AIAA 2009-4933
45th AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit
2 - 5 August 2009, Denver, Colorado
Supersonic Transport (SST) Engines
Mr. Daniel T. Jensen1
Rolls-Royce Heritage Trust, Indianapolis, IN 46206 USA
Mr. Gerald L. Brines2
Pratt & Whitney (Retired), South Glastonbury, CT 06073 USA
and
Dr. John C. Blanton3
GE Energy, Greenville, SC 29615 USA
During the 1960s and 1970s much work was done to develop viable supersonic transport
(SST) engines by a number of manufacturers. Rolls-Royce developed the Olympus engine
for the Concorde, while General Electric and Pratt & Whitney competed to power the larger
and faster U.S. SST. This historical paper reviews technical highlights of SST engine
development for the three manufacturers during this time period.
Nomenclature
ATC
DOC
HP
IGV
PNdb
RFP
SST
TBO
TSFC
VSV
=
=
=
=
=
=
=
=
=
=
air traffic control
direct operating cost
high pressure
inlet guide vane
perceived noise decibel
request for proposal
supersonic transport
time between overhauls
thrust specific fuel consumption
variable stator vane
I. Introduction
E
VER since Captain Charles “Chuck” Yeager became the first aviator to demonstrate supersonic flight in 1947,
interest in eventual commercial supersonic flight has grown. As airlines began to transition from propeller
planes to first-generation jetliners such as the Boeing 707, McDonnell Douglas DC-8, Convair 880 and 990 and
deHavilland Comet, aircraft manufacturers began to explore the feasibility of the next logical step in commercial
aviation – the supersonic transport (SST).
Europe, seeing an opportunity to reestablish dominance in aircraft manufacturing, launched the Concorde, to be
powered by Rolls-Royce Olympus engines. The Soviet Union launched its own SST, the Tupolev TU-144.
Refusing to cede such a prestigious and valuable market to others, the United States launched its own SST
development program in 1963. Pratt & Whitney and General Electric competed for the contract to provide the
powerplants for the American SST, which was to be larger and faster than both the Concorde and TU-144.
The substantial technical advances required to develop a viable powerplant for a supersonic transport are worth
documenting. Each of the three major engine manufacturers, Rolls-Royce, Pratt & Whitney and General Electric,
overcame difficult technical obstacles in order to enable the first generation of supersonic transports. While the
Pratt & Whitney and General Electric engines did not ultimately see revenue service, their technological
achievements are well worth remembering.
1
Allison Branch Secretary, P. O. Box 420, Indianapolis, IN 46206-0420, Associate Fellow.
Consultant, 11 Overshot Drive, South Glastonbury, CT 06073-2231, Associate Fellow.
3
Consulting Engineer, GE Energy, 300 Garlington Rd., Greenville, SC 29615-4615, Associate Fellow.
1
American Institute of Aeronautics and Astronautics
2
Copyright © 2009 by Gerald L. Brines, General Electric Company and Rolls-Royce Heritage Trust. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.
II. Rolls-Royce
The engine that would ultimately power the Concorde had its origins in 1946 at Bristol Aero Engines Ltd. when
the first speculative layouts were completed. Initially, the engine was conceived as the powerplant for a bomber
which would fly at 600 mph and 40,000 feet, not a supersonic civil transport. Originally intended for both the
Bristol Aeroplane Type 172 bomber and the Avro Vulcan, construction of six prototypes was authorized in 1947.
An initial setback for the Olympus was the cancellation of the Bristol Aeroplane Type 172, but momentum was
regained on the project when Dr. Stanley Hooker (Fig. 1) joined Bristol Aero Engines Ltd. from Rolls-Royce in
1949 and championed Bristol’s work on turbine engines. The first Olympus (Model B.01.1) ran in 1950, achieving
9,140 lb. of thrust with a sfc of 0.83 lb/lb thrust/hour and weight of 3600 lb.
Figure 1. Dr. Stanley Hooker. Photo
courtesy of Rolls-Royce Heritage Trust.
Development of the Olympus in the early 1950s for the Avro Vulcan steadily increased its thrust from that of the
B.01.1. The next model, B.01 1/2 achieved 9,500 lb. and the B.01 1/2A reached 9,750 lb. This model was an early
candidate for the Convair F102 and was provided to Curtiss-Wright for testing in a fixed nacelle under a B-29
testbed during 1951.
The Olympus 99 was a derated (8,000 lb thrust) version which was flown in an English Electric Canberra B2
(Fig. 2) beginning in 1952. The Canberra with Olympus engines set a new world altitude record of 63,668 feet in
1953. The flight experience gained by this aircraft assisted the Bristol engineers in readying the engine for its
intended airframe, the Vulcan.
