Rocketdyne Engine Family Tree

From on page 15 of the Development of the Rocket Engine for the Jupiter Missile.
Dated 18 Dec 1958.

Rocketdyne Rocket Engines

North American Aviation was the nation's largest manufacturer of military aircraft during World War II, but the end of the war meant a substantial decrease in demand for such planes. Management decided that cruise and guided missiles like the German V-1 and V-2 would provide additional opportunities for the company, and experimented with their own small rocket engine designs.

In 1946, NAA's Aerophysics Lab received two German V-2/A4 engines, which they disassembled and studied. They built three copies of this engine, using American standards and fabrication techniques. These engines were never test-fired, although they were flow-tested with water.

With its large, spherical combustion chamber, tangle of LOX feed lines, and 18 burner cups, this engine was a nightmare to fabricate. NAA redesigned the engine, using a cylindrical combustion chamber and a flat-faced injector. They retained the V-2's 75% ethyl alcohol fuel mixture, hydrogen peroxide steam generator to drive the turbopumps, and double-walled steel thrust chamber with a conical, 15-degree expansion angle. With engine design still in its infancy, the V-2's steering jet vanes were also retained, rather than to produce a gimballed engine. This engine produced 75,000 pounds of thrust and was dubbed the XLR43-NA-1 or simply the 75K.

NAA planned to use this engine for the rocket booster for the Air Force's MX-770 Navaho, but had to abandon this course as the cruise missile's weight increased. Around then, the Army was looking for an engine for the Redstone missile under development at the time, and this engine met the specifications. While the engine itself produced 75,000 lbs of thrust, the engine is sometimes listed has having 78,000 lbs; this extra 3,000 lbs corresponds to the exhaust from the turbine exhaust.

The Redstone proved to be a reliable and versatile booster; it was fitted with two solid rocket upper stages to test scale models of the reentry vehicle to be used by the next-generation Jupiter missile. This version of the Redstone was called the Jupiter-C. Although the diagram above lists the Jupiter-C's fuel as alcohol, most other sources cite higher-energy Hydyne fuel. A fourth solid upper stage was subsequently added to the Jupiter-C, to be called the Juno I, and was used as America's first satellite launcher (still using Hydyne fuel), launching Explorer I on January 31, 1958. Reverting back to the less-toxic alcohol, the Redstone was also used in the manned Mercury program.

To meet the increased requirements of the Navaho, NAA engineers (who were now referred to as the Missile and Control Equipment, or MACE, division) developed a new engine with 120,000 pounds of thrust. The double-walled, steel thrust chambers of the previous generation of engines were durable, but heavy. Heat transfer through the relatively thick walls was poor; several rows of holes were drilled through the inner thrust chamber wall to provide film cooling to prevent burn-throughs.

The 120K engine's thrust chamber was made out of regenerative cooling tubes. These tubes were lighter and thinner than the steel walls, which increased the heat transfer, allowing both the elimination of the film cooling and increasing performance by allowing the use of 92.5% alcohol for fuel. The thrust chamber was still largely the same size and shape as the XLR43-NA-1.

The 120K also replaced the XLR43-NA-1's hydrogen peroxide steam generator with a bipropellant gas generator, burning the same propellants as the main engine, discarding the separate peroxide tank. The 120K also introduced a geared turbopump, the Mark 3 turbopump, which would go on to see service not only with the other Navaho engines, but also with the Atlas, Jupiter, Thor, and H-1 engines.

Two of these 120K engines were packaged together to form the G26 Navaho engine.

The Navaho program continued to demand larger engines, so the 120K was uprated to 135K, and three of them were packaged together; the resulting G38 engine system produced 405,000 pounds of thrust (the source of the 415,000 lbs in the diagram above is unclear, but the extra 10,000 lbs are presumably attributable to turbine exhaust, as with the 75K/78K engine). The thrust chambers in the G38 were hinge-mounted, allowing the deletion of the jet vanes.

