Sutton 2003
Uploaded by
Amjad Ali PashaSutton 2003
Uploaded by
Amjad Ali PashaJOURNAL OF PROPULSION AND POWER
Vol. 19, No. 6, November–December 2003
History of Liquid Propellant Rocket Engines in the United States
George P. Sutton
Los Angeles, California 90049
I. Introduction An estimated300–350 differentLPREs have been designed,built,
T he liquid propellant rocket engine (LPRE) is a proven means and static tested in the United States. Because of space limitations,
of propulsion.It was conceived over 100 years ago, but its rst only a few of them have been selected for this summary paper. For
actual construction in the United States (and in the world) was ac- each of those, only a few pieces of data or a gure will be given here.
complished by an American, namely, Robert H. Goddard, in 1921 If a signi cant LPRE or outstandingaccomplishmentwas omitted, it
(82 years ago). His rst static hot- ring test was in 1923 and the was not by intent, but by the lack of information available to the au-
historic rst ight with a LPRE occurred in 1926. Today this tech- thor and/or the space limitation for this paper. Although some of the
nology is suf ciently well developedand proven that we can design, ight vehicles driven by a LPRE (airplanes, missiles, or spacecraft)
build, and y with con dence any kind of LPRE. In 1940, there were are mentioned or identi ed here brie y, the emphasis in this work
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
only a few outstanding individuals and groups that were struggling is on the rocket engines and not on the rocket vehicles themselves.
with early research and development efforts. The LPRE capability Gaseous propellant engine systems are included because they are
has proliferated and grown, and today there are several active U.S. usually grouped with the LPREs. We will not cover solid propellant
companies and several government laboratories that have a mature rocket motors, electrical propulsion,hybrid propulsion,and combi-
broad LPRE technical base. nation rocket-airbreathingengines.
The reader shouldhave an understandingof LPREs, a background
or exposureto some aspectof the subject.For basic informationrefer II. Need for LPREs
to the general Refs. 1–8 and, for more detail, to a future book by the Why were LPREs used? Because they propelled certain military
author on a world wide history of LPREs, scheduled to be published and space vehicles better than any other type of chemical propulsion
by the AIAA in late 2004. and because they provided some operating characteristicsthat could
There is no single LPRE concept or type, but rather several that not be duplicated at the time by any other means of propulsion.2
are related and tailored to speci c applications. All have one or LPREs made it possible to build sounding rockets (1926–1960);
more thrust chambers (TCs), a feed system for providingthe propel- they propelledmilitary aircraftand assisted with their takeoff(1942–
lants under pressure to the thrust chamber(s), and a control system. 1970). They went into production for several early tactical missiles
There are signi cant differences between LPREs with high thrust (1951–1973) because solid propellant rocket motors could not meet
and low thrust, cryogenic vs storable propellants, monopropellants the operating temperature limit requirements during the 1940s and
or bipropellants,single use or reusable, one run per ight vs multi- 1950s. LPREs were selected for all the initial ballistic missiles,
ple restarts during ight, random variable thrust or nearly constant helping to build up the military missile inventory needed urgently
thrust, and those with pumps or gas pressure expulsion of propel- by the U.S. Government in the 1950s–1970s. Since 1960, LPREs
lants in their feed systems. The history of each of these types will be propelled all of the large space launch vehicles and just about all
discussed. the U.S. spacecraft and satellites. They constitute the propulsion
In this paper a “successful LPRE” is de ned as one that 1) has machinery that drove us into the space age.
been put into productionand/or 2) has own its mission satisfactorily The features and performance characteristics of LPREs that al-
more than once.After all, the ultimate objectiveis to propela vehicle. lowed their selection for the mentioned missions, were unique and
There have been many LPREs, engine components,and propellants, are brie y reviewed next.2 Liquid bipropellants generally give a
but for various reasons they were never successful,and most fell by higher speci c impulse than other chemical propulsion means, such
the wayside. Yet we learned some important lessons from them. as monopropellantsor those using solid or hybrid propellants.Cryo-
We will concentrate on some of the successful LPREs, but we will genic propellants give the highest speci c impulse. LPREs can be
also discuss some others that have interesting technologyor historic designed over a very wide range of thrust values to t speci c ap-
signi cance. plications (by a factor of 108 /. They are the only form of chemical
George P. Sutton has been active in the design, research, development, testing, teaching, installation and man-
agement of rocket propulsion since 1943 and was personally involved in several early historic liquid propellant
rocket engines and solid propellant rocket motors programs. In the aerospace industry he worked for three years
at Aerojet Engineering Company and for more than 25 years at Rocketdyne (now a part of The Boeing Company),
where he held several positions, including Executive Director of Engineering and Director of Long Range Planning.
His book Rocket Propulsion Elements (currently in its 7th edition) is the classical text on this subject, has been
translated into three other languages, and is used by more than 40 colleges world wide. First published in 1949 it
has been in print longer than any other aerospace text. For 11 years he was a member of the U.S. Air Force Scienti c
Advisory Board. In academia he was the Hunsaker Professor of Aeronautical Engineering at the Massachusetts
Institute of Technology MIT and has served on the faculty of the California Institute of Technology. He has worked
for the U.S. Government as Chief Scientist of the Department of Defense Advanced Research Projects Agency,
where he started major programs, and as a project leader at the Lawrence Livermore National Laboratory. For
a few years he worked in the commercial world and has been on the board of directors of two industrial private
companies. He is an AIAA fellow, an author of 50 technical articles, the recipient of several professional society
awards, and is listed in Who’s Who in America.
Received 11 March 2002; revision received 3 September 2003; accepted for publication 4 September 2003. Copyright ° c 2003 by the American Institute of
Aeronautics and Astronautics, Inc. All rights reserved. Copies of this paper may be made for personal or internal use, on condition that the copier pay the $10.00
per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923; include the code 0748-4658/03 $10.00 in correspondence
with the CCC.
978
SUTTON 979
propulsion that can be designed for quick restart, fast pulsing, and to 350 psi with a pump feed system in 1939. In the 1940sgas pressur-
ready reuse. They can be designed for a random thrust variation on ized feed systems allowed increases to 500 psi. With pumped feed
command. They have been uniquely suitable for controlling quick systems, these chamber pressures reached 1000 psi in the 1950s.
attitude (pitch, yaw, or roll) changes and minor velocity changes of There were some exceptions. Some small experimental TCs were
individual stages of missiles, space launch vehicles, spacecraft, and tested at more than 5000 psi in the 1970s. The highest chamber
satellites. A precise repeatable thrust termination permits an accu- pressure of a ying U.S. engine was 3319 psi in block I of the space
rate terminal ight velocity. LPREs can be functionallychecked out shuttle main engine (SSME), whose development started in 1972.
and even fully tested before they are used. An engine-out capability The higher pressure allows a higher nozzle area ratio (without ow
can be designed into engine clusters. A remarkably high reliability separation at sea level), which gives further performance increases.
has been achieved in production LPREs. Lightweight LPREs have Higher chamber pressures also allow the TC to be smaller, which
allowed ight vehicles to achieve a high propellant fraction and a makes it easier to place into a vehicle. There were some disadvan-
high vehicle mass ratio. Instant readiness has been achieved with tages, which prevented going to even higher values. Because heat
storable propellants.These propellantshave been stored for 20 years transferincreasesapproximatelylinearlywith the chamber pressure,
in a vehicle. cooling of TCs becomes much more dif cult at higher pressuresand
All common propellants used today can discharge a very clean the amount of gas ow or energy needed to drive the turbines in-
transparent exhaust gas without smoke. Gas from certain of the creases. Also the engines become heavier.
storable propellants can give a trace of smoke, but their particulates
do not usually form a noticeable deposit on sensitive vehicle sur- C. So Many Liquid Propellants
faces, such as windows. The exhaust gas of today’s LPREs is not An estimated 170 differentliquid propellantshave undergonelab-
toxic and environmentally friendly.
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
oratory evaluations and many also small TC tests.8¡10 This number
III. Technology Trends and Changes
Throughout this LPRE history, one can discern some technical
trends, growth patterns, or directions for improvement. Some are
brie y listed later. There are several other trends, but their write-up
had to be omitted from this paper. They were thrust vector control,
injectors, gas generators or preburners, major reductions of inert
engine mass, extending or uprating a family of existing engines,
and reducing lifetime costs. They are planned to be included in the
upcoming book version.
A. Expanding the Range of the Thrust, 0.01–1,800,000
Pounds Force (Refs. 1–4)
The thrust magnitudeis dictatedby the application.The rst thrust
chambers (Goddard 1921–1924) had between 40- and 100-lbf thrust
a) Thrust chamber with propellant tanks
and were intended for small sounding rockets. Some of his early mounted on wall
bipropellantTCs (roughly 1.2 in. diameter) are shown in Fig. 1. His-
torically the thrust levels went both up and down. By 1944, a series
of hydrogen peroxide monopropellantthrust chambers (for reaction
control) became available with thrust values as low as 0.1 lbf. With
inert or with warm gas as the propellant, the thrust levels went even
lower. It took 45 years to increase of the thrust to 1,800,000 lbf as
seen in Table 1.
The highesttakeoffthrust was with ve F-1 enginesat 7:5 £ 106 lb
for the S-IC booster stage of the Saturn V space launch vehicle b) Thrust chamber with two adjustment valves
(SLV). There have been no application requirements for higher
thrusts since about 1969. The F-1A did not y, and the program
was not continued.
B. Increasing the Chamber Pressure
The historical trend has been to raise the chamber pressure. This
makes it is possible to increase speci c impulse between 4 and 10%.
The exact values depend on the speci c design, chamber pressure, c) Injector has one hole each for fuel and oxidizer
nozzle area ratio, and application.Goddard started (1920s) with rel- Fig. 1 Small thrust chambers that were some of the earliest designed,
ativelylow chamberpressure,typically50–100 psi, but later went up built and tested by Goddard; some with ceramic inserts (from Ref. 23).
Table 1 Historical Increases in Thrust Level of U.S. LPREs
Thrust, lbf
Ground tests No. of TCs per TC (maximum) Application Developer
1923–1925 1 40–100 Experimental Goddard
1927–1940 1 150–1000 Sounding rockets Goddard
1942 1 1,500 Experimental RMI
1943 3 2,000 PB2Y-3 JATO Aerojet
1949 1 16,000 Hermes 3 missile General Electric
1950 1 75,000 Redstone missile Rocketdyne
1953 2 120,000 Navaho G-26 booster Rocketdyne
1955 2 150,000 Atlas Missile booster Rocketdyne
1960 2 210,000 Titan II booster Aerojet
1963 5 1,500,000 F-1/Saturn V booster Rocketdyne
1968 1 1,800,000 F-1A, Experimental Rocketdyne
980 SUTTON
does not include any minor changes in the propellant formulation
(such as small changes in NO2 percentage in nitric acid) and in the
propellant additives, such as gelling agents, corrosion inhibitors, or
stabilizers.1¡3 More than 25 different propellant combinationshave
been own in U.S. LPREs.
Early U.S. efforts were with liquid oxygen (LOX) gasoline
(Goddard1923),nitric acid/aniline[GuggenhiemAeronauticalLab-
oratory (GALCIT) and Aerojet 1940–1955], and LOX/75% alcohol
[Reaction Motors, Inc. (RMI) 1942]. There was no discernibletrend
in the early propellantselectionsand each project team, company,or
governmentagency picked the propellantthey thoughtmost suitable
for their application. Therefore, U.S. LPREs were developed, pro-
duced, and own with a variety of propellant combinations, such
as those just mentioned. Flying LPREs have also used ammonia,
90% hydrogen peroxide, nitric acid, kerosene, mixtures of nitro-
gen oxide and nitric acid, 75 and 92% alcohol, pure hydrazine,
mixtures of kerosene and unsymmetrical dimethylhydrazine
(UDMH), monomethylhydrazine (MMH), gelled propellants, and
others.
A yearning for higher energy denser propellants led to programs
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
of investigating uorine, chlorine, and/or uorine containing com-
pounds, boron materials, and several other exotic materials or mix-
tures in the 1960s and 1970s. The effort was extensive and included
ground ring of TCs and three complete engines with some of
these toxic, corrosive, and highly ammable high-energy propel-
lants. None were selected for a ying engine. Fluorine as an oxidizer
can give a higher speci c impulse and higher average densities than
LOX, but it and other high-energy chemicals were not considered
to be practical. The decision to stop these high-energy propellant
investigations was due in part to the potential drastic consequences Fig. 2 Rocketdyne J-2 LPRE used LOX/LH2 (courtesy of The Boeing
of a major engine failure and/or a spill of propellantsand their effect Co., Rocketdyne Propulsion and Power).
on people, equipment, or environment.
All of this effort did not lead to a universally acceptable single
liquid propellant combination. All selections were a compromise D. Large Liquid Propellant Rocket Engines
between good qualities (high performance, high density, easy start, A large LPRE consists of one or more TCs, usually one or more
low cost, or stable, long time storage) and bad qualities (corrosive, turbopumps to feed the propellants from the propellant tanks to the
ammable, toxic, prone to storage decay, high vapor pressures, or TCs, a source of medium-hot gas to drive the turbine(s), a control
combustioninstability)and dependedon the application.After years system that will include commanding the start and shutoff, provi-
of operationalexperience ve propellantcombinationsseem to have sions for lling or drainingpropellants,variouspipes and valves,and
emerged as being practical to use with current space applications, means for applying a small pressure to the propellanttanks. In addi-
and they are listed in Table 2. Not mentioned in this table are pro- tion, some engines have features to enable throttlingor restart,thrust
pellants for applications that are today obsolete, such as jet-assisted vector control, and self-monitoring of certain pressures, tempera-
take offs (JATOs), sounding rockets, aircraft propulsion, or tactical tures, or performance parameters. There are many ways in which
missiles with LPREs. these components have been designed to t together and meet the
High speci c impulse is very signi cant in space missions, where requirements of different missions.1¡4;6;7 There are four types of
the cumulative mission ight velocity is high. Here even a small in- feed systems to supply the propellants to the LPRE.
creasein speci c impulse leads to major increasesin payloador orbit The pressurizedgas feed system is the oldest ( rst tested in 1923).
height. This has led to the cryogeniccombination of LOX/liquid hy- Here the propellantis expelledfrom its tanks by pressurizedinert gas
drogen (LH2 ) with speci c impulse values between 410 and 467 s or by nonreactive gas created in a gas generator. Engines with this
depending on the design. This practical propellant combination has feed system will be shown subsequently. Most of the applications
been investigated experimentally since 1945. It has been preferred were with small LPREs using multiple TCs and with relativelysmall
for upper stages of SLVs [such as the Pratt and Whitney Aircraft up- total impulses. However, there were some large LPREs using this
per stage RL10 ying since 1963 with a speci c impulse of 466 s and system (for reasons of high reliability and fewer parts), such as the
shown in Sec. V.B.5 or The Boeing Company, Rocket dyne Propul- lunar takeoff engine or the Apollo Service Module engine.
sion and Power (Rocketdyne) J-2 LPRE used in Saturn V shown in The pumped feed systems are preferred with LPRE of high total
Fig. 2]. The extra performance usually overcomes the disadvantage impulse and large thrust. Three different engine cycles have been
of the low density of LH2 , which means very large insulated fuel distinguished, and they are shown in Fig. 3. These cycles refer to
tanks, extra tank weight, and more drag. the method of supplying medium-hot gas to one or more turbines,
the owpaths of the propellants and the method of handling and
discharging the turbine exhaust gas.
Table 2 Practical current U.S. propellants and their applications
The rst engine with pump feed used the GG cycle. It was rst
ground tested by Goddard in 1938 and own in 1940. Normally
LOX/kerosene (RP-1) Some SLV booster stages the same propellants used by the main LPRE are also burned (usu-
LOX/LH2 Some SLV booster stages ally at a fuel-rich mixture) in a small secondary combustion de-
Most SLV upperstages vice called the gas generator (GG) to generate gas at temperatures
NTO/MMH Attitude or reaction control between 700 and 1650± F for driving the turbine. The turbine ex-
systems (for orbit change, haust gas is discharged overboard at a lower velocity than the ex-
reentry, or space rendevous);
haust gas from the TC (shown in Figs. 3 and later). LPREs with
Post boost control systems
Hydrazine monopropellant Some reaction control systems this cycle usually have the lowest cost and often the lowest inert
NTO/50% hydrazine C 50% UDMH Older SLV and missiles mass, but the performance is 2–7% lower than with the other two
cycles.
SUTTON 981
a) b) c)
Fig. 3 Simpli ed diagrams of three common engine cycles. The spirals are a symbol for the hydraulic resistance of an axisymmetric cooling jacket,
where heat is absorbed by the cooling uid. From Ref. 2.
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
The expander cycle relies on the evaporation and heating of a
cryogenic propellant. This propellant (usually LH2 ) is heated and
gasi ed in the TC’s cooling jacket, and it is then expanded to drive
the turbine. There is no GG. The turbine exhaust gas is injected into
the combustion chamber, where it is burned ef ciently with all of
the oxidizer at high pressure and high temperature.It was rst tested
in 1960, was developed by Pratt and Whitney and is explained in
Figs. 3 and Sec. V.B.5. The rst ight was in 1963.