2
American Institute of Aeronautics and Astronautics
Figure 2. English Electric Canberra B2
with Olympus 99. Photo courtesy of RollsRoyce Heritage Trust.
The first model intended for the Vulcan, the Olympus 100, first went to test in 1952. Later that year, the first
production model, the Olympus 101 (Fig. 3), received its airworthiness certificate with a rating of 11,000 lb thrust.
This model weighed 3,650 lb and had a Time Between Overhauls (TBO) of 250 hours.
Figure 3. Olympus 100 for Avro Vulcan. Illustration courtesy of Rolls-Royce Heritage
Trust.
In 1953, flight clearance was obtained for the Olympus 100 rated at 9,250 lb. thrust and the Olympus-powered
Vulcan was first flown (Fig. 4). The Vulcan was put into service by the Royal Air Force in 1956, and was finally
retired in 1984.
During the 1950s, significant interest emerged in adapting the Olympus for supersonic operation. This initiated
some investigations into adding afterburning (“reheat”) both internally and with Solar Aircraft Company of San
Diego, California. Additionally, collaboration was begun with French manufacturer Snecma on thrust reversers and
noise supressors. Even with an afterburner and variable nozzle, the Olympus was relatively light-weight and high
thrust for its time.
3
American Institute of Aeronautics and Astronautics
Figure 4. Avro Vulcan. Photo courtesy of
Rolls-Royce Heritage Trust.
Bristol Aero Engines Ltd. was already pursuing the international market during this time. The Olympus Model
551 was conceived as a candidate powerplant for the Boeing 707, with proposed manufacture by American aircraft
engine company Curtiss-Wright. The Olympus 551 was derated to 13,500 lb. for the 707 application but the
proposal was not successful. Commercial development of the Olympus was initially hampered by its use in the
Vulcan bomber and the associated export restrictions.
The breakthrough needed to push the Olympus into the supersonic regime came with the British Tactical Strike
and Reconnaissance (TSR2) solicitation in 1957. The TSR2 (Fig. 5) was to be the replacement for the Canberra,
with a mission requirement which included a maximum cruise at altitude of Mach 2.05. The British Ministry of
Supply, however, did not feel that a single domestic company could manufacture either the aircraft or the engine,
and made it clear that the contract would be awarded only to merged firms – forcing consolidation in the British
aerospace industry.
Figure 5. TSR2. Photo courtesy of RollsRoyce Heritage Trust.
In response, Vickers of Weybridge and English Electric Warton joined together and absorbed the Bristol
Aeroplane Company to become British Aircraft Corporation and subsequently received the TSR2 contract.
Similarly, Bristol Aero Engines Ltd. merged with Armstrong Siddeley Motors of Coventry to form Bristol Siddeley
Engines Ltd. in 1959. The company was then selected to provide the Olympus Mark 320 engines (Model B.01 22R)
for the TSR2.
The Olympus Mark 320 (Fig. 6) represented the first Olympus with a cooled turbine, more advanced materials
and an afterburner, all required to make supersonic operation of the Olympus feasible. Rated at 30,610 lb thrust
4
American Institute of Aeronautics and Astronautics
with afterburning, the engine first flew in the TSR2 during 1964. By the program’s cancellation in 1965, the
Olympus Mark 320 had completed 6,000 hours of testing including 800 hours with afterburner operating, 61 hours
on the Vulcan flying test bed, and 26 hours in the actual TSR2 prototype.
Figure 6. Olympus Mark 320.
courtesy of Rolls-Royce Heritage Trust.
Photo
In the early1960s, interest began to grow in Europe in a supersonic civil transport aircraft. Under consideration
in Britain at the time was the Bristol Type 198, which specified six Olympus 593/3 engines rated at 28,000 lb of
thrust as well as the Bristol 223 which required four engines of the same type. France was proposing a “Super
Caravelle” which also specified the Olympus as its powerplant. It soon became apparent that a partnership would
help spread development risk and costs, and the Concorde project was born.
The initial Olympus model under development for the Concorde was designated the Mark 593/3 or 593D (for
“Derivative”), rated at 28,800 lb. First run in July of 1964, the 593D featured an added compressor stage, turbine
cooling and a variable nozzle. It had soon become apparent, however, that in order to achieve adequate range, the
Concorde would need more powerful engines. Olympus Model 593B (for “Big”) provided an increased thrust rating
of 32,800 lb dry and 35,000 lb with afterburner.
