A History of American Rocket Engine Development

It is quite common to associate rockets with cutting edge technology. But oddly enough until the last decade or so, the rocket engines used by most of America’s launch vehicles have actually been not much more than incrementally modernized versions of power plants originally developed over half a century ago. Engineers tend to be conservative by nature and when they find designs that work, they stay with them. This philosophy is especially valuable when complex, high performance machines, like modern liquid propellant rocket engines, are involved.

American rocket pioneer Robert Goddard working on a stainless steel rocket engine in 1915. (NASA)

Since the first flight of a liquid propellant rocket in 1926 by American rocket pioneer Robert Goddard, rocket engine development had been based as much on trial and error as on the application of engineering principles. Throughout the prewar years, individuals and small groups of rocket enthusiasts in America and Europe endlessly tinkered with rocket engine design concepts slowly improving on their performance and reliability. By the late 1930s, however, rocket technology had developed to the point where it became of interest to the military especially in Germany.

The ultimate achievement of this government-funded development program was the A-4 rocket more widely known as the V-2. Capable of hurling a ton of high explosives over a range of 320 kilometers, the V-2 was a quantum leap in rocket technology. While the V-2 could not reverse Germany’s failing fortunes in the closing months of World War II, it would become the grandfather of the rockets that literally propelled us into the Space Age.

The Early Years

Before captured V-2 rockets were sent to the US for study after World War II, American rocket engines were relatively modest in size generating no more than several kilonewtons of thrust. The turbopump-fed rocket engine in the V-2 burned alcohol and LOX to produce an unprecedented 265 kilonewtons of thrust. Captured V-2 hardware would serve as the starting point for all subsequent large rocket engines developed not only in the US but in the Soviet Union as well.

From left to right, the rocket engines for the V-2, Redstone and Navaho booster. (Boeing)

Companies like General Electric and Reaction Motors were among the first to exploit V-2 technology to build engines for early military research rockets like Hermes and Viking (see “America’s First Space Rocket: The Origin and First Flights of the Viking Rocket“). Another important innovator in the development of large rocket engines was North American Aviation. By 1951, North American working with the US Army had built the 350-kilonewton A-1 engine to power the first tactical ballistic missile deployed by the US – the Redstone (see “Redstone: The Missile That Launched America into Space“). While the Redstone and its engine were highly improved versions of the V-2 built to American design standards, even larger engines were needed to power the longer range missiles the military had on their drawing boards. To handle the development of these new power plants, North American formed the Rocketdyne Division in 1955 (which, following a series of corporate sales and mergers over the past decade, is now part of Aerojet Rocketdyne).

Rocketdyne’s first leap beyond the V-2-class rocket engine was the XLR89. Built from lessons learned with the V-2 and Redstone, a trio of these engines were used in the rocket booster of the Navaho supersonic cruise missile. Producing 600 kilonewtons each, the XLR89 was the most powerful engine of its time and would serve as the basis of a family of rocket engines used to this day.

The Rocketdyne engines developed for (left to right) the Redstone, Thor and Jupiter rockets. While there are major differences in the turbomachinery of the latter two, their thrust chambers are almost identical. (Rocketdyne)

The first rockets developed in the 1950s to benefit from Rocketdyne’s developments were the USAF Thor and US Army Jupiter IRBMs as well as the USAF Atlas ICBM. Instead of using alcohol like earlier large engines, these new engines burned RP-1 grade of kerosene in combination with LOX. The Thor and Jupiter used 668-kilonewton engines of similar design called MB-3 and S-3D, respectively. Both missiles would eventually serve as the first stages of some early American launch vehicles starting in 1958.

Diagram of the Atlas’ innovative MA-3 propulsion system with its pair of jettisonable outboard booster engines and a central sustainer. (Rocketdyne)

While Rocketdyne was busy studying larger rocket engines, no single power plant was available in the 1950s with the thrust needed for an ICBM. Drawing on its experience with Navaho, Rocketdyne came up with an innovative engine cluster concept for the Atlas yielding the required 1,600-kilonewton liftoff thrust. Designated MA-3, it consisted of a pair of LR89 booster engines similar to Rocketdyne’s IRBM engines producing 667 kilonewtons each. These would be jettisoned shortly after liftoff when they were no longer needed. A scaled down design optimized for high altitude operation called the LR105 served as the sustainer engine. Producing 267 kilonewtons, the sustainer was ignited on the launch pad with the pair of LR89 boosters and would propel the now lightened rocket using the remaining propellants after the booster engines were jettisoned (see “The First Atlas Test Flights“).

A pair of Aerojet LR-87 engines were employed in the Titan II and later the Titan III and IV family of space launch vehicles for about a half a century. (USAF)

Rocketdyne was not the only American company developing large rocket engines at this time. As a backup to the innovative Atlas and its MA-3 power plant, the USAF approved the development of the Titan ICBM with a more conventional two-stage design. Aerojet-General (which is now also part of Aerojet Rocketdyne after a long series of corporate mergers), which had built smaller rocket engines for sounding rockets and the upper stages of some early launch vehicles, got the contract to build the engines for Titan. The Titan II, unlike its predecessors, used storable, hypergolic propellants that ignite on contact. The twin chambers of the LR-87 in the first stage produced a total of 1,900 kilonewtons at liftoff. The single chamber LR-91, which was adapted from the LR-87 design for high altitude use, powered Titan’s second stage and produced 445 kilonewtons at altitude. The Atlas and Titan II would both be used as launch vehicles including in the Mercury and Gemini manned spacecraft programs.