The staged combustion cycle has two combustion chambers in
Fig. 4 Wyld’s simpli ed sketch of rst U.S. fully regeneratively cooled
series. The rst, called the preburner, burns all of the fuel and some TC (1938), (from Astronautics, a Journal of the American Rocket
of the oxidizer to give medium-hot gases at very high pressure, and Society, July 1938).
these gases drive the turbine. The turbine exhaust goes through the
injector into the main combustion chamber, where it is burned with
the bulk of the oxidizer and expanded ef ciently with the rest of the
propellant; all of the reaction products are discharged through the erative comes from the fact that heat absorbed by the coolant (in
nozzle. This cycle was rst tested in the United States by Pratt and the cooling jacket) is not lost, but is regenerated and augments the
Whitney in the 1960s (with a simulated turbine and without turbop- heat of combustion.1;2;12 Because the design did not have a feature
ump) and with a complete engine by Rocketdyne around 1974. The to allow thermal axial growth, the inner wall yielded or wrinkled
rst and only US engine with this cycle to y (1981) was the SSME, and may have cracked on cooldown after several rings. Improved
described in Sec. V.B.3 and shown subsequently. versions of this thrust chamber design were used by RMI, Aero-
jet, and other U.S. contractors for about 10 years. Improvements
included higher cooling velocities in the throat region, where the
1. TC, Heart of the Large LPRE heat transfer was the highest, and better provisions, such as expan-
The TC consists of a combustion chamber, where the propellants sion joints, for compensatingfor the thermal growth of the hot inner
are burnedand form hot gas, a supersonicnozzle,where the gases are wall.
expanded and accelerated,and an injector, where the propellantsare The thermal stresses in the inner wall can be greatly reduced if
introducedto the chamber, broken up into small droplets,and evenly the inner wall is thin. This led to the tubular cooling jacket idea, an
distributed in ow and mixture ratio over the cross section of the advancedform of regenerativecooling. Three U.S. companies RMI,
chamber.1¡3 Because the gas temperature is usually twice the melt- Aerojet, and Rocketdyne implemented experimental TCs using a
ing point of steel, the heat transfer and the coolingof the TC structure bundle of attened and shaped tubes for the coolant passages during
have historically been critical issues, and they are discussed next. the late 1940s. RMI started studies in 1947, built an experimental
Originally the early investigators (1920s–1940s) used uncooled tubular TC in the 1950s, and used a single-pass concept of nickel
TCs. The duration was limited to a few seconds by the heat- tubes in the TC of the XLR-99 (ground tested around 1959) for the
absorbingcapacityof the wall, beforethe wall material would locally X-15 research aircraft ( ew 1960). In 1948 Aerojet built a small TC
melt or readily oxidize. Uncooled TCs are still used for experiments, with shaped welded and soldered aluminum tubes, which may have
short-duration applications, and with special materials as well as in been the very rst oneeverbuilt. Aerojet rst applieda tubulardesign
some small TCs. Historically lm cooling was the rst method of in the Titan I TCs, which were designed in 1955. Rocketdyne tested
extending the duration to more than a minute.11 It was invented by small tubes around 1949, and in 1950 began to design large TCs that
Goddard around 1925. He called it “curtain cooling,” where most of used a brazed together bundle of thin-walled stainless-steel tubes,
the fuel was injected through slots or openings located in a circular which had been double tapered and shaped to the nozzle/chamber
pattern at the outer diameter of the injector. It caused a reduction of contour. Rocketdyne rst tested it (about 1953) with the large TC
performance (2–20%) due to incomplete mixing of the two propel- (120,000-lbf thrust) of the G-26 Navaho engine, and it was own
lants. Film cooling is still used today in some large TCs, but as a rst in 1956. Figure 5 shows such a TC and its at plate impinging
supplementarymethod to augment other cooling methods at critical stream injector. The fuel ows from the inlet manifold through every
locations of high heat transfer rates. other tube down to the fuel return manifold. It then ows up through
Regenerativecooling was rst demonstratedin 1938 in the United alternate tubes (between the down ow tubes) and through openings
States by James H. Wyld, an amateur rocketeer and one of the into the injector. The three sketches on the right show how the tube
founders of RMI. His drawing is shown in Fig. 4. Here one of the cross sections change with chamber/nozzle diameter. The pressure
propellants(usually the fuel) is circulatedthrough the cooling jacket forces were taken by steel hoops brazed to the outside of the tubes.
around the chamber and nozzle and absorbs heat. The name regen- Brazing was done by coating or supplying those surfaces, which are
982 SUTTON
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Fig. 5 Early version of large TC at 120,000-lb thrust with tubes, which are formed, shaped, and then brazed together (from Ref. 2).
to be joined, with brazing material, holding the tubes and hoops in a Several types of supersonicdivergingnozzle exits have been used
xture and heating this assembly in a special reducing atmosphere on TCs.1;2;13¡15 Some are compared in Fig. 7 together with their
furnace. Tubular cooling jackets have been used successfully in ow patterns at sea level and at high altitude. The earliest versions
many large engines. (1921–1936) by Goddard and other pioneers had a straight, long,
Another successfuladvancement of regenerativecooling for very conicaldivergingnozzlesection with a small half-angleof 4 or 5 deg.
high heat uxes was developed in the 1960s. It uses straight milled (Fig. 1 and shown subsequently). Analysis done in the late 1930 at
channels (of variable width) machined into a forged or cast metal GALCIT and other organizationsshowed that a shorter nozzle with
piece with the shape of a nozzle throat region. The outer wall can a half-angle around 15 deg was best. This nozzle exit cone angle
be brazed or electroformed to the milled center piece. The milled was used between 1938 and 1957 in all types of rocket propulsion
channel design was, for example, used by Rocketdyne in throat including solid propellant types as shown, for example, in Fig. 5.
region of the TC of the SSME and by Aerojet in the TC of the Between 1956 and 1958, several peoplein my section at Rocketdyne
orbital maneuver engine (6000 lbf) of the Space Shuttle Orbiter, proposed and investigated a bell-shaped nozzle contour for the di-
which is shown in Fig. 6. Its injectorhasresonancecavitiesexplained verging nozzle section. Two are shown in Fig. 7. This contour was
later. derived by analysis;its shape is close to a parabola.1;2;13;14 It was val-
Regenerative cooling is not suitable for large TCs that have deep idated by tests at Rocketdyne of two different opposing nozzles on a
throttling. At very low-thrust level (or low ow) the coolant would pendulum in 1956/1959 and by full-scale ring tests. The bell shape
boil, and the mixture ratio would change. However, an ablative liner gives a little more speci c impulse (reduces divergence losses) for
in the chamber and nozzle (without a cooling jacket) has been sat- the same nozzle length as an equivalent 15-deg cone. Alternatively,
isfactory. Such an ablative material absorbs heat by evaporation it can be made shorter and still have good performance. This con-
and chemical cracking/decomposition of the material; the resulting tour has been used since about 1960 in all rocket propulsionnozzles,
gases seep out of the material and form a relatively cool boundary large or small, liquid or solid propellant. Some of the large engines
layer, which gives a reduction of the heat ow. Later, an example is (Thor or Atlas) that ew originally with a straight 15-deg cone noz-
shown. zle were then modi ed to the new bell-shaped contour. The lower
SUTTON 983
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Fig. 6 Simpli ed partial section of one of the two TCs of the orbiting maneuvering system (OMS) used on the space shuttle vehicle (courtesy Aerojet).
Fig. 7 Several supersonic nozzle exit sections with different lengths and ow patterns (partly from Ref. 2).
portion of the diverging nozzle exit segment has a relativelylow heat ground tested at Pratt and Whitney in the 1960s, and the rst ight
transfer and does not require a regenerative cooling jacket. Instead, was in 2000 on the Pratt and Whitney RL 10 B2 LPRE, shown
three lower cost single-wall uncooled designs for nozzle extensions later; its movable nozzle exit skirt was made from carbon bers
have been used and own. An ablative material was used by Aero- with a carbon ller. The third approach for a nozzle exit is called
jet in the 1950s for upper-stage engines and in 1962 for the Titan dump cooling, and it has a single wall with an inside boundary layer
sustainer engine and in 1998 on the Rocketdyne RS-68 engine for of medium-hot turbine exhaust gas (700–900± F), which is dumped
an SLV. This has usually been the lowest cost approach. Alterna- through a manifold and slots into this lower portion of the diverging
tively, a radiation-cooledthin wall made of niobium or carbon ber nozzle. It was used on the Rocketdyne F-1 engine (designed 1959)
material have been used effectively for various upper-stage engines and is shown later.
since about 1960 (Fig. 6). A special version is the extendible nozzle, Unique special types of nozzles were developed, namely, the
which is stored around an upper stage engine during ascent through aerospike nozzle and the expansion/de ection (E/D) nozzle, shown
the atmosphere and then extended or moved into position at alti- in Fig. 7.2 Their main merit is to expand the exhaust gases at op-
tude before engine start. The rst extendible nozzle of a LPRE was timum value at all altitudes. (The effective area ratio changes with
984 SUTTON
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Fig. 8 Technician is holding two of the turbopumps developed by Goddard (from Ref. 24).
altitude.) This allows a slightly higher time-averaged speci c im-
pulse or thrust during ight, when compared to a xed conven-
tional nozzle. The engine length can be very short with a cutoff
spike or a shortened E/D nozzle, saving some inert vehicle mass
and drag. Five different experimental aerospike engines have been
ground tested by Rocketdyne (using LOX/LH2) with thrusts be-
tween 50,000 and 400,000 lbf includinga linear aerospikeversion.15
Rocketdyne also ground tested two versions of an E/D engines [us-
ing nitrogen tetroxide (NTO)/Aerozine 50] in the early 1960s at
50,000- and 10,000-lbf levels. None of these projects with special
nozzles have been continued.
2. Turbopumps1;2;16;17
The turbopump (TP) is a key component for a pump-fed LPRE
and an engineering intensive, high-precision, high-speed piece of
rotating machinery. The rst turbine-drivencentrifugal pumps were
tested by Goddard in 1934 (Sec. IV.A). Two of his TPs are shown in
Fig. 8. They had ball bearings, shrouded turbine blades, and pump
outlet diffusers.(One is shown extendingfrom the pump in the upper Fig. 9 Sectioned view of the type of turbopump with a gear case used
right.) A small arc of the blades of the two turbines were immersed in the Rocketdyne Thor, Jupiter, Atlas booster, and H-1 engines (from
in the exhaust gas of the main nozzle (over 5000± F), and he experi- Ref. 2).
enced frequent turbine failures. Therefore,Goddard developeda GG
(1938) that had a lower gas temperature. The rst version had three
propellants (LOX, gasoline, and water as a diluting/cooling uid).
The TPs were small (low ow) and inef cient. The rst LPRE with a The steel alloy turbine was usually driven by a GG, which used the
TP and a GG was tested by Goddard in 1939 and own in a sounding same propellants as the main TC (but usually at a fuel rich mixture
rocket in 1940. In 1942 he used a fuel-rich GG without water. ratio), resulting in a gas between 1300 and 1650± F. A gear case was
The early TPs for JATO and aircraft superperformance (1943– also used on the Pratt and Whitney family of RL-10 engines to allow
1950)had a turbineand both the propellantpumps on the same single the oxidizer pump to rotate slower than the fuel pump.
shaft. The rst large U.S. TP (Redstone engine designed 1949) had An inducer impeller ahead of and on the same shaft as the main
two in-line shafts, a coupling, and an aluminum turbine because pump impellerwas used duringWorld War II in the TP of the German
this was a proven German technology on the V-2 engine. The GG Walter aircraft rocket engine. It provided for better cavitation resis-
at that time used monopropellant 80% hydrogen peroxide with gas tance of the main pump impellers, and it allowed the tank pressure
temperatures of about 700± F. to be lowered, resulting in a weight reduction of the propellant tank.
Historically the large U.S. TPs of the 1950s and 1960s used a gear It can also allow the main pump to run at a higher speed, which in
case that allowed a turbine to rotate at a higher speed than one or turn allows a reduction of inert TP weight. The United States was
both of the propellantpumps because this allowed better turbine and late in adopting this clever innovation.Several of the U.S. LPRE TP
pump ef ciencies. This resulted in a lower GG ow and a slightly that were already in productionwere changed in the 1950s to use re-
better engine performance than a single-shaft TP. Figure 9 shows a designed pumps, which included inducer impellers. This happened
geared turbopump as used with the Atlas/Thor/ H-1 (Saturn I SLV) to the Thor and Atlas engines. An inducer can be seen in Fig. 9.
family of booster LPREs. Initially oil was supplied from a small oil Goddard’s concept of separate TP assemblies for the fuel and the
pump to lubricate and cool the gears and the bearings, but the oil oxidizer pump was revived several decades later for propellantcom-
was then replaced by kerosene fuel. A gear case was used for the binations where the fuel and the oxidizer have very different densi-
Titan family of LPRES to drive the two pumps at different speeds. ties. It was used with LOX/LH2 engines,such as the J-2 (Rocketdyne
SUTTON 985
1960 design), the SSME (1972 design), and the RS-68 (1997 de-
sign), in part because it gives a smaller and lighter design and if
avoids the complexity of a gear case. Major design advanced were
made in TPs in recent years.2;16;18
E. Small Liquid Propellant Rocket Engines
These small engines with multiple thrust chambers (often called
thrusters) have a very important role for the vehicles’ ight
control.2;3 They were and are still used for trajectory changes, atti-
tude control (pitch, yaw, and roll), deorbit maneuvers, station keep-
ing, rendezvousmaneuvers,or y wheel desaturation,or for settling
liquid propellantsin a zero-gravity ight before main engine restart.
Although a large LPRE is usually assembled and delivered in a sin-
gle package, a small LPRE, with multiple TCs (placed in several
different locations in a vehicle), is normally delivered in several
pieces. Thrust levels are typically between 1 and 100 lbf, but there
were some that were larger (1000 lbf or more). They generally use
storable propellants and a pressurized gas feed system. There were
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
several differentways to obtain a small thrust, and they are explained Fig. 10 Monopropellant hydrazine 0.1-lbf TC assembly with valve to
brie y in their approximate historical sequence. More details are in the left of the mounting ange and an electric heater (courtesy, Aerojet).
Sec. V.
The rst solution for attitude control was the orderly expulsion
of an inert cold gas, such as air or nitrogen, which was stored at
high pressure and exhausted through simple valves, regulators, and hydrazine and 50% UDMH) and a few years later by MMH. These
multiple nozzles. Cold gas for attitude control was used starting in fuels had a lower freezing point, but slightly lower performance.
the late 1940s and continuing sporadically until about 1980. These However, these fuels can, under certain conditions, cause thin un-
systems were simple, low cost, reliable, and ran at ambient tem- desirable deposits of solid particles in the combustion products on
peratures. However the speci c impulse was low (around 70 s) and sensitive vehicle surfaces (windows, solar cells). These deposits
the systems were heavy, adding to the inert mass of the vehicle. have prevented the use of MMH and UDMH in certain satellite
They were used on many early satellites and for roll control on applications.
some upper stages. Several companies have built and own cold-gas With the high gas temperatures, some form of cooling of the TC
thrusters. walls is needed. Regenerativecooling can no longer be used because
In the 1947–1966 period, small monopropellanthydrogen perox- the heat capacity of the low fuel cooling ow would not be adequate
ide thrusters became popular. The relatively low gas temperatures to absorb all of the heat rejected by the hot gas to the inner walls.
(600–1300± F depending on the peroxide concentration)allowed the The cooling fuel would boil, causing a drastic change in mixture
use of simple single-walllow-carbon-steelconstructionand avoided ratio. The thrusters often use some lm cooling, but by ityself, this
the need for a cooling jacket. It was usually decomposed by a silver is not suf cient. One good solution came with a small experimental
screen catalyst and used a pressurizedgas feed system. Thrust levels radiation-cooledthrusters, which were developed in 1958 and 1959
were between 0.1 and 100 lbf. The two suppliers of H2 O2 thrusters by MarquardtCorporation,one of the predecessorsof Aerojet’s Red-
were Walter Kidde & Company (out of business) and Bell Aircraft, mond Center. Its thrust was 25 lbf, and it used NTO/hydrazine and a
which today is Atlantic Research Corporation (ARC). They were molybdenumchambernozzlewith an insidecoatingof molybdenum
used extensively, for example, on the Mercury manned space cap- disilicide for oxidation protection. Molybdenum was soon replaced
sule, and more than 1000 thrusterswere own. Next came hydrazine by niobium (also called columbium), which is lighter and easier to
monopropellant thrusters (1958 to present) with pebble-type cata- fabricate. It has a niobium disilicide inner coating for oxidation pro-
lysts, again a pressurized feed system, and uncooled alloy steel TC tection. A later design Marquardt’s 100-lbf thruster shown in Fig. 11
walls. They offered more than a 50% improvement in performance was rst used for the auxiliary propulsion on the Saturn S IV B up-
over the peroxide.Hydrazinethrusterswere made possibleby the de- per stage (later in other applications), and it ew for the rst time
velopment of a suitable catalyst and by making ultrapure hydrazine, in 1965. Most of these radiation-cooledbipropellant thrusters with
which did not poison the catalyst. The example in Fig. 10 shows niobium chambers have been produced by the predecessorof Aero-
the nozzle exit at the lower right, and the radiation shield hides the jet’s Redmond Center, Northrop Grumman (NG) (formerly TRW),
TC and the catalyst bed. Some of these thrusters could demonstrate or ARC.
more than 100,000 start/stop cycles over a typical ight mission pe- Ablative liners were also an early solution for small thrusterswith
riod. Suppliers were the Rocket Research Corporation (today part many starts. The ablative liner is made of glass, Kevlar® , or carbon
of Aerojet), TRW (today part of Northrop Grumman Corporation) bers woven in a ber cloth in a plasticmatrix, and the cloth is laid in
W. Kidde (no longer in business), and Hamilton Standard (today layers before heating and compressing the material and surrounding
the product is sold by ARC). The advantage of these monopropel- it by a metal shell. Sometimes a ceramic sleeve or a graphite nozzle
lants are the inherentsimplicity of the system (good reliability),high insert is used to minimize erosion. Small ablative type thrusterswere
propellantdensities(small propellanttanks),and clean exhauststhat developedmostly by NG (TRW) and Rocketdynebetween 1960 and
will not fog up sensitive surfaces (window, mirrors, solar cells). Its 1973. Figure 12 shows a 25-lbf thruster (left) and a 100-lbf thruster
principaldemerits are the lower performance compared to bipropel- used on the Gemini manned capsule, its maneuvering system mod-
lants, resulting in a heavier system, and hydrazine’s high freezing ule, or the Apollo command module. They were gradually replaced
point (34± F), requiring heating of all components. Multithruster hy- by radiation-cooledmetal thrustersbecauseablativeswere relatively
drazine monopropellant systems have been used on hundreds of heavy and had dirty exhausts,which have caused unwanted deposits
spacecraft or upper ight vehicle stages and are still popular today. on mirrors or solar cells.