While the Navaho never entered production and its engines weren't used by any future program, the 120K engine did beget several lines of rocket engines.

An early design of the Atlas intercontinental ballistic missile called for 840,000 lb of thrust, which could be met by using seven of the 120K engines; even at this point, the Atlas was a stage-and-a-half design, with five of the engines to be jettisoned after the initial boost phase. The Air Force wanted to switch from alcohol to JP-4, a kerosene-based fuel. While JP-4 worked in jet engines, it had a very broad specification which was too variable for reliable rocket engine performance. NAA agreed to switch to another kerosene fuel, RP-1, which had a tighter specification and which has become a standard rocket engine propellant.

The weight of the warhead to be delivered by the Atlas shrunk as the Atomic Energy Commission was able to produce more compact devices. The final Atlas configuration required only 360,000 lbs of thrust. With minor improvements and the change to RP-1, the 120K engine was now producing 150,000 pounds of thrust. Two of these 150K engines were packaged together for booster engines, and a new, 60,000 lb engine was developed for use as a sustainer.

Before the Atlas became operational, a need was identified for an intermediate-range ballistic missile. The Air Force began development of the Thor and the Army developed the Jupiter. Both missiles had similar capabilities and enjoyed relatively quick design cycles, benefitting from ICBM developments. The engines for these two missiles were also very similar, although there seems to be some confusion as to exactly how similar they are. Some sources claim that the Jupiter and Thor engines were "similar in many respects," differing primarily in the more precise thrust control system of the Jupiter engine and the use of ground-mounted start tanks on the Jupiter rather than the engine-mounted Thor tanks (which were also used for vernier engine operation). At least one source states that the same thrust chamber was used in Atlas, Jupiter, and Thor engines. Other sources enumerate additional differences.

There doesn't appear to be much agreement as to what to call the Thor engine, either (although, to be fair, the Thor had a long life and was upgraded several times). Although the occasional source calls it an S-3D (i.e., the same as the Jupiter engine) and the Air Force refers to both the Jupiter and Thor engines as the LR-79, I've also seen the Thor engine referred to as simply the S-3, the S-3E, and various engines in the MB series (most commonly the MB-3). But, eventually, the Jupiter and Thor engines led to an engine named the S-3D.

Although Rocketdyne was busy making engines for the Atlas, Jupiter, and Thor, it also recognized the value of pure research and development. A group of engineers began experimenting with a number of concepts, including the simplification of rocket engine subsystems, in an engine program named X-1.

The X-1 program experimentally introduced a number of concepts which would become standard in later engines:

• The use of a hypergolic igniter (wherein hypergolic fluid, delivered through the injector, provides spontaneous ignition).
• Earlier rocket engines had start tank systems to bootstrap the turbopumps, prior to gas generator operation; the X-1 introduced a solid-propellant gas generator to provide the initial turbine operation.
• The X-1 replaced the electrically-controlled pneumatic system for actuating valves with valves actuated by fuel: as the turbopump came up to speed, the buildup of fuel pressure was used to open propellant valves. This was referred to as a "pressure ladder sequence", and ensured that valves were actuated only when the engine was ready.
• Rather than using oil stored in a separate tank to lubricate the turbopump, the X-1 used RP-1, blended with a small amount of lubricant additive, for this purpose.

When the Army Ballistic Missile Agency went shopping for an engine to power their Juno V (soon to be renamed Saturn) rocket, Rocketdyne offered the H-1 engine, based on the S-3D but incorporating many of the improvements from the X-1. The vastly-simplified H-1 would also mount its turbopump on the thrust chamber (rather than on the missile body, as with previous engines). This allowed the propellant high-pressure ducts to be very short and rigid, replacing the longer ducts which had to provide enough flexibility to allow for engine gimaballing.

The bulk of the information on this page (other than the family tree diagram itself and a comparison of the Thor and Jupiter engines) was taken from Rocketdyne: Powering Humans into Space.