Challenges for the powerplant associated with supersonic cruise necessitated novel design solutions. Fuel was
used to cool the engine oil, hydraulic fluid, constant speed drive and alternators prior to being sent to the burner for
combustion. In the worst-case scenario, the fuel entered the burner already at 150 deg C. Thermal management
issues placed special requirements on engine fuels and oils for satisfactory viscosity at temperature characteristics.
While development of the Olympus for the Concorde was well underway, Bristol Siddeley Engines Ltd. agreed
to be acquired by Rolls-Royce in 1966. As part of Rolls-Royce, development continued in earnest, leading to a
successful first flight of Concorde on March 2, 1969.
The initial Concorde engine production standard was the Olympus 593-602 (Fig. 7). Compared to the 593B, the
602 replaced the cannular combustor with an annual combustor design, incorporated improved turbine cooling,
added squeeze film damping on all five main bearings, redesigned the IGVs to provide a 5% increase in mass flow,
and incorporated improvements to the thrust reverser and fuel system.
5
American Institute of Aeronautics and Astronautics
Figure 7. Olympus 593 on Display in the James A.
Allison Exhibition Hall, Rolls-Royce Heritage Trust,
Indianapolis, Indiana. Photo courtesy of Rolls-Royce
Heritage Trust.
In 1973, a further enhancement, Olympus 593-610 was announced. The 610 incorporated control system
improvements and was rated at 32,000 lb thrust dry and 38,050 lb thrust with afterburner. First deliveries of the
Concorde went to Air France in December, 1975 and British Airways in January, 1976 (Fig. 8). Airline service was
initiated by both carriers on January 21, 1976.
Figure 8.
Concorde with four Olympus 593 engines in
British Airways livery. Photo courtesy of Rolls-Royce Heritage
Trust.
Further improvements were proposed but never realized. The Olympus 593-621 was to be rated at 38,275 lb
with afterburner and was planned to be incorporated beginning with the 41st aircraft. The Olympus 593-631 would
have been rated at 41,360 lb with afterburner by adding a zero stage to the compressor and redesigning the HP
spool.
Ultimately, only fourteen Concordes entered airline service and were all retired in 2003. The Concorde’s
Olympus engine fleet had accumulated 930,000 flight hours, more than 500,000 of which were supersonic. The
experience gained in the development and operation on Olympus will be used to enable future high-speed flight in
the 21st century.
6
American Institute of Aeronautics and Astronautics
III. Pratt & Whitney
A. Cycle Selection
In June of 1960, the House Committee on Science and Astronautics recommended that Congress support a
Federal program for the development of a commercial supersonic transport (SST). Numerous analytical studies
were conducted by the U.S. aircraft and engine community during 1961 and 1962. Then, in November, 1962,
Britain and France signed an agreement to develop the Mach 2 Concorde SST. The following June, President
Kennedy announced his support for a government/industry partnership to develop a SST superior to that being built
in any other country. In August, 1963, the Federal Aviation Administration (FAA) issued an RFP for SST engine
and aircraft designs, thus launching the formal U.S. SST program. The FAA received extensive technical support
from experts at NASA and Air Force Research Centers.
At Pratt & Whitney (P&W), Phases I, II, IIA and IIB of the FAA SST program were conducted at the main
engineering center in East Hartford, Connecticut. The 18 month Phase IIC demonstrator engine program was
conducted at P&W’s Florida Research and Development Center in West Palm Beach, home of the J58 engine for
Lockheed’s Mach 3.2 YF-12A/SR-71 Blackbird. This paper will discuss the highlights of the Phase IIA and Phase
IIC programs.
The objective of the Phase IIA program was to arrive at an airplane-engine combination which could meet the
overall performance and economic requirements of the SST. P&W’s efforts in Phase IIA were concentrated in two
areas: (1) design and performance studies conducted in close cooperation with Boeing and Lockheed to determine
the optimum powerplant cycle and design and (2) a comprehensive component test and analysis program aimed at
verification of component design and performance. These efforts resulted in a new turbofan design incorporating
more advanced state-of-the art components and having a development objective of considerably higher turbine inlet
temperatures than the Phase I powerplants.
The Phase IIA effort at P&W was planned to utilize maximum latitude in ingenuity and explored many
promising alternatives. A major portion of the design study effort consisted of a re-examination in greater depth of
all engine cycles that appeared attractive for a SST cruising in the range of Mach 2.7 to 3.0. The results of this study
indicated that one cycle, the duct-heating turbofan, with an airflow size of approximately 640 to 700 lbs/sec could
best meet the overall requirements of the SST in the Phase IIA aircraft designs.