The low-thrust bipropellantthrusters also started in the late 1950s The third type of bipropellant thruster called Interregen was de-
and are still used today on many upper stages, spacecraft, and satel- veloped by Rocketdyne in the late 1960s. It uses a relatively thick
lites. Bipropellants give higher speci c impulses (250–320 s) than wall of beryllium (a low-density, high-conductivity metal) for the
hydrazine monopropellant(210–250 s). Initial propellants were ni- chamber nozzle material. The beryllium conducts the heat away
tric acid (and later NTO) as oxidizer and hydrazine as a fuel. Around from the hot-throat region to a lm-cooled region in the chamber. It
1963, the hydrazinefuel was replacedby Aerozine 50 (a mix of 50% has own in postboost control propulsion systems.
986 SUTTON
Fig. 11 Two views of an R-4D radiation-cooled bipropellant TC (100-lbf thrust in vacuum) with integral valve (courtesy, Aerojet).
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Fig. 12 Section and external view of two ablative thruster of the Gemini manned spacecraft (courtesy, Rocketdyne).
F. Nemesis of Combustion Vibrations The last type of combustion vibration occurs at high frequency
Combustion instabilities have tormented LPRE developers for (above 1500 cps). It has since been linked to the burning process it-
perhaps 30 years beginning about 1950 (Refs. 2 and 19). They have self and to pressure waves and chamber acoustic resonances. When
caused sudden and unexpected failures of TCs and, thus, of LPREs. it did occur, it would cause high-frequency large-amplitude cham-
Therefore,all LPREs must be designedand proven to be free of such ber pressure oscillations, cause sudden increases in heat transfer
instabilities.Three types of vibrations have been identi ed. The rst or the forces exerted by the TC, and lead to a structural or heat
is a low-frequencychugging(10–400 cps) or interactionof the liquid transfer failure of the TC in less than a second of time. Often this
propellant feed system with the oscillating gas in the combustion instability would occur only in one test run out of perhaps 100 or
chamber.This includesoscillationsof propellantsin long feed pipes, 1000 ring tests. Therefore, the only method for assuring a stable
often called POGO instability. Remedies included modi cations design in these early days was to run hundreds of static tests on the
in the feed system, increasing the injection pressure drop, and for same identicalenginedesignwithout a singleincidentof combustion
POGO instability the addition of damping accumulators in the pipe instability.
lines.20 A rating technique was developed between 1957 and 1967. Arti-
The second type of instability is characterized by intermediate- cial disturbances are introduced into the combustion chamber (by
frequency oscillations(400–1500 cps), often called buzzing,associ- setting off speci c directional explosive charges) to induce a pres-
ated with mechanical vibrations and resonances of pieces of the en- sure surge and trigger high-frequency vibrations.21 Accurate high-
gine structures, injector manifolds, pipes, and their interaction with frequency chamber pressure measurements can then determine if
gross combustion behavior, such as turbulence.Frequencies depend there is enough energy absorption for the magnitude of the pressure
on the size and structural resonances. Changes in the chamber ge- oscillations to be damped and diminish rapidly. The recovery time
ometry, injector con guration, and in the structural stiffness of the (in milliseconds) between the arti cial pressure surge and the re-
affected components became effective countermeasures. By about sumption of steady combustion is a rough measure of the inherent
1956, the understanding of the rst two vibration types was good stability. This rating technique reduced the number of static ring
enough to diagnose incidents and take effective remedial actions. tests needed to prove stability.
SUTTON 987
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Fig. 13 Main injector assembly of the SSME showing a six compartment acoustic baf e with protruding coaxial injector elements (copied from
Ref. 2).
Subtle small changes in the geometry of the injection holes, spray
injection elements, or their distribution over the face of the injector
have been effective. Each organizationuses its own experience base
and has its own favorite injector designs. In the past 25 years, the
incidence of combustion vibrations during engine development has
been greatly reduced. The expensive, extensive engine static rings
of the past (to get meaningful statistical data) have been replaced
by analyses of vibration behavior and a few directed bomb tests for
demonstrating stability and for rating the recovery time period.
IV. Early Efforts: 1923–1943
A. Pioneering Work of Robert H. Goddard
Fig. 14 Simple diagram of acoustic energy absorbing cavities at the
periphery of an injector; cavity entrance restriction is a slot in the shape This American physics professor, Robert Hutchinson Goddard
of a section of a circular arc or sometimes a hole. Details of chamber or of Clark University in Worcester, Massachussetts, was a very cre-
injector design are not shown (from Ref. 2). ative researcher and the important developer of rst LPREs in the
world.22 Born in 1882, he died in 1945. He was the rst to develop,
build, and test key LPRE components and static test or re small
Several remedies or xes have been used for avoiding high- LPREs with gas pressurized feed systems between 1921 and 1925.
frequency combustion instabilities. The rst and earliest method Goddard was the rst to launch a sounding rocket with a simple
was to make empirical and somewhat arbitrary changes in the in- LPRE on 16 March 1926 in Auburn, Massachussetts.It is described
jection design, propellant, chamber volume, or chamber shape. For later. His signi cant inventions and contributions23;24 to LPREs are
example, in the Bomarc missile engine, the change from keroseneto summarized as follows.
a UDMH/kerosene fuel mixture solved a high-frequency vibration 1) One of his patents gave the rst plausible drawing and descrip-
problem in 1952. This method could not be scaled up in thrust, was tion of a LPRE with a pump feed system. The drawing is shown in
not fully reliable, and required a lot of expensive testing and time. Fig. 15 (1914).
Next came baf es, which were introduced in about 1958 and proved 2) Between 1921 and 1925, he designed, developed, and ground
to be a reliable remedy for the destructiveradial and circumferential tested the rst liquid propellant thrust chambers. Initial thrust levels
vibration modes. These baf es were built into the injector, as seen were between 40 to 100 lbf for a 1.0 in.-diam chamber (Fig. 1).
in Fig. 13, and cooled baf es were put into most large U.S. engines Later he built 5-, 6-, and 10-in.-diam chambers with up to 1000-lbf
until about the late 1970s. They were retro tted into the engines thrust. He used LOX as the oxidizer,initially with ether as a fuel. He
for the Atlas, Thor, and Titan II and designed into Saturn (H-1, then tested gasoline, alcohol and kerosene with LOX. Early nozzles
F-1), Titan III, Titan IV, and the initial version of the SSME. The were long and had an exit cone half-angle of 4–5 deg; he used larger
third remedy uses acoustic resonance cavities (to absorb gas vibra- nozzle half angle after 1941.
tion energy), was designed into LPREs beginning about 1963, and 3) From 1924 to 1930, using a “curtain ow,” a type of lm cool-
was reliable in preventing many high-frequency oscillations. This ing, he achieved ring duration of more than 1 min without burnout.
is shown in Figs. 6 and 14. They are designed or tuned for a speci c In 1943, he adopted regenerative cooling, which was invented by
vibration frequency, usually the estimated resonance frequency. To- others.
day these resonancefrequenciescan be predictedfor differentmodes 4) He used black powder igniter (1923) and then pioneered py-
of acoustic oscillations.In the United States a number of production rotechnic igniter with solid propellant (1927) and spark plug igniter
injectors, both for high and low thrust, were then redesigned to in- (1937) for restart.
clude resonancecavities. Some injector designs had both baf es and 5) From 1921 to 1923 he developedthe rst propellantfeed system
resonance cavities, such as the injector for the Apollo lunar ascent using high-pressuregas to expel liquid propellants from their tanks
engine. Resonant cavities have been more effective than the baf es, and into a thrust chamber. He later included valves or ori ces to
and in some cases, such as on the SSME, it was possible to later adjust line pressure drops for best mixture ratio and ows.
remove the baf es. It was also learned that certain injector designs 6) The rst ight of an LPRE in the world on 16 March 1926 was
and propellants give stable combustion without baf es or cavities. accomplished by him (described later).
988 SUTTON
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Fig. 15 Goddard’s U.S. patent 1,103,503 issued 14 July 1914 showing a LPRE with a pumped feed system.
7) Between 1925 and 1935, he built and ew rst lightweight Goddard launched sounding rockets with LPREs initially at
propellant tanks and high-pressure gas tanks using welded steel, Auburn, Massachusetts (1926–1930) and later at Roswell, New
often aluminum or sometimes brass sheets. Mexico (1933–1938 and again 1939–1941). Altogether through
8) In 1930, he was rst to use baf es to suppress sloshing in liquid 1941, he conducted hundreds of component tests and static ring
propellants tanks during ight to prevent excursions of the center tests of TCs, over 100 static (bolteddown) tests of an enginemounted
of gravity of the vehicle or to keep gas from entering the propellant in a vehicle, attempted about 50 ight tests, and of these, 31 resulted
pipes. He was rst to reinforce tanks with wound high strength wire in ights.
to reduce tank weight (1937). His early work with solid propellant rocket motors (1914–1920)
9) In 1924, he designed and tested several types of propellant was abandoned in favor of liquid propellant engines because his
pumps: At rst, he tried and abandonedpiston,vane and gear pumps. theoreticalanalysesshowed him that liquidswould give more energy
By 1933, small centrifugalpumps worked. He ground-testedthe rst per unit propellant mass. All of the engines and sounding rocket
TP (1934) with a separate turbine for each propellant pump (Fig. 8). vehicles were built and assembled in his own shop. Every one of
Later some of his pump bearings were running in LOX, which had these ight con gurations had some new features, improvements,
not been done before. or design changes,and he never ew exactly the same vehicle twice.
10) He did static ring of LPRE with the rst TP in 1939. In The vehicle for his historic rst ight on 16 March 1926 is shown
1939/1940, he used a GG to generate “warm gas” that would not schematically in Figure 16. It rose 41 ft above the launch stand and
melt the turbine buckets. In 1938, he developed and thus invented ew a distance of 185 ft in about 2.5 s. It had the thrust chamber
the rst U.S. GG. In 1940, he launched the rst ight of a LPRE at the front of the vehicle and the long propellant tanks (LOX and
with a TP feed system and GG. gasoline) at the aft end. The propellant feed lines also served as the
11) In the 1940s, he conceived and tested a novel ceramic-lined structure to tie the key components together. He used a crude simple
precombustion/ignition chamber (attached to main chamber) suit- cone as a heat shield to protect the tanks from being overheated by
able for restart. the rocket exhaust plume. The black powder igniter was in a tube on
12) During 1925 to 1941, he was rst to develop and y several top of the TC. Ignition of the powder was achieved by a ame from
lightweightvalves, includingsafety valves, propellantvalves, check some broken off match heads inside a copper tube, which in turn
valves, shutoff valves, and throttling valves, and several lightweight was heated externally by some burning cotton, which is not shown.
gas pressure regulators using bellows and springs. The two propellant tanks were both pressurizedby gaseous oxygen,
13) In 1937, he invented and ew vehicles with a “movable tail,” which was evaporatedfrom the oxygentank. The line pressuredrops
a type of gimbal for thrust vector control, actuated by four sets of and the mixture ratio were preset by two small needle valves near
dual pneumatic bellows. The TC was mounted in the tail and had the TC. The lower part of the nozzle had burned off during the last
exible feed lines. part of the rst ight.
14) In 1924, he developed and later improved the rst control Goddard led many patents, 48 were issued during his life time,
system for starting. First controls were manual (strings pulled by 35 more for which he had applied, but were issued after his death in
operatorsat controlstation),then mechanicalsequencersand a clock 1945, and 131 more led by Mrs. Goddard as his executrix after his
as controller. He then developed (1927) pneumatic valve actuation death, based on his notes, sketches and photographs. For example,
and later (1933) a pneumatic LPRE control. in 1914 he obtained a patent on a two-stage vehicle. In 1960, the
15) In 1942, his was the rst JATO of a ying boat on water with U.S. Government bought the rights from his widow to use 200 of
reusable LPRE (800-lbf thrust). these patents for $1 million dollars, and this payment was shared by
16) The rst US variable thrust LPRE was developed and tested her with the Guggenheim Foundation, which had supported most of
by Goddard, but not own (1943). Goddards work between 1930 and 1941.
17) Between 1923 and 1943, he developed and improved tech- From 1942 to 1945, Goddard worked with the U.S. Navy Bureau
niques for photographing gauges, indicator lights, and clock, of Aeronautics at Annapolis, Maryland. There he helped to develop
thus, recording ground-test data. He developed the rst ight a LPRE for JATO, which was successfully ight tested, and the rst
recorder. US variable thrust LPRE, which was quite complex; it was later
SUTTON 989
For his historical and outstandingaccomplishments Goddard has
receivedmany honors, unfortunatelyonly posthumously.For exam-
ple, the NASA Goddard Space Flight Center, at least six schools,
some university professorships,several society awards and lectures
were named after him.
B. Amateur Rocket Societies
A series of voluntary amateur technical groups, intrigued by the
prospects of space ight, sprang up in the United States during the
1930s. They debated space ight issues, published articles and/or
news bulletins, conducted some ground tests of LPREs, and ew
a few simple rockets.26 Their biggest contributions were the pop-
ularizing of space travel and rocketry (lots of free publicity) and
the attracting, educating, and identifying of technical personnel that
later became engaged in the emerging eld of rocketry.
The American Interplanetary Society (AIS) was founded in
March 1930 in New York. It soon grew in membership and started
to publish its own journal. The society designed and static tested
TCs and LPREs at its proving grounds.27 It ew simple and crude
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
rocket vehicles with pressurized gas feed systems. They had many
TC burnouts and ight failures. In 1934, the AIS changed its name
to the American Rocket Society (ARS) because they wanted to get
away from the word interplanetary, which the public and the press
viewed with suspicion and considered a fantasy. ARS had about
15,000 members in about 20 chapters 15 years after it started. I had
the privilegeto serve as a director and then as presidentof the ARS at
a time when the rst actual space ights occurred. Society member
Wyld designed and successfullytested ( rst time in the US in 1938)
a fully regenerative fuel-cooled thrust chamber as shown in Fig. 4.
This was hailed as a major step forward in the technology because
it allowed prolonged rocket operations without burnout failure.
In 1941 the experimental work of the ARS was discontinued, in
part because of the good and well-funded LPRE work being per-
formed elsewhere. The society concentrated on publishing refer-
eed professional papers and holding technical meetings on the sub-
jects of propulsion, space ight and rocket vehicles. About a dozen
other local amateur societies were founded in different parts of the
United States. Other LPRE efforts, including a small U.S. Navy
project,28;29 also sprang up. In February 1963 the ARS merged with
the Institute of Aeronautical Sciences to form AIAA. Today this is
Fig. 16 Diagram of Goddard’s historic rst ying rocket vehicle with a respected professional organization, is still very much concerned
a LPRE, launched on 26 March 1926 in Auburn, Masschussetts (from about LPREs, but rocket propulsion technology is now only one of
Journal of the British Interplanetary Society, Vol. 40, 1987, p. 307). many elds of interest to AIAA.
C. GALCIT: 1935–1943
was fully developed by a contractor. He also became a consultant to This laboratory, originally on the campus of the California In-
the Curtiss–Wright Corporation at Caldwell, New Jersey, and RMI stitute of Technology, was perhaps the rst in the United States
originally at Pompton Plains, New Jersey. to undertake theoretical and experimental work in LPREs. 30 The
Goddard was very reluctant to publish or disclose his concepts, Chairman of the Aeronautical Engineering Department and the
designs, test data, or ight results to other people. His 1919 paper head of GALCIT project was Theodor von Kármán, a reknowned
on “A Method of Reaching Extreme Altitudes” brought him some aerodynamicist.31 This laboratory performed laboratory tests of dif-
fame, but it did not describe his ideas about LPREs. 25 Although ferent propellants, designed and tested small TCs in their own off-
he had correspondence with many people, including other noted campus test facility(beginning1936),and was the rst to achievehy-
rocket experts, such as Hermann Oberth of Germany, he would not pergolic ignition using nitric acid and aniline as propellants (1940).
divulge very much useful information.He was concerned about oth- Different propellantsand thrust chambers with thrusts up to 1000 lb
ers using his concepts before they were fully proven and also about were investigated. In 1937, nitric acid was selected as a good po-
a disclosure of his ideas before the issue of patents. He published tential storable oxidizer. In 1939 GALCIT improved this nitric acid
very little about his work on LPREs during his lifetime. What he oxidizer by dissolving up to 30% nitrogen dioxide (a red colored
did publish had limited distribution.His collected works (including gas, that came out of solution and evaporated as reddish clouds),
diaries, photographs, sketches, and data) were more revealing and and this was henceforth known as red fuming nitric acid (RFNA). It
were published by his widow 25 years after his death (1970). By had greater density, slightly higher performance, and better ignition
that time, the U.S. contractors had reinvented or developed on their properties than nitric acid without the dissolved gas.
own much of what Goddard had previously achieved. It is an ironic They published some historic analysis on rocket propulsion and
twist of history that the LPREs, which were developed by General were the rst to use the concept of a thrust coef cient. They built and
Electric, Rocketdyne, or Aerojet, were designed and produced in ew the rst U.S. JATO32 in 1942 and developed the rst LPRE for a
the 1940s and 1950s without the bene t of the pioneering work sounding rocket. GALCIT was the progenitor of the Jet Propulsion
done by Goddard. He had relatively little impact on the U.S. LPRE Laboratory, which today is still administered by California Insti-
developments. We can only speculate what would have happened, tute of Technology for NASA. Its rst Director, Frank J. Malina
if Goddard would have allowed access to his development results was a key GALCIT member. Several GALCIT members started the
and know-how, while he was still alive in the early 1940s, while the Aerojet Engineering Company in 1942. Its rst chairman was von
LPRE industry was in its infancy. Kármán.
990 SUTTON
V. LPRE Industry in the USA: 1941–2003 12) The Lance TC con guration with the sustainer TC inside the
A. Innovations and Accomplishments annular booster TC is novel and compact, and the use of liquid pro-
A list follows of signi cant and historic U.S. industrial achieve- pellant side injection for TVC is unique. The highest known thrust
ments and key events in this eld. This list is not complete or in any variation of 300 to 1 (from 4400 down to 14 lb) was achieved in the
particular order. Lance sustainer TC assembly. It is discussed further in Sec. V.B.3.