The Golden Age

While the Atlas was obsolete as a weapon almost as soon as it was deployed in 1960, it proved to be an excellent satellite launch vehicle when used with upper stages. Improvements to the Atlas included the use of the more powerful MA-5 propulsion system producing 1,869 kilonewtons at liftoff. The Thor was also modified to better serve as a launch vehicle including in its most enduring role as the first stage of the Delta starting in 1960. Like all Thor-based launch vehicles, the Delta would continue to use the MB-3 for the next 14 years.

The Rocketdyne H-1 engine which was used on America’s first heavy lift launch vehicle, the Saturn I, and later the Saturn IB. (Rocketdyne)

Although the Jupiter was phased out as a weapon and a launch vehicle by the early 1960s, its S-3D engine was substantially simplified by Rocketdyne engineers to produce the H-1 in 1959 (see “Juno V: The Birth of the Saturn Rocket Family“). A cluster of eight H-1 engines, eventually producing a total of 7,300 kilonewtons after a long series of design improvements, were used on the Saturn I and IB which launched the first Apollo missions into orbit. Later Rocketdyne’s experience was used to manufacture the most powerful rocket engine flown by the US: the F-1. Producing 6,675 kilonewtons each, a cluster of five F-1 engines was used in the first stage of the Saturn V which first flew in 1967 (see “Apollo 4: The First Flight of the Saturn V“).

A Rocketdyne F-1 engine on display at KSC. (A.J. LePage)

While more powerful first stage engines can translate into larger payloads, the performance of the upper stage engines can have an even greater impact. The ultimate propellant combination employed today is liquid hydrogen and LOX which has almost twice the punch by mass of most conventional propellants. The first engine to use this powerful cryogenic combination was the RL-10 built by Pratt & Whitney (which, after a merger with Rocketdyne in 2005, is now also part of Aerojet Rocketdyne). Based on a design successfully tested in 1957 at Lewis Laboratory (now NASA’s Glenn Research Center), a pair of RL-10 engines producing about 67 kilonewtons each were used in the high-performance Centaur stage. First successfully flown with the Atlas in 1963, the Centaur had more than double the performance of any other upper stage for missions into high Earth orbit or beyond.

A cutaway model of the Pratt & Whitney RL-10 engine.  Originally employed on the Centaur, more advanced versions are still in use on several launch vehicles after more than half a century. (A.J. LePage)

A cluster of six RL-10 engines was also employed by NASA in the second stage of the Saturn I starting in 1964 (see “The Coolest Rocket Ever“). Despite their performance, the RL-10 was too small for practical use in larger rockets. NASA gave Rocketdyne a contract to build the more powerful 890-kilonewton J-2 engine. A single J-2 was used in the improved second stage of the Saturn IB first flown in 1966 (see “The First Flight of the Apollo-Saturn IB“). The same stage was also used as the third stage of the Saturn V which also employed a cluster of five J-2 engines in its second stage. The use of the high-performance J-2 in the upper stages of the Saturn V helped to give this rocket its unprecedented payload capability.

A cluster of five Rocketdyne J-2 engines being prepared for a static firing of the Saturn V second stage. (Rocketdyne)

The ultimate achievement in cryogenic rocket engine development during this era was Rocketdyne’s Space Shuttle Main Engine or SSME (today known as the RS-25). With development starting in 1971, the SSME operated at higher temperatures and pressures than any other engine in order to make it the most efficient engine flown. Capable of being throttled and producing 2,278 kilonewtons each, a trio of these engines were used by the now-retired Space Shuttle. While a pair of recoverable solid rocket boosters did most of the work to get the Shuttle off the launch pad, the efficient SSMEs provided most of the energy to actually achieve orbit.

Test firing of the Space Shuttle’s SSME. (MSFC/NASA)

The Lull

Rocket engine development in the United States has typically been tied to the development of specific rockets to use them. With a stable of proven launch vehicle designs established by the mid-1960s, space planners were more inclined to make incremental improvements to boost performance rather than design new rockets from scratch. In many Thor-based launch vehicles, especially the Delta, larger clusters of increasingly more powerful solid rocket boosters coupled with stretched upper stages were used to incrementally improve performance.

Launch of a Titan IIIC which used a pair of 120-inch solid rockets to boost the performance of the basic Titan II. (USAF)

The use of solid boosters was taken to the extreme in the Titan rocket family. In 1965, a pair of 120-inch solid rocket boosters were attached to a modified Titan II core to produce the first Titan IIIC quadrupling its orbital payload capacity in the process (see “The First Missions of the Titan IIIC“). Over the next three decades, the Titan core was stretched and longer boosters attached to slowly increase performance in successive versions of the Titan III and later the Titan IV. Even the Centaur was adapted for use with the Titan IIIE and eventually the Titan IV to further improve its payload capability.