An objective assessment was made of two different turbofan cycles and three different turbojet cycles. The
engines included: a turbofan with duct heater augmentation, a turbofan with common afterburner augmentation, a
turbojet without augmentation, a turbojet with full afterburning augmentation and a turbojet with a low
augmentation afterburner. Every effort was made to incorporate the same advanced levels of technology in all of
the various engine components to ensure that the comparisons would indicate the cycle best suited to meeting the
performance and economic objectives of the SST. The process of optimizing an engine for lowest airplane direct
operating cost (measured in cents per seat mile) necessitated the creation of several “paper” SSTs of the required
range, passenger capacity and take-off capability designed around a number of engine configurations embodying a
full range of the variables being optimized.
The total operating cost to fly a 3,500 nautical mile trip was then
calculated for each hypothetical airplane. P&W’s SST studies based on such a method indicated: (1) that
unaugmented engines were clearly not competitive with the augmented turbofan or turbojet; (2) that cost, although
relatively insensitive to overall pressure ratio, is minimized at a design pressure ratio of 9:1; (3) that an optimized
duct-heating turbofan can operate at lower costs than an afterburning turbojet at the same turbine temperature; and
(4) that a fan pressure ratio of 2.5 is near optimum.
In addition to the clear advantage in direct operating cost shown for the duct-heating turbofan, two other
characteristics which confirmed its choice for the SST were revealed in the P&W studies. The first of these was the
turbofan’s appreciably lower noise levels. Fig. 9 shows a clear cut advantage for the duct-heating turbofan over the
afterburning turbojet in a fixed sweep aircraft both in reduced airport noise and reduced community noise. Also, a
substantial degree of noise suppression is inherent in the design of the blow-in-door exhaust ejector. This important
reduction in noise was achieved without performance loss, weight penalty or additional mechanical complexity. The
second major advantage of the turbofan over the turbojet was its superior subsonic performance. Installed TSFC of
7
American Institute of Aeronautics and Astronautics
the duct-heating turbofan was approximately 12 to 14 percent lower at Mach 0.9, 36,000 feet and 10 to 15 percent
lower at Mach 0.6, 15,000 feet than the afterburning turbojet. Thus, the turbofan powered aircraft would be better
able to cope with an emergency which might require a portion of the scheduled supersonic trip to be flown at lower
altitudes and flight speeds.
Figure 9. Noise Comparison Between
Afterburning Turbojet and Duct-Heating
Turbofan.
The P&W studies also revealed certain intrinsic advantages of the turbofan cycle over the turbojet cycle beyond
the considerations of DOC: (1) To be competitive with the turbofan, the turbojet must operate at higher
temperatures and thus represents a greater development risk; (2) The turbofan provides a more rapid rate of thrust
growth as advancing technology makes higher turbine temperatures feasible; (3) the turbofan offers a greater
operational flexibility, suffering less range loss from ATC altitude changes and from operating at above standard
temperatures and under one engine-out conditions; and (4) The turbofan characteristically operates at lower external
case temperatures, a potential factor in safety.
The five study engines were also provided to the airframe contractors for their use in studying each cycle in an
optimized engine-airframe combination to form a basis for their own cycle selection. After an extended period of
mission studies, utilizing all of the turbojet and turbofan engines supplied by P&W, both Boeing and Lockheed
selected the duct-heating turbofan as the optimum P&W cycle for their Phase IIA designs. The optimization
considered the airplane-engine combinations with respect to the parameters of range-payload, flight speed, take-off
and landing conditions, sonic boom overpressures, noise levels and overall utility and flexibility. In Lockheed’s
fixed wing aircraft studies, the duct-heating turbofan was clearly superior to the other fan and jet cycles. Boeing
also selected the duct-heating turbofan for Phase IIA, although the superiority of this cycle was not as clear cut in
Boeing’s variable sweep wing configuration. The basic operational concept of the duct-heating turbofan at take-off,
subsonic cruise and supersonic cruise is shown in Fig. 10.
8
American Institute of Aeronautics and Astronautics
Figure 10.
Turbofan.
Operational Concept of the Duct-Heating
The final effort in Phase IIA consisted of detailed work to optimize the duct-heating turbofan engine selected by
Boeing and Lockheed to the requirements of each installation. Intensified effort was expended in coordinating the
areas of inlet compatibility, exhaust system optimization (including secondary ejector airflow), mount design, fuel
and lubrication interfaces, accessory drive and air bleed requirements.