1) The demands for new technologyand missile production were
very high during the cold war with the Soviet Union during the B. Companies in This Business
1950–1970 period, and the U.S. LPRE industry successfully met Since the 1940s at least 14 U.S. companies have engaged in the
these demands and did its share to put missiles into the arsenal. design, development, manufacture, testing, and ight support oper-
2) In 1965, the United States launched Saturn V with the high- ations of some types of LPRE. Table 3 lists their names (roughly in
est thrust engine at that time, namely, a cluster of ve F-1 LPREs the order of the years of their start) and shows that there have been
at 1:5 £ 106 -lb thrust each. This record stood until 1985 when the mergers and consolidation.Most of these companies have gone out
Soviets ew a somewhat larger engine. The 1:8 £ 106 -lb of the ex- of the LPRE business, were acquired, or merged. Today there are
perimental F-1A is the highest known thrust of a LPRE ground test. ve that are active at the time of this writing, and each will be brie y
3) There were a good number of inventions, innovations, or rst discussed. In the book currently being written, there are discussions
implementationsof technologythat can be credited to organizations of LPREs of all of the 14 listed companies.There were other compa-
in the United States. This includes the rst ight of an engine with nies, but they are not shown in Table 3. They include aerospacecom-
liquid hydrogen as a fuel, the rst expander engine cycle, the theory panies and subsystem suppliers, where the development of LPREs
of bell-shaped nozzles, the rst TP, the rst booster pump, the rst was a sporadic or sideline activity. None of their engines has as yet
resulted in a production. The data in this section come from the
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
expander engine cycle engine, and the rst applications of gimbals
to TCs for thrust vector control (TVC) (1947). Furthermore, there author’s recollection and from personal communications.
was the development of the tubular thrust chambers for regenerative Employment in the LPRE business was high between 1955 and
cooling, the rst use of ablativeson LPREs, the developmentof spe- 1968 during the cold war period and historically the busiest time
cial materials for the hot-TC walls and for turbine buckets, at plate in the LPRE industry. The real funding available to LPREs has
machined injectors made of forgings, certain clever valve designs, since decreased, as have the number of employees and the number
certain injection patterns, Aerojet’s platelet injectors, the electronic of companies. For example, the peak total LPRE employment at
engine controls, pressurizing propellant tanks with gas generators, Rocketdyne was about 20,000 people in 1964; it hit a low in 1971
variable position pintle injectors, rst ight of an extendible nozzle, but went up to about 5500 in 1985 and was about 2800 in 2001.
rst applicationof liquid side injection for TVC, and very fast small Aerojet’s employment in LPREs personnel peaked in 1963 at about
propellant valves mounted on an injector of a small thruster. 10,000 and was about 1000 in the year 2000.
4) There are a number of clever innovations that were conceived Although some good research and development (R&D) work has
and implemented with pride in the United States, but very similar in- been done in the United States by certain universities, some private
novations were actually accomplishedat an earlier date in Germany research organizations, and by government laboratories, they are
and/or the Soviet Union. The LPRE work in these two countries was not discussed here because they did not develop and qualify LPREs
secret and advancedat the time and not known to U.S. LPRE person- that went into production or became operational. This includes for
nel until years later. Examples include the staged combustion cycle, example R&D work at Princeton University, Cornell University,
the use of inducer impellers ahead of the main pump impellers, early Purdue University, Pennsylvania State University, University of
reinforced concrete test facilities, the highest thrust large ying en- Alabama, Naval Postgraduate School, and California Institute of
gine, earliest LPREs speci cally for aircraft installation, the rst Technology.The graduate and undergraduateeducation of quali ed
airplane ights with LPREs, prepackaged LPREs with storable pro- technical LPRE personnel is perhaps the universities’ most impor-
pellants, or TVC by auxiliary or vernier thrust chambers supplied tant contribution,and more than 25 U.S. universitieshave at one time
by the main engine feed system. It also includes a large LPRE with taught courses concerned with LPREs. Research organizationsthat
the highest chamber pressure and pump-fed experimental LPREs have worked on LPRE issues include Batelle Memorial Institute,
with certain high-energy propellants. The Aerospace Corporation, and SRI (formerly Stanford Research
5) The RL-10B2 of Pratt and Whitney is the rst pump-fed Institute). Government organizations doing or having done work
LOX/LH2 engine that has own (in 2000) with the highest known on LPRE includes the Rocket Propulsion Laboratory (now part of
speci c impulse, namely, 467 s. It was the rst ight application of Philips Laboratory of the U.S. Air Force) at Edwards Air Force
an extendible nozzle exit segment with a LPRE. Base, Edwards, California, the U.S. Air Force Arnold Engineering
6) The United States had in 1965 a small radiation cooled thruster Test Center, at Tullahoma, Tennesee, NASA Marshall Space Flight
that could demonstrate over 100,000 cycles or restarts, a record at Center at Huntsville, Alabama, NASA-funded Jet Propulsion Lab-
the time. oratory in Pasadena, California, NASA Stennis Space Center in
7) Several of the propellants originated in the United States. This Mississippi, and NASA John H. Glenn Research Center at Lewis
probably includes RFNA, inhibited RFNA, Aerozine 50, mixed ox- Field, Cleveland, Ohio. These and other government organizations
ides of nitrogen,gelled propellants,hydroxylammoniumnitrate, and are indeed useful in testing LPREs and de ning, selecting, funding,
ultrapure hydrazine that would not contaminate its catalysts. guiding, and monitoring work at U.S. companies, doing R&D and
8) The United States was a leader in the applicationof computers testing, and in providing propellants, testing facilities/services for
and software to LPREs. Computers became part of the controls large and small LPREs, hover test facilities, or simulated altitude
for LPRE beginning in the 1970s. The Rocketdyne RS-68 was the test facilities.
rst LPRE to be fully designed by computers in the 1990s. This
covers not only design or analysis programs, but also integration 1. Reaction Motors, Inc. (RMI)
with manufacturing, test operations, engine controllers, cost and RMI was the rst American LPRE company.32¡34 It was founded
schedulecontrol,spare parts inventory,test data reductionor display, by four amateur experimentersof the American Rocket Society and
and other areas. incorporated in August of 1941. The words rocket motor were used
9) The Nike –Ajax antiaircraftmissile used a LPRE for the upper in those days for what today is designated as a rocket engine. In
stage, and it was the rst such military air defense system to be 1958, RMI was acquired by and became a division of Thiokol Cor-
deployed for this purpose. poration, a manufacturer of solid propellant rocket motors. In 1972
10) The United States has probably built more different JATO the Reaction Motor Division was shut down and ceased operations
LPREs than any other country. due to a lack of business. Because of the limit on the length of this
11) The production of 50,000 Bullpup engines was a unique ac- summary, only two of their historic LPREs will be discussed.
complishment; it represents the largest number of LPREs ever pro- The best known and perhaps the historically most signi cant of
duced anywhere. their engines was the RMI 6000-C4 aircraft rocket engine35 with
SUTTON 991
Table 3 U.S. companies in the LPRE Business (1941–2002)
Company Typical LPRE work Comments
RMI Engines for experimental aircraft, Started Dec. 1941,
Viking, prepackaged, vernier merged into Thiokol 1958,
reaction control thrusters stopped operating 1972
Aerojet All types of LPRE Started in 1942, bought by General Tire
was Aerojet Engineering Corporation, 1944, spun off as General
Corporation Corporation, bought General Dynamics
propulsion operation 2002
Curtiss-Wright Corporation LPRE for research aircraft 1943–1960a Work stopped
General Electric Company Hermes and V-2 operation, 1944–1967a
(Rocket Section) Vanguard booster Section was dissolved
Rocketdyne Propulsion All types of LPRE Started 1945 as part of North
and Power, since 1996 a American Aviation, merged into
part of The Boeing Company Rockwell International Corporation 1964
Walter Kidde & Company H2 O2 monopropellant 1945–1958a Work stopped
ARC Agena upper stage, small Formerly Bell Aircraft Company
Liquid Division. attitude control LPRE started LPREs in 1947,
became division of Sequa Corporation bought Royal Ordnance 1997
Acquired hydrazine monopropellant
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
line from Marquard 2001
(originally Hamilton Standard)
M. W. Kellogg Company JATO units 1945–1953a Work stopped
Pratt and Whitney, a United Upper stage LOX/LH2 Started 1957/1958
Technologies Company LPREs, Russian RD 180
Marquard Corporation, became Small bipropellant LPRE Started LPRE 1958, bought by Kaiser
Kaiser–Marquard, 1990 or reaction control systems 1990, bought hydrazine TC line
Sold to Primex 2001 from Hamilton Standard, in 1996
Northrop Grumman Corporation Attitude control LPRE, Started 1960 as part of Space
Propulsion Products Center lunar lander and others Technology Laboratory, name TRW
formerly TRW, Inc. adopted in 1965, bought by
Northrup Grumman 2003
Rocket Research Corporation Hydrazine monopropellant Started LPREs in 1963,
spun off as Primex, 1996 TCs and aquired by Olin Corporation 1985,
bought by General bipropellant TC (in 2000) Primex bought Kaiser–
Dynamics Corporation in 2001 Marquardt in 2000
Hamilton Standard Division Hydrazine monopropellant Started 1964, technology
United Technologies thrust chambers bought by Marquardt in 1995,
Corporation (small group) divested to ARC in 2001
General Dynamics Corporation (GD) All types of small TCs Bought Primex in 2001
ACS/maneuver systems sold to Aerojet in 2002
a
Date is not con rmed.
four fuel-cooled thrust chambers shown in Fig. 17. It was designed Table 4 Bullpup LPRE data
for the Bell Aircraft manned-research airplane X-1. This engine
Engine designation LR58 or Bullpup A LR62 or Bullpup B
propelled this aircraft on 14 October 1947 to a record speed of
Mach 1.06. The engine had four thrust chambers at 1500 lbf each Diameter, in. (cm) 12.1 (102.7) 17.3 (43.9)
(total 6000 lbf) with LOX/75% alcohol at a chamber pressure of Length, in. (cm) 40.5 (102.7) 61.2 (155.4)
220 psia and a speci c impulse (sea level) of 209 s. The initial Weight, loaded, lb (kg) 203 (92.3) 563 (255.3)
versions ew with a pressurized feed system. Later versions had a Weight, dry, lb (kg) 92 (41.8) 205 (92.9)
TP feed system with a GG supplied with hydrogen peroxide, which Thrust, lbf (kN) 12,000 (52.8) 30,000 (132)
Duration, s 1.9 2.3
was decomposed by a catalyst; the propellant tank pressures and Total impulse, lb ¢ s (kN ¢ s) 22,800 (101) 69,000 (307)
total propulsion system weights were lower. Each thrust chamber
had a small igniter chamber in the center of the injector designed
to allow multiple starts. The igniter used a spark plug to ignite a
small ow of fuel and gaseous oxygen that had been evaporated
in coils around the fuel feed pipe. The engine was improved and temperatureof ¡65± F. The most signi cant of the RMI prepackaged
used to y several later versions of the Bell X-1 research aircraft, LPRE was for the Bullpup air-to surface missile. Work started in
the Douglas D558-2 Skyrocket research aircraft, several unmanned 1958. It had the largest productionof any LPRE, and approximately
research lifting bodies, and as a dual engine for an interim power 50,000 units were delivered between 1960 and 1967. As shown in
plant for the North American X-15 research airplane. An up-rated Fig. 18, it used storable propellants (RFNA and a fuel consisting of
version of this four-barrel engine (at 435-psi chamber pressure and 50.5% diethylenetriamine, 40.5% UDMH, and 9% acetonitrile), a
7600–8400 lb total thrust) launched the small scale model of the central solid propellantgas generator (double-basesolid propellant)
Navaho missile (Project MX-774), but without restart capability. for pressurizingtwo annular propellanttanks, burst diaphragms,and
It was the rst U.S. engine with hinged thrust chambers, which a very short (inef cient) bell nozzle. As seen in Table 4 there were
allowed ightpath control. There were engine-related problems in two versions. It started with a powder cartridge moving a piston
two of the three ights. (the only moving part), which then sheared or broke a diaphragm
RMI developed several storable propellant prepackaged rocket and initiated the burning of the solid propellant grain; full thrust
propulsion systems. In the 1940s and 1950s, solid propellant motors was achieved in about 0.1 s. The thrust chamber was regeneratively
had problems operating at ambient temperatures lower than about cooled. The Bullpup had a design storage life of 5 years minimum
¡40± F, but the liquid engines could meet the required minimum and a safestoragetemperaturerangebetween ¡80 and C160± F (¡62
992 SUTTON
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Fig. 17 RMI 6000C-4 LPRE for research aircraft with four TCs; each could be turned on or off individually,thus, giving a stepwise change in thrust
(from Ref. 3).
Fig. 18 Cutaway section of the Bullpup A (LR-58) (from Ref. 34, Part III, 1983).
and C71± C). Bullpup became an operationalmissile in 1959. There Corp, Inc. It was the second U.S. company dedicated to rocket
were thousands of test rings, both on the test stand and in ight. propulsion, was started 1942, and it grew quickly.36¡38 It was men-
Reliability was rated at 0.9972. Military cutbacks and changing tioned in Sec. IV.C to have been a direct outgrowth of rocket propul-
military requirements eventually caused the Bullpup to be taken out sion projects done earlier at the GALCIT under the guidance of
of service. von Kármán, a famous aerodynamicist.31 For several decades, the
work on LPREs represented about one-quarter of Aerojet’s busi-
2. Aerojet General Corporation ness. Aerojet was also engaged in solid propellant rocket propul-
This company (originally known as Aerojet Engineering Com- sion, nuclear propulsion, ordnance, and other areas. In August of
pany) is commonly called “Aerojet”; it is a subsidiary unit of Gen- 2002 Aerojet acquired the rocket propulsionassets, LPRE products,
SUTTON 993
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Fig. 19 Schematic diagram of the Aerotojet LPRE concept (courtesy Aerojet).
skilled personnel, inventory, LPRE know-how, and facilities of the (which already had their main turbojet engine) to get quickly to
General Dynamics operation located at Redmond, Washington. In altitude and augment the ight performance.36¡41 This application
the past Aerojet has worked on a large variety of LPRE schemes, is differentfrom the JATO mission, where aircraft are given an extra
but only a few are presented here. push during takeoff(with heavy loads), while getting off the runway.
Aerojet developed more different JATO engines than any other Aerojet developed and ew such engines for three ghter aircraft, a
U.S. company.36¡41 The rst JATO LPRE was actually developed P-51 Mustang, a P-80, and an F-86. All used storable propellantsand
by its predecessor GALCIT in 1939, but improved by Aerojet in sophisticated pumped feed systems. Flight tests generally showed
1942. It had a thrust of 1000 lbf for 25 s duration, an uncooled TC, some improved aircraft performance during rocket operation, but a
and RFNA with aniline as propellants. The rst successful takeoff reduced aircraft range and/or reduced weapons load.
of a manned aircraft using two liquid propellant JATO units was on The government, through its intelligence agencies, knew in late
the Douglas A-20A attack bomber in January 1943 at Muroc Dry 1943 that the Germans were soon going to have a rocket-propelled
Lake in California. The rst productionof a JATO LPRE of Aerojet ghter airplaneand wanted a U.S. rocket-propelledairplane.Aerojet
was an improved, lighter, reusable,and more compact version of the conceived, designed, built, and partially tested a very unique LPRE
JATO that ew on the A-20A airplane.32 All of these early JATOs for an aircraft power plant, called the Aerotojet,37 and it was quickly
had three spherical tanks, one each for RFNA, aniline fuel, high- put under contract in 1944. As shown in Fig. 19, it had two stationary
pressure nitrogen pressurant, and an uncooled TC. By 1944 some TCs of 750 lb each and two rotating throttlable TCs at about 300 lbf
64 of these were delivered at a cost of $3450 each. each. The small TCs were mounted to a hollow shaft, but offset from
In the 1946–1954 period, Aerojet developed a series of different the shaft axis and slightly inclined to produce a torque. The power
JATO units for different military aircraft, and they were own suc- was transmitted through a gear case to four pumps: an aniline fuel
cessfully in experimental airplanes of the F-84 ghter-bomber, the pump, an RFNA pump, an oxidizer booster pump, and a lubricating
PB2Y-3 ying boat, and the B-29, B-45, and B-47 bombers. Most oil pump. An electric motor initiated the rotation and the start. It was
JATOs had cooled TCs, some used TPs feed systems with a GG, but intended to propel a new ying wing design of John Northrop, the
the early models used gas pressurized feed systems. The LPRE for founder of Northrop Aircraft Company. The author was a develop-
the PB2Y-3 is unique because its propellant pumps were driven by a ment and test engineerin the early phasesof this unique project.This
separate gasoline engine. Several of these Aerojet JATO units were was the rst known applicationof a boosterpump for preventingcav-
put into a limited production. The JATOs for the B-47 bomber had itation at the impeller of the main pump with a high vapor pressure
the highest thrust of any JATO and represent a historic achievement. propellant. The rotation of the small chambers caused a pumping
Two units, one on each side of the aft fuselage, with two TCs each, effect or additionalpressure rise and a mismatch of the impingement
gave total thrust of 20,000 lbf. The turbines of the two propellant of the fuel and oxidizer jet streams. In turn this caused incomplete
pumps were driven by warm air bleed from the compressors of the combustion and the accumulation of unburned and poorly mixed
aircraft’s jet engine. (Pumping reduced the weight of inert hardware propellants in the chamber. Two explosions of these accumulated
of the propellant tanks.) Liquid propellant JATO units were not used propellants broke experimental chambers. There were a number of
often by the military services because there was relatively little real other developmentproblem, which are brie y described in the book
need for takeoff assistance, and the servicing and refurbishing of length version. The delays in this development forced Northrop to
used units (with remnants of ammable, corrosive, toxic propel- go to an alternate propulsion scheme for the ying wing, and the
lants) was considered hazardous. None were put into operation by Aerotojet project was canceled in 1946.
the military services. A small-scale model of this Northrop ying wing aircraft was
Between 1945 and 1956, Aerojet also developed auxiliary rocket built earlier in this program (1944) to test the aerodynamics of this
engines or aircraft superperformance engines for ghter airplanes novel winged aircraft design. It was the rst piloted U.S. rocket
994 SUTTON
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Fig. 20 Flow diagram of one version of the Bomarc booster LPRE (courtesy of Aerojet).
airplane that was propelled by rocket power only.3;5;36;37;40 It was
known as Project MX-324. Aerojet built, tested, and ew a small
LPRE for this model airplane, and I was a part time member of
the team. It used RFNA and aniline as propellants, had a single,
small regeneratively cooled (with acid) thrust chamber of about
200-lb thrust (aluminum with a copper nozzle), and a pressurized
feed system, and it was restartable in- ight. The model aircraft was
air launched from a larger airplane and ew successfully for the
rst time in July 1943 and several times thereafter. This ying wing
experimental aircraft was so small that the pilot had to lie down on
his stomach with his head in the plastic nose of the airplane.
