lems. Obviously, engineers had not expected them and had to adjust designs to correct the difficulties. In any event, engine deliveries did not occur in mid-December, as initially planned, but started in April 1964.

Aerojet engineer and manager Ray Stiff recalled that after engine deliveries began, the air force started to impose new requirements. Because Transtage needed to perform a 6.5-hour coast while in orbit and then be capable of “a variety of firing, coast, and refire combi­nations," there had to be “unique insulation requirements," to pro­tect propellants from freezing in the extreme cold of orbit in space, especially when shaded from the sun. But this insulation retained the heat from combustion, which built up around the injector with presumed dire consequences for continued performance. Stiff does not reveal how Aerojet solved this problem, stating only that the engine’s injector was “baffled for assurance of stable combustion."56

Other sources reveal that the injector used an “all-aluminum flat 170 faced design" with a “concave spherical face, [and] multiple-orifice Chapter 4 impinging patterns." The baffle was fuel cooled, so perhaps an ad­justment in this feature solved the heating problem. According to an Aerojet history written by former employees and managers, “The injector design has undergone two performance upgrade programs which resulted in the very high specific impulse value of 320 lbf-sec/ lbm, and the design has been carried over into later versions of the Delta."57 (Most sources do not rate the specific impulse this high.)

In any event, the two initial Transtage engines each yielded 8,000 pounds (lbf) of thrust with a specific impulse of more than 300 lbf-sec/lbm. Pressurized by cold helium gas, each of the hyper – golic propellants was stored in tanks of a titanium alloy that the prime contractor, Martin, machined in its Baltimore Division. The titanium forgings came from the Ladish Company of Cudahy, Wis­consin. Although titanium was difficult to machine, it was gaining increasing use for liquid-propellant tanks. With a fuel tank about 4 feet in diameter by 13.5 feet in length and an oxidizer tank mea­suring about 5 by 11 feet, Transtage’s propellant containers were hardly huge but were reportedly some of the largest yet produced from titanium. Each overall engine was 6.8 feet long with its diam-

eter ranging from 25.2 to 48.2 inches. Its rated burning time was a robust 500 seconds, and its total weight was only 238 pounds.58

Transtage advanced storable-propellant technology but also rep­resented a further example of trial-and-error engineering. Other up­per stages used the technology developed for Transtage and for the Apollo service module’s engine.