The Rocketdyne RS-27, which incorporated a number of improvements first employed on the H-1, replaced the MB-3 on the Delta family of launch vehicles in 1974. (Rocketdyne)

But during the 1970s and 1980s only minor changes were made to the liquid rocket engines used by the Atlas, Delta, and Titan to improve performance, reliability, and manufacturability. In 1974 Rocketdyne introduced the RS-27 to replace the aging MB-3 in the Delta. Incorporating components of the venerable MB-3 and the H-1 designs, the RS-27 was only a modernized version of the same basic design used for two decades. But with plans for the Space Shuttle to replace all expendable launch vehicles (ELVs) in the 1980s, there was little need seen to develop new engines.

Test firing of the MA-5A propulsion system used in modernized versions of the Atlas starting in 1991.

After the Challenger accident in 1986, the folly of relying totally on the Shuttle for launch services was finally realized. As a result, the US started developing more evolved designs of existing launch vehicles along with the rocket engines to power them. The Delta II and III (the former of which is still flown today) use the improved RS-27A engine. Starting in 1991, the Atlas II used the upgraded MA-5A where the original YLR89 boosters of the MA-5 were replaced with a pair of RS-27 thrust chambers. The Russian engine manufacturer Energomash working with Pratt & Whitney developed the RD-180 (derived from the Russian RD-170 that powers the Zenit and the retired Energia boosters) to replace the MA-5A in the larger Atlas III introduced in 2000. Burning kerosene and LOX and built with a throttle capability, the two-chamber RD-180 can produce up to 3,800 kilonewtons. This same engine is still used in the Atlas V and has been the source of increasing controversy in recent years.

Test firing of the RS-68 engine currently used in the first stage of the Delta IV. (NASA)

The first totally new large liquid propellant rocket engine developed since the SSME was Rocketdyne’s RS-68. Burning liquid hydrogen and LOX to produce up to 3,314 kilonewtons, this engine was designed from the start to minimize manufacturing costs. The RS-68 has 93% fewer parts than the SSME and has been advertised to provide more thrust per dollar than any other rocket engine. The RS-68 has been used in the first stage of the Delta IV launch vehicle family since its introduction in 2003 (see “The Largest Launch Vehicles Through History“).

Since the Centaur had a role even in the Shuttle-era, Pratt & Whitney continued upgrading the RL-10 to meet new demands over the years. The most recent version of the Centaur flown as an upper stage with the Atlas V typically uses a single RL-10A4-2 engine. A newer version, the RL-10B-2 producing 110 kilonewtons of thrust, was first used in the second stage of the now-retired Delta III introduced in 1998 and is currently employed in the second stage of the Delta IV. The RL-10 was also modified for other roles as well: Four throttleable, 65-kilonewton RL-10A-5 engines were used on the DC-X experimental rocket in the mid-1990s.

Test firing of a Merlin 1D engine used on the SpaceX Falcon launch vehicle. (Space Exploration Technologies)

While there have been a number of rocket engines developed and tested over the past couple of decades for various heavy-lift launch vehicles concepts, the trend has been to employ evolutionary upgrades of older engines (like the RS-25/SSME for the SLS) or inexpensive Russian-built engines in launch vehicles such as the Atlas V, mentioned earlier, and Orbital Science’s Antares. Between this trend and a string of corporate mergers during the last decade that combined the original three major American rocket engine manufacturers into a singe company, Aerojet Rocketdyne, as well as a single major manufacturer of large launch vehicles, United Launch Alliance (ULA), the development of new engines and launch vehicles hoped for a couple of decades ago was slow to materialize. Continued uncertainty and lack of sufficient funding for the proper development of heavy-lift rockets after the retirement of the Space Shuttle has only compounded the problem. This has left the United States dangerously dependent on Russia at a time when relations with that country are becoming increasingly strained. The only major exceptions have been the development of the inexpensive Merlin family of engines by SpaceX used in their affordable Falcon launch vehicle family and Blue Origin’s BE-series of engines being considered for future launch vehicles. Only time will tell if American rocket engine development, which was once the pride of this country, will be revived.

Follow Drew Ex Machina on Facebook.

Related Reading

“The Largest Launch Vehicles Through History”, Drew Ex Machina, February 19, 2018 [Post]

General References

David Baker, The Rocket: The History and Development of Rocket & Missile Technology, Crown Publishing, 1978

David Baker (editor), Jane’s Space Directory 2001-2002, Jane’s Information Group, 2001

J.D. Hunley, U.S. Space-Launch Vehicle Technology: Viking to Space Shuttle, University Press of Florida, 2008

J.D. Hunley, Preludes to U.S. Space-Launch Vehicle Technology: Goddard Rockets to Minuteman III, University Press of Florida, 2008

Werhner von Braun and Frederick I. Ordway, History of Rocketry and Space Travel, Thomas Y. Crowell Co., 1966