The remarkable booster LPRE 37;38 for the Bomarc Area De-
fense system (ramjet-powered supersonic missile) was developed
by Aerojet beginning in 1951. The second stage was propelled by
a supersonic ramjet. It is a historic LPRE because it successfully
overcame combustion vibration problems, had a novel tank pres-
surization scheme, and was the rst Aerojet LPRE to use a gim-
baled uncooled TC (with a ceramic lining) in a military operation.
A ow diagram of one version of the Bomarc booster LPRE is in
Fig. 20. The propellant tanks were preloaded with propellants mak- Fig. 21 Titan booster engine LR87-AJ-11 with two TCs, 151 in. tall
ing a sealed prepackaged system. Originally it used white fuming with a 15:1 nozzle area ratio (courtesy, Aerojet).
nitric acid and jet fuel propellants. It delivered 35,000 lb of thrust
for 4 s. The gas for the tank pressurization system came from two
GGs (one fuel rich gas for the fuel tank and one oxidizer rich), ori ce. The oxidizer tank is pressurized by NTO, which has been
both supplied from a separate set of small gas-pressurized small gasi ed in a superheater; a cavitating venturi regulates the liquid
propellant tanks. This saved considerable inert pressurization hard- NTO ow. There are no separate propellant valves for the GG, only
ware weight and minimized signi cant chemical reactions in the check valves to prevent reversed ow. The propellant ows to the
propellant tanks. The control (start/stop) was accomplished by time GG are controlled by two cavitating venturi tubes. Venturis were at
sequencing the operational steps and using high-pressure gas with the time a popular way to control the propellant ow and mixture
pilot valves to actuate the main valves. This method of timing con- ratio. The two main rotary propellant valves are on the same shaft
trol was common at the time, but is no longer used today. Early and use a single actuator. Each engine has a regenerative cooled
tests failed due to a high-frequency combustion instability causing gimbaled tubular thrust chambers and its own TP and GG. The gear
excessive heat transfer and erratic operation. After extensive inves- case oil is cooled by fuel. Each TP has a single high-speed turbine
tigations a change in fuel to UDMH (or really a mix of 60% JP-4 geared to two centrifugal propellant pumps and a lubricating oil
jet fuel and 40% UDMH) solved these vibration problems. Bomarc pump. This geared TP scheme optimizes the shaft speeds and oper-
was deployedand operatedwith the military servicesuntil 1972. The ates at high turbine and pump ef ciencies, thus, minimizing the GG
prime contractor, The Boeing Company, delivered 366 missiles. ow and the performance loss. The sustainer engine has one geared
The Titan LPREs were probably Aerojet’s most successful large TP (similar to the larger booster engine’s TP) and a single cooled
LPREs.37;38;42;43 A family of those were installed in the booster and gimbaled tubular thrust chamber with an ablative exhaust nozzle
sustainerstages of the several versionsof the Titan vehicle. The twin liner extension at the larger nozzle exit diameters. The turbine ex-
booster LPREs is shown in Fig. 21. A ow diagram of the sustainer haust gas from the sustainer engine was used for roll control of the
engine in Fig. 22 shows the fuel tank is pressurized by fuel-rich hot second stage. Titan I work startedin 1954, and the initial thrustswere
gas taken from the turbine exhaust; this gas is cooled by fuel in a 300,000 lb for the two booster engines and 60,000 lb for the sus-
separate heat exchanger, and the gas ow is controlled by a sonic tainer engine; the propellants were LOX and RP-1 (kerosene). The
SUTTON 995
Table 5 Several key parameters of recent Titan LPREs canceled. Modi ed NK-33 engines ( tted with a gimbal) were sold
by Aerojet to Kistler Aerospace for their unique SLV.44 Several
Application Booster LPRE Sustainer LPRE
dozen of this large LOX/kerosene engine were taken out of Russian
Engine designation LR87-AJ-11 LR91-AJ-11 storage and shipped to Aerojet.
Thrust, lbf, in vacuum 548,000a 105,000 In August 2002, Aerojet acquired the rocket propulsion orga-
Speci c impulse, s vacuum 301 316 nization of General Dynamics in Redmond, Washington. It had
Nozzle area ratio 15 49.2 two roots. The rst is the Rocket Research Corporation (RRC)
Mixture ratio 1.91 1.86 of Redmond, Washington, a leader in hydrazine monopropellant
Chamber pressure, psia 857 860
thrustersand GGs. RRC also built electricalpropulsionsystems.The
a
Dual engine. other is the Marquardt Corporation (later Kaiser–Marquardt Corpo-
ration) of Van Nuys, California, a leader of small, storable bipro-
pellant thrusters. Marquardt’s business was relocated to Redmond,
Washington, in 2001.45;46 General Dynamics and its predecessors
sold mostly assemblies of these small thrusters with their special
control valves, but also developed a few complete LPREs with pres-
surized feed systems.
RRC was founded in 1960, but became active in small hydrazine
monopropellantrocket engines only in 1963.45¡48 A major step for-
ward was the development of a suitable catalyst for hydrazine de-
composition by California Institute of Technology’s Jet Propulsion
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Laboratory in a joint effort with the Shell Development Company.
Thereafter, Shell produced the Shell 405 catalyst, which has since
been successfully used by all U.S. and some foreign competitors.
The Aerojet Redmond Center now produces this catalyst. Figure 10
shows one of their small hydrazine monopropellantthruster assem-
blies. The rst contracts for monopropellanthydrazine LPREs were
for transstage attitude control system (ACS) (25-lbf thrust) and for
the ACSs of Titan I and Titan II. Aerojet acquired a stable of proven
catalytichydrazinemonopropellantsthrustersranging in thrust from
0.1 to over 300 lbf and a history of successful applications in more
than 60 different space launch vehicles and spacecraft. By 2000 the
company had delivered 10,000 such monopropellant thruster and
GGs. Two to four individual thrusters and their valves were often
packaged into a subassembly or rocket engine module because this
simpli es their vehicle assembly, their checkout, and servicing.
Marquardt, the other predecessor company, started in the LPRE
businessin 1958 by conductinga study of the futurerequirementsfor
reaction control systems.45;46;49 They then performed in-housework
on a thrust chamber using the bipropellantcombination of NTO and
hydrazine. Marquardt’s rst contract for a bipropellant rocket en-
gine was for the Advent satellite of General Electric Company at
25-lb thrust. This LPRE experienced some failures during devel-
opment, but did pass the quali cation tests. However, for unrelated
reasons it never ew. After Marquardt experimented with a variety
of liquid storable propellants,they selected NTO and MMH for their
Fig. 22 Flow diagram of the Titan sustainer engine (LR-91-AJ-11) families of bipropellant thrusters. The 100-lb thrust R-4D thruster
with a GG cycle (courtesy, Aerojet). for the lunar orbiter became a versatile thruster for the company.
One version is shown in Fig. 11. Different models of this thruster
design have been successfully employed in several applications,
U.S. Air Force changed the propellants and the Titan II, a ballistic such as the Apollo service module (16 thrusters) or the Apollo lu-
missile, and subsequent versions used the storable combination of nar lander (16 thrusters). This R-4D thruster has an eight on eight
NTO with Aerozine 50 (50% hydrazine and 50% UDMH) (Ref. 43). unlike doublet impinging stream injector with eight small lm cool-
The rst production engines were delivered in 1961 and an initial ing holes and very closely coupled valves. Another version of this
operationalintercontinentalballistic missile (ICBM) capability had radiation-cooled thruster had a high nozzle area ratio of 375 and
been established by 1963. The thrusts of the Titan engines were up- an altitude speci c impulse of 323 s. It can be seen in Fig. 23 and
rated progressivelyfor the Titan III and IV SLVs to the values listed uses three different metals for the nozzle: rhenium (with an iridium
in Table 5. More than 1500 individual LPREs have been delivered protective coating can be used up to 4000± F) for the chamber and
for all of the versions of Titan over a period of 47 years. Unfortu- nozzle throat, niobium (for the upstream portion of the nozzle exit
nately high-frequencycombustion instability was observed in a few section, up to 3400± F), and titanium (near the nozzle exit, up to
tests, and this caused a major diversion to the program and an in- 1300± F). This improved costs and reduced weight. Over the years,
tensive R&D effort. After the injectors were equipped with baf es, a series of bipropellant thusters was developed, ranging in thrust
the combustion appeared to be stable. from about 1 (5 N) to about 900 lbf, with several materials, nozzle
Titan II ICBM, was installed in heavily armored undergroudsilos geometries, thermal insulation, durations, duty cycles, total number
and was operational between 1963 and 1987. When these Titan of pulses, and very short pulse widths (minimum total impulse bit
II missiles were decommissioned from their service as a military of 0.01 lbf ¢ s).
deterrent, they became available as SLVs. Titans III and IV are
bigger,upratedSLVs with larger enginesspeci cally for heavyspace 3. The Boeing Company, Rocketdyne Propulsion and Power
ight payloads. This company (Rocketdyne) has been the largest LPRE com-
In 1995 Aerojet obtained the right to sell in the United States pany in the United States. It has developed LPREs in every area of
several LPREs designed by the Kuznetsov Design Bureau in the application.50¡52 It was started in 1945 as a section of the aircraft
1960s for the Soviet N-1 moon ight vehicle, whose program was company North American Aviation, which in 1964 became part of
996 SUTTON
Table 6 Performance data for selected large Rocketdyne production engines/engine families
Missile/launch Redstone Navaho Atlas Jupiter Thor and Saturn I Space
vehicle MRBM cruise ICBM IRBM Delta IRBM and IB Saturn V Saturn IB shuttle Delta IV
application and SLV missile and SLV and SLV and SLV SLV SLV and V SLV SLV SLV
Engine family A-6 G-26 B-2C, MA-1, S-3D S-3E, MB-1, H-1 F-1 J-2 SSME RS-68
designation (data and A-7 MA-2, MA-3, MB-3, RS-27, A, B, C, phase 1 and 2
from last-in-family MA-5, and and and D block 1, 1A,
engine MA-5A RS-27A 2A, and 2
Initial family engine 1948 1950 1952 1953 1953 1958 1959 1960 1972 1997
design year
Thurst
Sea level, thousand lbf 78 240 430 (B) 150 200 205 1,522 DNA 374 650
60 (S) (100%)
Vacuum thousand lbf 89 278 480 (B) 174 237 237 1,748 230 470 745
84 (S) (100%)
Speci c impulse, s
Sea level 218 229 265 (B) 248 255 263 265 DNA 361 365
220 (S) (100%)
Vacuum 249 265 295 (B) 288 302 295 305 425 452 410
309 (S) (100%)
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Oxidizer LOX LOX LOX LOX LOX LOX LOX LOX LOX LOX
Fuel Alcohol Alcohol RP-1 RP-1 RP-1 RP-1 RP-1 LH2 LH2 LH2
(75%) (92.5%)
Mixture ratio 1.354:1 1.375:1 2.25:1 (B) 2.4:1 2.24:1 2.23:1 2.27:1 5.5:1 6.03:1 6.0:1
(oxidizer/fuel) 2.27:1 (S)
Chamber pressure, 318 438 719 (B) 527 700 700 982 717 2,747 1,460
psia 736 (S) (100%) (100%)
Nozzle area ratio 3.61 4.6:1 8:1 (B) 8:1 12:1 8:1 16:1 27.5:1 69:1 21.5:1
(exit/throat) 25:1 (S)
Nominal ight 121 65 170 (B) 180 265 150 165 390 (S-II) 520 400
duration, s 368 (S) 580 (IVB) Max
Dry mass, lbm 1,478 2,501 3,336 (B) 2008 2,528 2,010 (C) 18,616 3,454 7,774 14,850
1,035 (S) 2,041 (D)
Engine cycle GG GG GG GG GG GG GG GG SC GG
Gimbal angle, deg None None 8.5 7.5 8.5 10.5 6 7.5 11.5 10-MPL
circular 6-FPL
Diameter/width, in. 68 77 48 (B T/C) 67 67 66 149 81 96 96
48 (S)
Length, in. 131 117 101 (B) 142 149 103 230 133 168 205
97 (S)
Operating temperature ¡25 to ¡20 to ¡30 to C40 to 0 to 0 to ¡20 to ¡65 to ¡20 to ¡20 to
limits ± F C110 C110 C130 C130 C130 C130 C130 C140 C130 C140
First ight date 08-20 11-06 06-11 03-01 01-25 10-27 11-09 02-26 04-12 11-20
in family -1953 -1956 -1957 -1957 -1957 -1961 -1967 -1966 -1981 -2002
Comments Restart Throttlable Throttlable
in power level power level
space 67–109% 57–102%
B D booster, S D sustainer, Gas generator cycle, SC D staged combustion cycle, DNA D does not apply, MPL D minimum power level, FPL D full power level, A and C D inboard
engines, B and D D outboard engines, MRBM D medium-range ballistic missile, IRBM D intermediate range ballistic missile, ICBM D intercontinental ballistic missile, SLV D space
launch vehicle.
Fig. 23 RD-4-15 bipropellant thruster with a large nozzle used for operation in a vacuum (from Ref. 2).
SUTTON 997
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Fig. 24 First large U.S. LPRE was own on the Redstone missile (courtesy, Rocketdyne).
Rockwell InternationalCorporation. In 1955, the name Rocketdyne and performance, and a smaller relative chamber volume. The rst
was adopted, and it became a separate division of the parent com- hot ring of the new large TC (with a pressurized test stand feed
pany. Rocketdyne also was in the solid rocket motor business be- system) occurred in January 1950, which at 75,000 lb was then the
tween 1959 and 1978, is today in the space power supply business, highest rocket thrust in the United States. I was the designer and de-
and was in the electricaland nuclear propulsionareas. Together with velopment engineer for this TC. The rst static test of the complete
some other divisions,Rocketdynewas sold to The Boeing Company Redstone engine took place in late 1950, and the rst ight was
in December 1996, when the name listed as the title of this section in August 1953. It was the LPRE of the rst U.S. ballistic missile
was adopted. to become operational, and it also was deployed overseas in June
In its 58-year history Rocketdyne put about 15 large LPREs 1958. This engine launched the rst U.S. satellite (Explorer on 31
(1500–1,500,000 lbf thrust) and 17 smaller LPREs (1.0–1500 lbf) January 1958), and it also launched two U.S. astronauts, each in
into production,and several of these had between one and ve major their Mercury capsule, on their rst suborbitalspace ights in 1961.
redesigned or upgraded models. Up to 1 June 2001 Rocketdyne en- The preliminary design of the engines for the Convair Atlas bal-
gines had boosted1516 vehicles.In addition,Rocketdynedeveloped listic missile was started in 1952 in my engineeringsection and was
and tested more than 36 experimental engines or thrust chambers unique.53¡55 Detail design was in 1954. The two booster engines of
aimed at demonstrating feasibility or advances in technology. The 150,000-lb thrust each were mounted in a ring or doughnut-shaped
author worked for Rocketdyne from 1946 until 1975. structure at the missile’s tail, and this ring structure was dropped
Table 6 lists summary data for several large Rocketdyne LPREs from the ying vehicle after booster cutoff at about 170 s. The sus-
that have own. For engines that had several models of the same tainer engine with 60,000-lbf (located in the center of the aft end of
engine family the performance data in Table 6 refer to the most the vehicle) is also started at launch,but runs continuouslyfor a total
recent version. The two dates (design year and rst ight) give a of about 368 s. Figure 25 shows a ground test of one version of this
clue to the historical sequence. Because of the limitations on the three engine con guration. The nozzles of the two-booster engine
length of this paper, only 8 of the more than 20 historical engines are owing full, and their bright radiatingplumes have sharp bound-
will be brie y discussed. aries. The center sustainer engine is slightly overexpanded with its
The rst large engine development effort was a pump-fed LPRE 25 to 1 nozzle area ratio when operating at the low altitude of the
of 75,000-lb thrust, which soon became known as the engine for the test facility. Its jet has separated from the nozzle wall. Steam clouds
U.S. Army’s Redstone ballistic missile.53;54 This historic rst large from the test facility water spray are being aspirated into the central
U.S. engine is shown in Fig. 24. The Redstone LPRE had many sim- jet. They obscure the exhaust jet, which has a smaller diameter than
ilarities to the V-2 engine (same propellants, heavy thick low-alloy the nozzle exit. This one-and-one-half set of stages was selected
steel walls in the TC with lm cooling, a similar GG, an aluminum in part because Convair (the vehicle developer) and Rocketdyne
turbine, aluminum fuel and oxidizer pumps). Like the V-2, carbon were not sure, at that time, if altitude ignition of a sustainer could
jet vanes were used for TVC during powered ight. However, there be reliably achieved. The total three-nozzle thrust of 360,000 lb at
were some signi cant improvements and differences. It had 33% sea level (about 414,000 at altitude) was increased in steps in the
more thrust, 44% more chamber pressure,a new type of large diam- several subsequent modi cations and uprated versions of this Atlas
eter ( at surface) injector similar to the one in Fig. 10, cylindrical engine until it reached the performance given in Table 6. The TP
chamber geometry (not pear shaped), better combustion ef ciency for the booster-stage engine is based on the TPs for the Navaho
998 SUTTON
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Fig. 25 Static ring test of the MA-5A LPRE for the Atlas vehicle (courtesy, Rocketdyne).
booster and uses a gear case (Fig. 9). The TCs for the Atlas, Thor,
or Jupiter missiles had a tubular cooling jacket and originally a con-
ical nozzle exhaust section. Later these cones (and the tubes) were
changed to a bell-shaped nozzle exit. The fabrication and brazing
of tubular cooling jackets requires precision xture and precision
parts.
The Atlas was the rst U.S. ICBM and was operational by the
U.S. Air Force between 1961 and 1965. Several versions of this
Atlas engine also served in propelling satellite launches for mili-
tary spacecraft and space exploration payloads. This included the
Surveyer, Pioneer, or Intelsat satellites. The Atlas/Centaur engines
boosted the astronauts in the Mercury manned space ight program.
The Atlas LPRE was an active engine program for 46 years, in pro-
duction between 1956 and 1996, and 482 engine sets (consisting of
two boosters and one sustainer engine) have been delivered.
Rocketdyne engines launched both of the two launch vehicles
used in the U.S. Apollo (moon) program, namely, Saturn I and
Saturn V.5;6 Saturn I was boosted by eight H-1 engines, each ini-
tially at 165,000 lbf, but later upgraded in steps to 205,000-lbthrust
each. There were 19 ights including the rst international space
rendezvous. Saturn V has ve F-1 booster engines in the rst stage
(at 1:5 £ 106 lb thrust each) with a total launch thrust of 7:5 £ 106 lb,
ve J-2 oxygen/hydrogen engines at 230,000 lb (vac) each in the
second stage and one J-2 engine in the third stage. Saturn V was
launched 13 times in connection with the world renowned Ameri-
can Apollo program’s moon circumnavigation, landing, and return
programs. During one of these ights, one of the ve J-2 second-
stage engines exceeded an operational limit and was shut off pre-
maturely, but safely. The computer controller allowed the ight to
continue to the planned cutoff velocity with the remaining four en- Fig. 26 F-1 LPRE, the highest thrust U.S. engine (1.5 £ 106 lb) when
gines. This was an unscheduled ight con rmation of the engine-out it rst ew in 1967; 220 in. long and 144 in. wide (courtesy, Rocketdyne).
capability.53
The large F-1 LPRE has the highest thrust of any ying U.S. were encountered. There were 2771 single F-1 engine tests plus 34
engine and for more than a decade the highest thrust in the world54 tests of a ve-engine cluster. Static tests indicated a reliability of
(Fig. 26 and Table 6). Engine detail design started in 1962. It was 99.7%, and the ights were 100% reliable. Altogether 98 engines
the rst U.S. LPRE where the bottom nozzle exit section (between were built, and 65 of these have own successfully.
area ratio of 10 and 16) is lm cooled with warm (about 800± F) The J-2 engine was the world’s rst large engine to use LOX and
turbine exhaust gas. It has the largest U.S. single-shaft TP. After LH2 as propellants.54 Design started in 1960 (Fig. 2). It was the
an extensive investigation, the cooled baf es extending from the rst large Rocketdyne LOX/LH2 engine with two separate direct-
injector face (Fig. 5) solved the combustion vibration problems that drive TPs (no gears), namely, the oxidizer pump and a seven-stage
SUTTON 999
relatively simple, the engine is heavy and large, and it costs less than
a comparable staged combustion cycle engine, but its performance
is lower. This RS-68 is the rst LPRE that was fully designed on
computers using computer-aided design and analysis programs.
The LPRE development for the Lance surface-to-surfacemissile
started in 1964. It used a prepackagedstorable propellant [inhibited
RFNA (IRFNA) and UDMH] pressure-fed LPRE. 53;54 The 20.5-ft-
long Lance missile, developed by the LTV Aerospace System’s
Division, had integral preloaded propellanttanks, a piston-typepos-
itive expulsiondevicein the oxidizertank, and a solid propellantGG.
The one-of-a-kind compact concentric dual-thrust chamber assem-
bly (outer annular booster TC and smaller center sustainer TC) with
its valves and TVC was developed by Rocketdyne, and it is shown
in Fig. 28. The ablative TC liners were fastened to a forged steel
outer wall. The pitch and yaw TVC was accomplished by pulsed-
liquid-fuel side injection at four places on the outer nozzle exit of
the booster engine. It is the rst and only known production LPRE
with liquid side injection. The booster thrust was 46,200 lbf (at sea
level), chamber pressure at full ow was about 950 psi. The variable
thrust of the sustainer (from 4400 down to 14 lb) was achieved by
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
a movable pintle (with an ablative face) actuated by a servocon-
Fig. 27 Simpli ed ow diagram of SSME (courtesy, Rocketdyne).
trol valve and fuel pressure. This large throttling ratio of 300 to
1.0 is the highest known anywhere in LPRE history. About 3229
axial- ow fuel pump with its own turbine. This axial pump is more engines were produced. The missile was deployed with the U.S.
ef cient than a centrifugal pump for this range of volumetric ow, Army, and some Lance batteries were stationed in several overseas
but it is not quite as adaptable to offdesign ow conditions. countries.
The historic Rocketdyne LOX/LH2 SSME was designed in 1972,
and its design is a radical departure from prior U.S. engines.53¡57 It 4. Propulsion Products Center, Northrop Grumman Corporation
is still in limited productionat the time of this writing. Three of these Today the LPRE work is done at this Propulsion Products Center,
engines are used in a Space Shuttle Orbiter. Table 6 indicates it has which is part of the Space TechnologySector of Northrop Grumman
a high performance. It is the rst and only ying U.S. LPRE with a in Redondo Beach, California. In the fall of 2002, the Space and
staged-combustion-engine cycle, where the turbines are driven by Electronics Group of TRW including its propulsion capability was
high-pressuregas from two fuel-rich preburners and the turbine ex- acquired by Northrop Grumman. For decades this center was part
haust gases are fed to the injector into the combustion chamber. It of TRW, Inc., where many different LPRE were developed.60;61
gives a higher performance by about 6% compared to a GG cycle Today’s Propulsion Center had its origin in the LPRE work done
engine of the same thrust and nozzle area ratio. It is a reusable man- at the NASA Jet Propulsion Laboratory (JPL) in the mid-1950s.61
rated U.S. large LPRE. It can be throttled between 67 and 109% of Around 1960, several of the JPL engineers, who had worked on hy-
rated power level, and the temporary thrust reduction prevents ex- drazine monopropellant engines and the predecessors of the pintle
cessive loads and heating of the vehicle during ascent. The SSME injector, moved to the Space Technology Laboratory (the predeces-
simpli ed ow sheet in Fig. 27 shows four TPs, namely, a three- sor organization of TRW), and they continued their work there. The
stage high-pressure fuel pump, a single-stage oxidizer pump, and discussionof historicLPRE effortswill be limited to just a few items.
two boosterpumps. It was the rst U.S. engine where a boosterpump Inert cold-gas attitude control thrusters were historically the rst
(oxygen) was driven by a hydraulic turbine (by LOX) and not a gas TVC method used for steering space vehicles.61 Several of the initial
turbine or not through a gear case. The original chamber pressure spacecraft (1958–1959) built at TRW used this reliable simple cold-
of 3319 psia (initial version or block 1) was the highest of any U.S. gas system for attitude control, for example, in Pioneers 1 and 2,
productionLPRE. This allowed a relativelysmall chamber and noz- Nimbus, or the Earth Resources Satellites. Flights with their cold-
zle and a high nozzle exit area ratio (68.8) without excessive nozzle gas systems continued sporadically on other spacecraft until about
ow separation.It had a uniqueengine controland healthmonitoring 1974. A number of other U.S. companiesalso built and ew cold-gas
system.58 Figure 13 shows the initial successfulinjector design with TVC systems.
baf es, which preventedhigh-frequencycombustioninstability.The
SSME thrust chamberuses tubularconstructionin the nozzle diverg-
ing section, where heat transfer is lower. In the chamber and nozzle
region (with high heat transfer) it uses a milled channel design.
A unique design feature of the SSME is its power head design
made of nickel-based superalloy 718 forgings. It contains the main
injector assembly with a novel oxygen/hot-gas heat exchanger. It
is the backbone or key structure to which the two high-pressure
TPs, the thrust chamber, and two preburners are attached. Changes
were made to the engine in 1995 and again in 1996–1997. In the
block 2 version of the SSME, a larger throat was built into the
thrust chamber; this reduced the chamber pressure by 11%, reduced
the feed pressures, reduced the heat transfer, and provided more
margin.57 There were some TP problems, such as the turbine blade
cracks after a few static ring tests of the early version. Two safer,
more ef cient, more robust, and heavier new main TPs, developed
by Pratt and Whitney, replaced the original Rocketdyne TPs in the
current version.
The RS-68 booster engine (Table 6) was designed for low cost in
1997 and 1998. It ew rst in November 2002 (Ref. 59). A single
engine launches the rst stage of the Delta IV family of SLVs. It is Fig. 28 TC assembly of Lance surface-to-surface missile, 19.4 in. high
the highest thrust existing LOX/LH2 engine. Its GG engine cycle is and 173 lb weight (courtesy, Rocketdyne).
1000 SUTTON
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Fig. 29 Half-sections through the Lunar Module descent engine with a lm-cooled ablative liner and an enlarged detail of the variable area pintle
injector (courtesy, Northrop Grumman Corporation, Propulsion Products Center).
This was the rst U.S. LPRE organizationto y hydrazine mono- acts as a dual valve. Most other small pulsing TCs have a small
propellant LPREs with a gas-pressure-feed system.61 It was rst volume of propellant trapped between the injector surface and the
own in the Able 4A (also known as Pioneer P-1) in 1959. Because valve seat, and this trapped propellant dribbles out and causes some
a reliable catalyst did not exist in 1959, a slug start was used, that is, afterburning at low thrust.
an initial injection of a small amount of NTO, a hypergolic oxidizer; The historic lunar landing decent engine of the Apollo lunar
this created the high gas temperature necessary for the subsequent module63;66 (developed at predecessor TRW) had a throttling pintle
thermal decomposition of the hydrazine. The next spacecraft, Able injector and was capable of a 10–1 thrust reduction with only a 4%
4B, carried enough NTO for six slug starts. In the 1960s, Shell 405, loss of speci c impulse. This engine had an ablative TC with a metal
an effective catalyst, became available, and a series of hydrazine nozzle exit extension and lm cooling. It is shown in a section view
monopropellantLPRE systems (with pulsing, multiple thrustersand in Fig. 29. The trick in deep throttling is to maintain a high injector
a gas-pressure-feed system) were then developed. These LPREs pressure drop and the proper mixture ratio. This is accomplishedby
were capable for periodic rapid pulsing operation for attitude con- varying the annular injection area (by hydraulically positioning the
trol over a long period of time. Sizes from 0.1 to 150 lbf thrust were single movable pintle sleeve) and at the same time vary the throat
developed. Their pulsing hydrazine monopropellant engines were area of the two cavitating venturis with movable center bodies in
own on Pioneer 6, 7, 8, and 9 spacecraft between 1965 and 1968 the propellant feed lines. These venturis are labeled as ow control
and in several military and NASA spacecraft since that time. In the valves in Fig. 29. The cavitating venturis control the propellant ow
last several years this center developed a unique TC that could op- and the mixture ratio at any particular thrust level. A relatively thick
erate interchangeablywith bipropellants (NTO/hydrazine, 4–14 lbf ablative liner with lm cooling is used because regenerativecooling
thrust) or with hydrazine as a monopropellant (0.9–4 lbf thrust).61 was not feasible at low thrust, as mentioned in Sec. III.D.1. The
The rst ight of this TC was on a GEO-LITE satellite in 2002. gimbal ring around the throat limits the space needed for turning
This propulsion organization has re ned the pintle injector tech- the engine during pitch and roll maneuvers.
nology over a period of 40 years and has designed and tested over The U.S. Army and this center at Northrop Grumman have been
60 different pintle con gurations over a thrust range between 5 and leaders in the chemistry and applicationof gelled propellantsand in-
650,000lbf thrust.61;62 Most have a xed (not movable)pintle.An in- vestigateddifferentpropellantformulationsand operatingcharacter-
jector with a movablepintle sleeveis shown in Fig. 29 (Refs. 63–65). istics with several experimental LPREs beginning TC tests in about
As of October 2001, there have not been any combustion instabil- 1983 (Ref. 67). Gelled propellants have additives that make them
ity incidence over a wide range of chamber pressures and with 25 thixotropic(jellylike) materials. The merits are enhanced safety be-
propellant combinations. It gives good performance, but requires a cause it is less likely to leak, be spilled, or react violently to impact.
relatively large combustion chamber volume. Eight of their LPREs By adding powdered aluminum or small carbon particlesto the fuel,
with pintle injectors have own, most of them with xed (nonmov- the fuel density and the combustion energy can be increased. Its
able) sleeves. Another feature of the pintle injector is the ability principal disadvantages are poorer atomization and combustion ef-
to shut off the propellant ow at the injector surface and to reduce ciency, causing a small decrease in performance and a somewhat
the propellant dribble volume essentially to zero. The pintle sleeve higher amount of residual (unused) propellant. Figure 30 shows
SUTTON 1001
Table 7 Characteristics of two versions of RL10 rocket engines
Design year
Characteristic 1958 1997
Designation RL10A-3 RL10B-2
Thrust in vacuum, lbf 15,000 24,750a
Chamber pressure, psia 300 644
Nozzle expansion are ratio 40 280a
Speci c impulse (vacuum), s 427 465.5a
Mixture ratio 5.0:1 5.88:1
Design life (number rings/cumulative 100/1.25 300/10
duration), h
a
With extendible diverging nozzle segment.
Fig. 30 Compact, prepackaged,preloaded LPRE for propellingexper-
imental smart ground-to-groundmissile (courtesy, Northrop Grumman
Corporation, Propulsion Products Center).
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
a unique prepackaged LPRE (6 in. diameter) for an experimental
smart missile, which is in an advanced development program.68 It
uses gelled propellants and a face shutoff-type pintle injector capa-
ble of some throttling and restarting. It includes features that allow
a control of the total impulse and of the time of ight. It uses a solid
propellant gas generatorwith multiple grains for tank pressurization
under differentambient temperaturesand ranges. Pistons in the con-
centric propellant tanks provide positive propellant expulsion and
separate the oxidizer liquid from the fuel-rich gases of the GG. The
propellants, IRFNA and MMH, are gelled, and the fuel has been
loaded with suspended carbon particles making it denser and more
energetic. It has own successfully in two experimental launches
from a military vehicle. There has as yet not been a production of
a LPRE with gelled propellants, even though these propellants had Fig. 31 Simpli ed schematic ow sheet of early version RL10 upper-
been investigated for more than 30 years. stage LOX/LH2 engine (courtesy, Pratt and Whitney, a United Tech-
nologies Company).
5. Pratt and Whitney, a United Technologies Company
The main product line of Pratt and Whitney, a United Technolo-
gies Company (P&W) was and is turbojet engines. The decision around it. It was patterned after a similar injector developed by the
by P&W/UTC to enter into the LPRE business was made in 1957, NASA Lewis Laboratory, now called NASA John H. Glenn Re-
and serious work began in 1958 (Ref. 69). This entry was based in search Center at Lewis Field. An innovation attributed to P&W was
part on their prior experience on rotating turbojet machinery and on the porous stainless-steelmaterial (called Rigimesh). It was used for
having used, handled, pumped, and burned LH2 in a special turbojet the injector face; hydrogen oozed through the pores of this material
engine, which at the time was a secret project. In the past their work providing a transpiration-cooled surface. This basic injector design
has concentrated on LOX/LH2 LPREs.70 The discussion is limited with the coaxial injection elements and the Rigimesh were used on
to three items. subsequent U.S. LOX/LH2 LPREs.
The RL-10 LPRE was the most successful and historic engine This engine had several improved and upgraded versions. The
of P&W. As of June 2003, the several versions of this LPRE thrust was increased in steps until it reached the value shown
logged more than 352 space ights and accumulated 659 in-space in Table 7. It has seen service as a dual engine in the Centaur
starts.69¡72 It was the rst ying LPRE in the world to use LOX/LH2 upper stage, was used in unmanned lunar landings, planetary
cryogenic propellants.The initial version had 15,000-lbfthrust. The yby/orbiters, Mars lander, astronomical observatories,and a num-
initial application was the Centaur upper stage with two gimbaled ber of other space programs. A variable thrust version (throttled
RL10 engines. Design of the RL10 LPRE started in 1958. The rst down to 30%) with four RL10 engines was developed for the
static engine ring occurredin 1959, and the rst ight was launched McDonnel–Douglas experimental Delta Clipper (DC-X) vertical
in a Centaur upper stage in January 1963. Data of two very different takeoff and vertical landing test vehicle. These engines have own
models of this LPRE are given in Table 7. 12 times, but these ights are not included in the total number of
The RL10 was the rst ying LPRE in the world to use an ex- space ights listed earlier. P&W was the rst in the world to y a
pander enginecycle.73 Its ow sheet is shown in Fig. 31. The gasi ed large extendiblenozzle with a high area ratio on LPREs The RL10B-
hydrogen(heated in the cooling jacket from ¡423 to about ¡165± F) 2 engine shown in Fig. 32 is the most recent. This extendible nozzle
powers the turbine of the TP, and the hydrogenturbine exhaustgas is concept was invented at the UTC Research Laboratory in 1948 and
then injected into the combustion chamber, where it is burned with had been used by P&W in solid propellantupper-stagerocketmotors
the LOX. The two-stage turbine and the two-stage hydrogen pump beginning about a decade earlier. Its extendible nonporous nozzle
are on one shaft, and the oxygen pump is driven at reduced speed segment was made of a three-dimensional weave of heat-resistant
through a gear case. The gear case and the main bearings are lubri- strong carbon bers in a matrix of carbon. The two-piece exten-
cated and cooled by oil. SubsequentlyP&W designed the fuel pump sion was developed and built by Snecma in France. The movable
bearing to be cooled by LH2 , which had not been done before. A nozzle extension is stowed around the engine during the ascending
formed, double-tapered,and attened set of 347 stainless-steeltubes ight and then lowered into position after the dropoff of the lower
is used for the cooling jacket of the thrust chamber and the nozzle. vehicle booster stage, but before the second-stage engine starts at
It is basically similar to the tubular TCs developed years earlier by altitude. The RL10-B2 with the large extendible nozzle rst ew in
RMI, Aerojet, and Rocketdyne. The injector featured multiple con- August 2000.
centric injectionelements with the oxygen coming through the small When Rocketdyne had technical problems with their own design
inside tubes and the gasi ed fuel coming through the annular sleeve of the SSME TPs (such as turbineblade cracking),NASA decided to
1002 SUTTON
and two booster pumps. The thrust chamber has rectangular cool-
ing channels, three lm cooling injection slots, and its assembly
requires brazing and multiple welds.
6. Liquid Propellant Division, ARC
The Liquid Propellant Division of ARC, a Sequa Company, has
three sources for its technology.75 The former Bell Aircraft Com-
pany started its work on LPRE in the 1950s; in 1960, Textron bought
this company and establishedthe Bell Aerospace Division as a sepa-
rate organization.In 1983 ARC acquired its liquid propellantopera-
tion (together with its inventory, personnel, and facilities at Niagara
Falls, New York) to supplement its own solid propellant rocket mo-
a) Half section of nozzle extension in stowed position tor product line. The second source came to ARC in November
1997, when it acquired the Royal Ordnance Company of Westcott,
England. With it they received some skilled people, facilities, a line
of proven storable bipropellant modern apogee LPRE, and small
thrusters for reaction control systems (RCSs), all operating with
hydrazine as a fuel, and an entry into the European market. Pure hy-
drazine gives a few more seconds of speci c impulse and a higher
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
average propellant density than MMH or UDMH fuel, and these
are needed in some high-performancespace ight applications.The
third source was the hydrazine monopropellantrocket engine prod-
uct line originallydevelopedby Hamilton StandardDivision of UTC
in the late 1960s. This product line had been acquired by Kaiser–
Marquardt Corporation,but when this company merged into Primex
in 2001, the government required Marquard to divest itself of this
hydrazine line. ARC then acquired this Hamilton Standard mono-
propellant line, its designs, manufacturing know-how, test data, and
customers. It lled a signi cant gap in the ARC thruster product line
with reliable proven products.75¡77 Hamilton Standardstarted in this
eld in about 1964 and had quali ed thrusters with integral valves,
good performance, and endurance in sizes of 0.2-, 1.0-, 2.0-, 5.0-,
20-, and 100-lbf thrust. Figure 34 shows the 5-lbf thruster. It has
a steady-state speci c impulse of 231.5 s at 200-psi feed pressure.
b) Nozzle extension in deployed position The reaction chamber contains the catalyst. A standoff perforated
Fig. 32 RL10B-2 LPRE with an extendible nozzle skirt (courtesy, metal sleeve minimizes the heat conduction to the mounting ange
Pratt and Whitney, a United Technologies Company). and the valve. The electric heaters avoid freezing of the fuel and bad
starts with a cold catalyst bed.
During the 1950s and 1960s, hydrogen peroxide monopropellant
have P&W evaluate the Rocketdyne turbopumps. P&W developed thrust chambers were produced and own in several sizes between
an alternate version of the high-speedmultistage high-pressureLH2 1.0- and 500-lb thrust.78;79 Over 2500 were built. A silverplated
TP and the high-pressure LOX TP for assembly into the SSME. screen catalyst was used. Bell-ARC delivered such RCSs for of the
This engine has been developed and is being built by Rocketdyne. X-1B and X-15 research airplanes, the rst U.S. manned ights
These alternate TPs are more reliable, interchangeable,and 350 lbs with the Mercury space capsule (1-, 6-, and 24-lbf thrusters), or the
heavier than the original Rocketdyne versions. These large TPs are Centaur upper stage for SLVs (1.5-, 3-, and 50-lbf thrust). Most had
still being built by P&W and delivered to Rocketdyne for assembly pulsing capability. Hydrogen peroxide monopropellantis no longer
into the SSME block 2. used because of its low performance.
About 10 years ago P&W reached an agreement with the largest Bell Aerospace (ARC predecessor) developed a series of good
and most experiencedRussian LPRE developing organization NPO bipropellant (mostly NTO and MMH) low-thrust radiation-cooled
Energomash at Khimky (Moscow Region) to obtain a license and rocket engines for reaction control, attitude control, station keeping,
later also to produce some of the proven Russian LPREs for pro- or ywheel desaturation. One of them is shown in Fig. 35; it has
pelling U.S space launch vehicles. This includes the RD-170 (high- some external thermal insulation to protect adjacent components
est thrust existing engine, 1,770,000 lb) liquid oxygen-kerosene from excessive thermal radiation and to keep the outer temperatures
booster engine, the RD-180 (described later), and the RD-120 below 400± F. A single triplet set of injection holes is used, and a
(187,400-lbf thrust). All run on LOX/kerosene. The RD 120 is a torque motor operates the bipropellant valve. After testing different
second-stageLOX/kerosene LPRE, and it was test red at the P&W materials, ARC settled on niobium for the chamber/nozzle walls
test facility in Florida in October 1995. It was the rst test ring of a because of its manufacturability and relatively low weight. A sili-
large Russian LPRE in a U.S. rocket static test facility.74 Of these en- cate coating to minimize oxidation by the hot reaction gases, and
gines, only the RD-180 found a U.S. application.It is a scaled-down a process to apply this coating was selected from various alterna-
version of the RD-170, which has four hinged thrust chambers, but tives. With valves closely mounted to the injector, the start time
the newer RD-180 has only two gimbaled thrust chambers. The spe- delay can be less than 5 ms, and the pulse duration can be as low
ci c impulse is 311 s at sea level, 337.8 s (in vacuum), is 140 in. as 10 ms for the low-thrust units. ARC now has a stable of small
high, has a maximum diameter of 118 in. and weighs 12,081 lb ight-proven bipropellant thrust chambers assemblies in sizes be-
(dry). It was developed by NPO Energomash of Moscow in coop- tween 0.2 and 350 lb and today is a leader in this eld. Beginning
eration with P&W. Figure 33 shows this engine. Its rst ight in in August 1965, ARC has designed, built, and delivered several dif-
the Atlas 5 SLV was in August 2002. It has a sea level thrust of ferent completely integrated LPRE with tanks, pressurizing system
860,000 lb (933,000 lb in vacuum), runs at a very high chamber and support structures,such as the Minuteman postboost propulsion
pressure of over 3720 psia and a nozzle exit area ratio of 36.4:1.0. system.
It uses a staged combustion cycle, an oxygen-rich preburner, an ox- The Agena rocket engine (Fig. 36) was probably the best known
idizer lead during start with an initial ow of a hypergolic starting and the largest developed by Bell/ARC. 80 It propelled an upper
liquid, high-pressurehelium for valve actuationand system purging, SLV stage (Agena), developed by the Lockheed Corporation in
SUTTON 1003
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
Fig. 33 RD-180 LPRE has two gimballed TCs (pointing up) central common TP, two booster pumps, high-pressure turbine exhaust pipe which feeds
gas to the injectors; TVC actuator on right (courtesy, NPO Energomash and Pratt and Whitney, a United Technologies Company).
Fig. 34 Typical 5-lbf hydrazine monopropellant thruster with valve (courtesy ARC).
Sunnyvale, California. The TP-fed engine used a GG cycle and hoops (an egg crate pattern). Two ARC positive expulsion piston
NTO/UDMH as its hypergolic propellants. An early version used tanks provide multiple restart capability in space to feed the GG.
high-density nitric acid (with a high percentage of dissolved NO2 ) Cavitating venturis in the feed lines leading to the GG provided
and UDMH. It has a unique aluminum (6061 T6) thrust chamber the control of the GG propellant ow and, thus, the thrust level.
(oxidizer cooled) with a relative thick wall, which contains long The propellant centrifugal pumps were geared to the turbine. The
drilledholes (inclinedto the axis) as coolingpassages.The radiation- Agena engine Model 8096 has these characteristics: altitude thrust
cooled nozzle exit section (between area ratio of 12 and 45) is made 16,000 lbf, altitude speci c impulse 290 and later 300 s, chamber
of titanium reinforced externally with molybdenum stringers and pressure 506 psia, chamber mixture ratio 2.8, GG mixture ratio 0.15,
1004 SUTTON
dry weight 296 lb, nozzle area ratio 45, exit diameter 32.5 in., and it with con dence. Certainly there have been recent new technical
a height of 83 in. A total of 418 Agena engines have been produced ideas that have improved LPREs, such as better materials, lighter
and 363 have own with only one engine failure. This LPRE pro- extendible nozzles, or simpler TPs. There have also been a few new
pelled many different payloads, including the rst U.S. satellite in a potential requirements, such a microminiaturized LPREs, combin-
circular low orbit, the rst into a polar orbit, and the rst to perform ing LPREs with other means of propulsion, or reusable LPRE for
a signi cant orbit velocity vector change. It also was the rst large reusable strapon boosters. There are still areas of this technology
upper-stage engine with a vacuum restart and the rst to participate where R&D can lead to further improvement. However, the oppor-
in a space rendezvous and docking operation with a manned space- tunities for developing a truly new LPRE are today not as plentiful
craft, namely, the Gemini. A version of an Agena engine was also as they used to be.
ground tested with liquid uorine and ammonia as propellants. The progress in the technology of LPREs has been truly remark-
able in the last 82 years. Many of the technical milestones and key
VI. General Findings, Comments, and Conclusions LPREs have been discussed in this paper. The rate of innovation,
The LPRE eld and its technology are today essentially mature. introducingnew materials or new designs, was higher in the rst few
The basic engine system and key components had been fairly well decades of this history than it has been in the more recent decades.
de ned about 40 or 50 years ago. High reliability numbers are The emphasis on the engine development criteria has changed.
recorded by all LPREs in production. A large amount of data is In the early decades the LPRE investigators were happy if the en-
available. For any particular kind of new LPRE, there are today gine held together, ran for at least 10 s, and did not fail. Making it
two or more organizations in the United States that can develop work was the key objective. Soon the emphasis shifted to running
for more than a minute. Then the aims became maximum possible
performance, high reliability, safety, reduced costs, and for some
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
applications also long life. Making it environmentally compatible
has become a goal more than a decade ago. Today the emphasis
is still on these same criteria, but more attention is being given to
cost.
The United States has the distinction of being identi ed with the
American Robert H. Goddard, a most important early pioneer in
this eld, the rst to build and run a LPRE and the rst to launch
a vehicle with a LPRE. His technical contributions are legendary.
It is unfortunate that many of his inventions did not reach the U.S.
industry in a timely manner.
Current space-relatedapplicationscan only be accomplishedwell
by using LPREs. A number of the ight applications, for which
LPRE were indeed appropriate, are obsolete today. In some ap-
plications they have been replaced by other means of propulsion.
For example, we no longer use LPREs for JATO, aircraft superper-
formance, or sounding rocket vehicles. We are no longer building
LPRE for ballistic missiles. The emphasis today is on applications
Fig. 35 TC assembly 23-lbf thrust (of the Minuteman III postboost for space launch vehicles,spacecraft maneuvers, or reaction control
control propulsion system) operates on NTO/MMH (courtesy, ARC). for the steering of ying vehicles.
Fig. 36 Agena 8096 engine (courtesy, ARC).
SUTTON 1005
In the last 30 years, the industry has settled on a few speci c barrierhave been remarkableaccomplishments.The space age could
practical propellant combinations,each for a speci c type of LPRE not have happened without LPREs. The dramatic advances of this
or application.A couple of propellantswith potentialfuture bene ts technology and the ight progress that LPREs have enabled are
are still being evaluated. In the rst 50 years of the U.S. LPRE his- indeed amazing and a source of satisfaction to the author and to
tory, about 170 different liquid propellants have been evaluated and others.
many have found their way into experimental TCs and/or LPREs.
More than 25 combinations have own. However, most propellant Acknowledgments
combinations are no longer used. This paper would not have been possible without the help and
The LPRE business in the United States has seen its peak in information received from about 20 people and organizations.Sev-
the late 1950s to the early 1970s. This was the period when LPRE eral are recognized by name and organization in the listings of the
employment and sales were at their highest. Although the volume of references as having provided valuable information, comments, or
new LPRE developmentshas greatlydiminished,there is today still a data through personal communications.Thanks are due to the three
lot of activity worldwide.The decreasein businesshas broughtabout people, who kindly reviewed the draft of this paper and provided
mergers, acquisitions, hirings or layoffs, cooperative agreements, valuable suggestions for improvement. Special recognition is given
closures, and consolidations, and many of these were described. to Vince Wheelock of Rocketdyne (for assembling the data and
There are fewer U.S. companies and fewer people engaged with preparing Table 6 and for other liquid propellant rocket engine in-
LPREs today. To date the capability to develop and produce has formation), to Mark Fisher of NASA Marshall Space Flight Center
been maintained. (for arranging the drawing of Fig. 20), to Charles M. Ehresman of
In the last 70 years a lot of researchrelatedto LPREs has been done Purdue University, and to Mark Coleman of the Chemical Propul-
at U.S. universities, government or company laboratories. Some of sion Information Agency for old technical literature. The staff at
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
the researchprojectshave indeedbeen helpfuland resultedin a better several libraries and at the AIAA of ce were instrumentalin nding
understandingof the physicalor chemicalphenomena,the materials, and providing some of the backgroundliterature and old references.
or the analysis of LPRE related subjects. One university’s R&D has
led directly to the start of one of the key LPRE companies and to References
a government-supported laboratory. However, this author believes 1 Huzel, D. K., and Huang, D. H., Design of Liquid Propellant Rocket
that most of the university research related to LPREs in the last Engines, revised ed., Vol. 147, Progress in Astronautics and Aeronautics,
35 years has not been directly useful or only marginally helpful in AIAA, Reston, VA, 1992.
making a better or novel LPRE. There have been some exceptions. 2 Sutton,G. P., and Biblarz, O., Rocket PropulsionElements, 7th ed., Wiley,
The propulsion designer or typical engineering manager had not New York, 2000.
been very aware of the research that had been conducted, had not 3 Herrick, J., and Burgess, E. (eds.), Rocket Encyclopedia, Illustrated,
read a Ph.D. dissertation or research report, and had not interacted Aeropublishers, Los Angeles, 1959.
4 Jane’s Space Directory, 16th and 12th eds. Launch Vehicle Propulsions
with the researchers. However, several U.S. universities have been
effective in educating professional personnel for work in the LPRE Sections, Jane’s Information Group, Ltd., Coulsdon, England, U.K., 1996–
eld. 1997, 2000–2001.
5 Baker, D., Space Flight and Rocketry, a Chronology, Facts on File, Inc.,
It took a few years for each company to become pro cient in this 1996.
business. Each organization had to learn about LPREs from scratch 6 Ordway, F. I., and von Braun, W., History of Rocketry and Space Travel,
and acquire or develop a suitable set of knowledgeable employees, T. Y. Crowell Publishers, New York, 1975.
computer programs, facilities, and laboratories. Each usually made 7 Doyle, S. E. (ed.), History of Liquid Rocket Engine Development in the
mistakes, experiencedengine failures, and had to repeat some work, United States 1955–1980, American Astronautical Society, AAS History
which had already been done well by others. Series, Vol. 13, Univelt, Inc., San Diego, 1992.
8 Ehricke, K. A., “A Comparison of Propellants and Working Fluids for
The work done by the LPRE industry has typically become more
ef cient. Ef ciency measurements,such as sales per employee,have Rocket Propulsion,” Journal of the American Rocket Society, Vol. 23, No. 5,
generally risen. This is in part due to the learning curve (better 1953.
9 Chemical Rocket PropellantHazards, Vol. 1,General Safety Engineering
educated and more experienced personnel) and major advances in
Design Criteria, CPIA Publ. 194, Chemical PropulsionInformation Agency,
computers,databases, and software programs for engineering,man-
Columbia, MD, 1971.
ufacturing, or management. It is only recently that one company 10 Liquid Propellant Manual M4, Chemical Propulsion Information
designed a new LPRE completely by computer. Agency, Columbia, Maryland, 2000.
As in many other technical endeavors, the techniques for design, 11
Knuth,E. L., “The Mechanics of Film Cooling,” Journal of the American
manufacture, and testing have drastically changed. This author re- Rocket Society, Nov.-Dec. 1954.
members having designed engines in the 1940s using a slide rule, 12 Bartz, R. R., “Survey of Relationships between Theory and Experiment
inked drawings on vellum, and a handwrittennotebook and many of for Convective Heat Transfer in Rocket Combustion Gases,” Advanced in
the manufacturinginstructionswere verbal.It has graduallychanged Rocket Propulsion, edited by S. S. Penner, AGARD, Technivision Services,
to today’s sophisticated computer-aided design techniques, which Manchester England, U.K., 1968.
13 Rao, G. V., “Exhaust Nozzle Contour for Optimum Thrust,” Jet Propul-
often tie into a computer-aidedmanufacturingsystem, an accounting
sion, Vol. 28, No. 6, 1958.
system, or a data management system. Test recording, data analy- 14 Hageman, G., Immich, H., Nguyen, T. V., and Dumnov, E., “Advanced
sis, and data displays are now largely computerized. Engineering Rocket Nozzles,” Journal of Propulsion and Power, Vol. 14, No. 5, 1998,
analysis, such as heat transfer, plume behavior, or combustion vi- pp. 620–634.
bration, can all be helped by computer programs. The control of 15 Kinkaid, J. S., “Aerospike Evolution,” Threshold, an Engineer-
vendor performance, material traceability, and product quality is ing Journal, No. 18, Rocketdyne Propulsion and Power, Spring 2000,
routinely assisted by computers. Computer simulation has reduced pp. 4–13.
16 Strangeland, M. L., “Turbopumps for Liquid Rocket Engines,” Thresh-
the amount of testing and manufacturing that used to be required.
The way business is being conducted has changed radically during old, an Engineering Journal of Power Technology, No. 3, Rocketdyne
the 82-year LPRE history. Propulsion and Power, Summer 1988, pp. 34–42.
17 Ross, C. C., “Principles of Rocket Turbopump Design,” Journal of the
This author started in this LPRE business 60 years ago. It has
American Rocket Society, No. 84, 1951.
been exciting to watch the evolutionof this cutting-edgetechnology. 18 Minnick, A., and Perry, S., “Design and Development of an Advanced
These six decades have brought forth major new milestones and Liquid Hydrogen Turbopump,” AIAA Paper 98-3081, July 1998.
amazing achievements that a life time ago were believed to have 19 Yang, V., and Anderson, W. (eds.), Liquid Rocket Engine Combustion
been fantasy or impossible. The propelling of vehicles to the moon Instability, Vol. 169, Progress in Astronautics and Aeronautics, AIAA, Re-
and other space destinations,the harnessingof enormous propulsive ston, VA, 1995.
power, the routine handlingof hazardouspropellants,the dispensing 20 Oppenheim, B. W., and Rubin, S., “Advanced POGO Analysis for Liq-
of precise small impulse bits, and the rst ight through the sound uid Rockets,” Journal of Spacecraft and Rockets, Vol. 390, No. 3, 1993.
1006 SUTTON
21 Reardon, F. H., “Combustion Stability Speci cations and Veri cation 46 Product brochures and data sheets, Marquardt Co., Van Nuys, CA, 1990,
Procedure,” Publ. 247, Chemical Propulsion Information Agency, Laurel, Primex Aerospace Co. Redmond, WA, 1997, and General Dynamics Corp.
MD, Oct. 1973. Redmond, WA, 2001.
22 Lehman, M., This High Man, The Life of Robert H. Goddard, Farrar, 47 Morgan, O. M., and Meinhardt, D. S., “Monopropellant Selection
Strauss and Co., New York, 1963. Criteria—Hydrazine and other Options,” AIAA Paper 95-2595, June 1995.
23 Goddard, E. C., and Pendray, G. E. (eds.), The Papers of Robert H. God- 48
Schmidt, E. W., Hydrazine and its Derivatives, Preparation, Properties,
dard, Vol. 1, 1898–1924; Vol. 2, 1925–1937, Vol. 3, 1938–1945; McGraw– Applications, Wiley, New York, 1984 (2nd rev. ed. 2002).
Hill, New York, 1970, p. 1707. 49 Stechman, R. C., and Sumpter, D., “Development History of the Apollo
24 Goddard, E. C., and Pendray, G. E., Robert H. Goddard, Rocket De-
Reaction Control System Rocket Engine,” AIAA Paper 89-2388, July 1989.
velopment, Liquid Fuel Rocket Research, 1929–1941, Prentiss–Hall, En- 50 Murphy, J. M., “Liquid Rocket Propulsion System Advancements,
glewood Cliffs, NJ, 1948 (reissued 1961); also Goddard, R. H., “Liquid 1946–1970,” 14th History Symposium of the International Academy of As-
Propellant Rocket Development,” Smithsonian Miscellaneous Collections, tronautics, Sept. 1980.
Vol. 95, No. 3, 1936. 51 Iacobellis, S., “Liquid Rocket Engines, Their Status and Their Future,”
25 Goddard, R. H., “A Method for Reaching Extreme Altitudes,” Smithso-
AIAA Paper 66-828, 1966.
nian Miscellaneous Collections, Vol. 71, No. 2, 1919. 52 Vilja, J., “Rocketdyne Advanced Propulsion System Overview,” AIAA
26 Winter, F. H., Prelude to the Space Age, The Rocket Societies: 1924 to
1940, National Air and Space Museum, Smithsonian Inst., Washington, DC, Paper 97-3309, 1997.
53 Personal communications, Wheelock, V., Schuster, E., Minami, H.,
1983.
27 “The Flight of Experimental Rocket No. 2” Astronautics, No. 26, 1933, Halchak, J., and others, Rocketdyne, 1997–2003.
54 Data sheets or data folder on Lunar Module Ascent Engine, Redstone
pp. 1–11.
28 Pendray, G. E., “Early Rocket Developments of the American Rocket Engine, F-1, J-2, H-1, Atlas, Thor/Delta, Lance, and others, The Boeing Co.,
Society,” First Steps Toward Space, Vol. 6, American Astronautical Society, Rocketdyne, Propulsion and Power, Conogu Park, CA, 1987–2003.
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
55 Smith, R. A., and Minami, H. M., “History of the Atlas Engine,” edited
Univelt, San Diego, CA, Chap. 14; also Truax, R. C., “Annapolis Rocket
Motor Development,” First Steps Toward Space, Vol. 6, American Astro- by S. E. Doyle, American Astronautical Society History Series, Vol. 13,
nautical Society, Univelt, San Diego, CA, Chap. 27; also Durant, F. C., III, Univelt, San Diego, CA, 1992, Chap. 3.
and James, G. S. (eds.), First Steps toward Space, American Astronautical 56 Biggs, R. E., “Space Shuttle Main Engine—the First Ten Years,” edited
Society History Series, Vol. 6, 1985. by S. E. Doyle, American Astronautical Society History Series, Vol. 13,
29 Truax, R. C., “Liquid Propellant Rocket Development by the U.S. Navy
Univelt, San Diego, CA, 1992, Chap. 4.
during World War II: A Memoir,” American Astronautical Society History 57 Colbo, H. I., “Development of the Space Shuttle Main Engine,” AIAA
Series, Vol. 12, Univelt, San Diego, CA, 1991, pp. 326–332. Paper 79-1141, July 1979; also Harris, S. L., “Block II: The New Space
30 Malina, F. J., “The U.S. Army Air Corps Jet Propulsion Research
Shuttle Main Engine,” AIAA Paper 96-2853, 1996.
Project, GALCIT Project No. 1, 1939–1946: A Memoir,” 3rd History Sym- 58
Mattox, R. M., and White, J. B., “Space Shuttle Main Engine Con-
posium of the International Academy of Astronautics, 1969. troller,” NASA TP 1932, 1981, p. 19; also Seitz, P. F., “Space Shuttle Main
31
von Kármán, T., and Edson, L., The Wind and Beyond, Theodore von Engine Control System,” Society of Automotive Engineers, SAE Paper 73-
Kármán, Little Brown, Boston, 1967. 0927, 1973.
32 Summer eld, M., Powell, W., and Crofut, E., “Development of a Liquid
59 “RS-68 First Flight Edition,” Threshold, an Engineering Journal,
Jet Unit and its Operation on an A-20 Aiplane,” Jet Propulsion Lab., JPL No. 20, the Boeing Co., Rocketdyne Propulsion and Power, Winter
Rept. 1–13, California Inst. of Technology, Pasadena, CA, Sept. 1942, 2003; also Conley, D., Lee, N. Y., Portanova, P. L., and Wood, B. K.,
33 Shesta, J., “Reaction Motors, Inc., the rst Large Scale American
“Evolved Expendable Launch Vehicle System: RS-68 Main Engine De-
Rocket Company: A Memoir,” 12th History Symposium of the International
velopment,” 53rd International Astronautical Congress, Paper IAC-2.S.01,
Academy of Astronautics, Oct. 1978; also American Astronautical Society
Oct. 2002.
History Series, Vol. 19, Univelt, San Diego, CA, 1995, Chap. 7. 60 Sackheim, R. L., “TRW’s Family of High Performance Spacecraft En-
34 Ordway, F. I., III, and Winter, F. H., “Reaction Motors, Inc.: A Corporate
History, 1941–1958, Part I, Institutional Development,” American Astonau- gines: Status and Impact,” 1996 JANNAF Propulsion and Joint Subcommit-
tical Society History Series, Univelt, San Diego, CA, 1994, Chap. 7; also tee Meetings, Albuquerque, NM, Chemical Propulsion Information Agency,
Winter, F. H., and Ordway, F. I., “Part II, Research and Development Ef- Laurel, MD, Dec. 1996.
61 Personal communications, Dressler, G. A., Sackheim, R. L., and Hodge,
forts,” American Astronautical Society History Series, Univelt, San Diego,
CA, 1994, Chap. 8; also Ordway, F. I., “Reaction Motors Division of Thiokol K. F., TRW, 1997, 2000, 2001, and 2002.
62 Dressler, G. A., and Bauer, J. M., “TRW Pintle Engine Heritage and
Corporation, an Operational History, 1958–1972 (Part II),” American As-
tronautical Society History Series, Vol. 12, Univelt, San Diego, CA, 1991, Performance Characteristics,” AIAA Paper 2000-3871, July 2000.
63 TRW Data Sheet on Lunar Module Descent Engine, 1979, and on other
Chap. 11; also Winter, F. H., “Reaction Motors Division of Thiokol Cor-
poration, A Project History, 1958–1972 (Part III),” American Astronautical LPREs, Redondo Beach, CA, 1999.
64 Mayer, N. L., “Advanced X-Ray Astrophysics Facility—Imaging
Society History Series, Vol. 12, Univelt, San Diego, CA, 1991, Chap. 12.
35 Winter, F. H., “Black Betsy,” The 6000C-4Rocket Engine,1945–1989— (AXAF-I) Propulsion System,” AIAA Paper 96-2869, July 1996.
Part 2,” American Astronautical Society History Series, Vol. 17, Univelt, San 65 Chazen, M. L., “High Performance Bipropellant Rhenium Engine,”
Diego, CA, 1991, Chap. 11. AIAA Paper 98-3356, 1998.
36 Winter, F. H., and James, G. S., “Highlights of 50 Years of Aerojet, 66
Staudhammer, E. P., et al., “The Descent Engine for the Lunar Module,”
A Pioneering American Rocket Company,” History of Rocketry and Astro- AIAA Paper 67-521, 1967.
nautics, edited by P. Jung, American Astronautical Society History Series, 67
Natan, B., and Rahami, S., “The Status of Gel Propellants in the
Vol. 22, Univelt, San Diego, CA, 1998, Chap. 4. Year 2000,” Combustion of Energetic Materials, edited by K. K. Kuo and
37 Umholtz, P., “Liquid Rocketry,” Aerojet, The Creative Company, Aero-
L. deLuca, Begel House, Boca Raton, FL, 2001.
jet History Group, Sacramento, CA, 1995, Chap. 3. 68 Hodge, K. F., Crofoot, T. A., and Nelson, S., “Gelled Propellants for
38 Personal communications, Fisher, C. B., 2001; also Nguyen, T. V., 2002;
Tactical Missile Application,” AIAA Paper 99-2976, June 1999.
also Calhoun, D., and Mueggenburg, H. H., 1990s. 69 Mulready, D., Advanced Engine Development at Pratt & Whitney, the
39 Schnare, C. W., “Development of ATO and Engines for Manned Air-
Inside Story of Eight Special Projects, 1946 to 1971, Society of Automotive
craft,” American Rocket Society Paper 2088-61, 1961.
40 Ehresman, C. M., “Liquid Rocket Propulsion Applied to Manned Air- Engineers, Warrendale, PA, 2001.
70 Heald, D., “LH Technology was Pioneered on Centaur 30 Years Ago,”
craft in Historical Perspective,” AIAA Paper 91-2554, June 1991. 2
41 Stiff, R., “Storable Liquid Rockets,” AIAA Paper 67-977, Oct. 1961. American Astronautical Society History Series, Vol. 21, Univelt, San Diego,
42 Aerojet’s Data sheets (several) and Aerojet Manuals, Engines of Titan II CA, 1997.
71 Personal communications, Pignoli, C., Pratt and Whitney Aircraft,
(1983), Titan III SLV (1992), and the Titan IV (2000).
43
Meland, L. C., and Thompson, F. C.,“History of the Titan Liquid Rocket United Technologies Corp., 2001 and Mills, P., Pratt and Whitney, a United
Engines,” AIAA Paper 89-2389, June 1989; also Adams, L. J.,“The Evo- Technologies Co., 2003.
72 Tucker, J. E., “The History of the RL10 Upper Stage Rocket En-
lution of the Titan Rocket—Titan I to Titan II,” American Astronautical
Society History Series, Vol. 19, Univelt, San Diego, CA, 1997. gine,” edited by S. E. Doyle, American Astronautical Society History Series,
44 Limerick, C., et al., “On Overview of the K-1 Reusable Propulsion Vol. 13, Univelt, San Diego, CA, 1992, Chap 5; also Santiago, R.,“Evolution
System,” AIAA Paper 97-3120, July 1997. of the RL10 Liquid Rocket Engine for a New Upper Stage Application,”
45 Personal communications, Stechman, R. C., Galbreath, J. J., and AIAA Paper 96-3013, 1996.
Morgan, O., Aerojet Redmond Center, formerly General Dynamics, 2001, 73 Brown, J. R., “Expander Cycle Engines for Shuttle Cryogenic Upper
2002, 2003. Stages,” AIAA Paper 83-1311, 1983.
SUTTON 1007
74 Croteau, M. C., Garbowski, R. C., and Minick, A. B., “Live Fire Testing Satellite Dispenser 0.2 lbf Rocket Engine Assembly,” Technical Rept. Oct.
of the NPO Energomash RD-120 Rocket Engine at Pratt & Whitney,” AIAA 1973, revised May 1974.
Paper 96-2608, 1996; also Data sheets on the RD-170, RD-180, and RD-120 77 Saenz, A., Jr., “Life Performance Evaluation of 22 N (5 lbf) Hy-
engines, Pratt and Whitney Aircraft, United Technologies Corp., West Palm drazine Attitude Control Engines,” U.S. Air Force Rocket Propulsion Lab.,
Beach, FL, 1997. USAF/RPL Technical Rept. Nov. 1980.
75 Personal communications, Marvin, M., and Boorady, F., ARC, 2001, 78 E. Gribbon, E., Driscoll, R., and Marvin, M., “Demonstration test of a
2002, and 2003; also ARC brochures on several products, and company 15 lbf Hydrogen Peroxide (90%) Thruster,” AIAA Paper 2000-3249, 2000.
historical information. 79 Bloom, R., Davis, N. S., and Levine, S. D., “Hydrogen Peroxide as a
76 Marcus, M., “Historical Evolution of the Hydrazine Thruster Design,” Propellant,” Journal of the American Rocket Society, Vol. 80, March 1950.
AIAA Paper 90-1838, July 1990; also Polythress, C. A., and Conkey, G. H., 80 DeBrock, S. C., and Rudey, C. J., “Agena Primary and Secondary
III, “Development of a Long Life 125-lbf Hydrazine Thruster,” AIAA Paper Propulsion Systems,” Journal of Spacecraft and Rockets, Vol. 11, No. 11,
80-1170, 1980; also Francis, A. E., “Quali cation Test for the Multiple 1974, pp. 769–777.
Downloaded by WASHINGTON UNIV IN ST LOUIS on May 2, 2013 | http://arc.aiaa.org | DOI: 10.2514/2.6942
You might also like
- History of Liquid Propellant Rocket Engines in The United States PDFNo ratings yetHistory of Liquid Propellant Rocket Engines in The United States PDF30 pages
- Lecture 2 - Rocket Science & Engineering - Franklin's LecturesNo ratings yetLecture 2 - Rocket Science & Engineering - Franklin's Lectures18 pages
- A Survey of Automatic Control Methods For Liquid-Propellant Rocket EnginesNo ratings yetA Survey of Automatic Control Methods For Liquid-Propellant Rocket Engines22 pages
- Chapter 4-Electrothermal Propulsion-Resistojet & ArcjetNo ratings yetChapter 4-Electrothermal Propulsion-Resistojet & Arcjet93 pages
- Design and Analysis of A Fuel Injector of A Liquid Rocket EngineNo ratings yetDesign and Analysis of A Fuel Injector of A Liquid Rocket Engine7 pages
- Modern Engineering For Design of Liquid-Propellant Rocket Engines (PDFDrive) - 20-41No ratings yetModern Engineering For Design of Liquid-Propellant Rocket Engines (PDFDrive) - 20-4122 pages
- Liquid-Propellant Rocket Engine Throttling: A Comprehensive ReviewNo ratings yetLiquid-Propellant Rocket Engine Throttling: A Comprehensive Review61 pages
- Rocket Propulsion Primer: Subramaniam Krishnan Jeenu RaghavanNo ratings yetRocket Propulsion Primer: Subramaniam Krishnan Jeenu Raghavan427 pages
- Solid Propulsion For Space Application An Updated Road MapNo ratings yetSolid Propulsion For Space Application An Updated Road Map19 pages
- Liquid Propellant: Theoretical InvestigationNo ratings yetLiquid Propellant: Theoretical Investigation4 pages
- AMCP 706-282 Ballistic Missile Series Propulsion and Propellats PDFNo ratings yetAMCP 706-282 Ballistic Missile Series Propulsion and Propellats PDF107 pages
- Unit 4 - Chemical Rocket Propulsion Course MaterialNo ratings yetUnit 4 - Chemical Rocket Propulsion Course Material28 pages
- Sunder Deep Engineering College: Seminar PresentationNo ratings yetSunder Deep Engineering College: Seminar Presentation19 pages
- Technician Powerplant Text E Book JeppesenNo ratings yetTechnician Powerplant Text E Book Jeppesen36 pages
- AdditiveManufactuingPropulsionOverview GradlNo ratings yetAdditiveManufactuingPropulsionOverview Gradl40 pages
- Modern Engineering For Design of Liquid (001-122)No ratings yetModern Engineering For Design of Liquid (001-122)122 pages
- Modern Design of Liquid Propellant Rocket Engines (1992)100% (2)Modern Design of Liquid Propellant Rocket Engines (1992)431 pages
- Editorial-from-the-guest-editors - 2022 - FirePhysChem (١)No ratings yetEditorial-from-the-guest-editors - 2022 - FirePhysChem (١)3 pages
- SPC Cardboard Packer 30 Speed 3149409 0105No ratings yetSPC Cardboard Packer 30 Speed 3149409 01051,435 pages
- Power Measurement in Three-Phase SystemNo ratings yetPower Measurement in Three-Phase System11 pages
- Complete The Bihar Provincial Kisan Sabha 1929 1942 A Study of An Indian Peasant Movement 1st Edition Walter Hauser PDF For All ChaptersNo ratings yetComplete The Bihar Provincial Kisan Sabha 1929 1942 A Study of An Indian Peasant Movement 1st Edition Walter Hauser PDF For All Chapters82 pages
- Keeping Records of Materials, Tools and EquipmentNo ratings yetKeeping Records of Materials, Tools and Equipment12 pages
- Test Bank For Sociology 13th Edition by Macionis DownloadNo ratings yetTest Bank For Sociology 13th Edition by Macionis Download41 pages
- Political Philosophy of Thomas Hobbes & State of Nature100% (2)Political Philosophy of Thomas Hobbes & State of Nature22 pages
- FIRESSENSE Direct Low Pressure System LPCB ApprovedNo ratings yetFIRESSENSE Direct Low Pressure System LPCB Approved6 pages
- Skylark Tiny III User Manual: A Skylark Soars Above The CloudsNo ratings yetSkylark Tiny III User Manual: A Skylark Soars Above The Clouds11 pages
- Grade 10.00 Out of 10.00 (100%) : Question TextNo ratings yetGrade 10.00 Out of 10.00 (100%) : Question Text69 pages