B. Engine Description
The Pratt & Whitney STF219 engine was a twin spool duct-heating turbofan powerplant. The low pressure rotor
consisted of two fan stages driven by a two-stage turbine. The high pressure rotor consisted of a five-stage
compressor driven by a single-stage turbine. The dual rotor design permitted use of a higher bypass ratio and higher
compression ratio without the added complexity of variable compressor geometry. This optimization permitted
higher compressor and turbine efficiencies and better component match than was obtained with single rotor designs.
Each rotor was supported on two bearings. A full annular primary burner incorporating advanced features to
provide high heat release rate and uniform temperature distribution was utilized. The duct heater incorporated a
low-loss aerodynamic type flameholder to provide high efficiency at both maximum and partial augmentation
conditions. A variable area nozzle in the fan duct heater exit provided engine airflow scheduling under augmented
and non-augmented conditions. The excellent subsonic, transonic and supersonic performance characteristics of the
blow-in-door ejector nozzle assured optimum performance throughout the flight range while also acting as an
exhaust nozzle suppressor. A reverser was provided which stowed conveniently in the ejector when not in use. The
design characteristics of the STF219-B (Boeing) and STF219-L (Lockheed) duct-heating engines are tabulated in
Table I and Table II, respectively. The basic (or ultimate) engine performance is given at 2200 F cruise turbine inlet
temperature and 2300 F turbine inlet temperature for max take-off and transonic acceleration. In addition, the initial
certification engine performance is also presented for 1900 F cruise and 2000 F for max take-off and transonic
acceleration.
9
American Institute of Aeronautics and Astronautics
Table I. Design Characteristics of STF219-B (Boeing)
10
American Institute of Aeronautics and Astronautics
Table II. Design Characteristics of STF219-L (Lockheed)
C. Turbine Inlet Temperature
The studies by the airframe contractors showed that the range-payload objectives stated in the Phase IIA
contracts required turbine inlet temperatures as high as 2200 F for cruise operation. The STF219 turbine design had
as its objective the ultimate attainment of a turbine inlet temperature of 2300 F for take-off and transonic
acceleration and 2200 F for supersonic cruise in commercial service operation. This was an extremely ambitious
goal and was, in P&W’s opinion, by far the greatest technical risk in the propulsion area of the SST program.
11
American Institute of Aeronautics and Astronautics
Successful commercial operations dictated the requirement for a high degree of reliability so that maximum
utilization of the aircraft would be realized. In order to make an SST airplane competitive on an economic basis
with subsonic commercial transports operating in the same time period, it was necessary to assume that the SST
would be available approximately the same number of hours per day, would be reliable, would have comparable
TBO’s, and would be as cheap to maintain and overhaul on a relative basis. These assumptions depend primarily on
the SST powerplant having approximately the same order of reliability as the commercial subsonic jet engines
which were being operated at turbine temperature levels of 1450 – 1500 F for cruise and 1700 – 1750 F for take-off.
For the above reasons, it was P&W’s firm conviction that a design and development program aimed at 2300 F
turbine inlet temperature for take-off and acceleration and 2200 F for supersonic cruise would be no more than
sufficient to assure acceptable engine durability at the initial airline service operating temperatures of 2000 F for
take-off and acceleration and 1900 F for supersonic cruise. The initial use of operational turbine temperature levels
considerably below the design level would provide the durability required to obtain the utilization necessary for
successful initial passenger carrying operations. Turbine inlet temperature levels would be increased to the 2300
F/2200 F level as soon as airline operating experience and the continuing engine development program showed that
it was practical to do so.
D. JTF17A-21 Demonstrator Engine
The cross section of the P&W demonstrator engine, designated the JTF17A-21, is shown in Fig. 11. The duct
heater augmentation system, the variable area convergent nozzle and the blow-in-door ejector nozzle and reverser
were new to commercial transport applications. However, significant development experience and military
operational experience would have been accumulated prior to their entry into service on the SST.
Figure 11. JTF17A-21 Demonstrator Cross Section
The duct heater, while new to commercial transport engines, was expected to be less of a durability challenge
than the high performance afterburners on which military operational experience had been accumulating for over ten
years. While the duty cycle in the SST would require operation of the duct heater augmentation system during
supersonic cruise, the level of temperature during cruise would be substantially reduced below the maximum
temperature used for transonic acceleration (average temperature was reduced at least 600 F, from 3100 F at
maximum to 2250 F - 2500 F at cruise). Relatively low temperature cooling air from the fan stream would further
contribute to duct heater durability. Accordingly, the duct heater durability was not expected to be a limiting factor
on engine or aircraft utilization. The development risk was considered less than for a high performance afterburner
which would normally depend on high temperature turbine discharge gas or relatively high temperature compressor
bleed air for cooling.
12
American Institute of Aeronautics and Astronautics
The variable area fan duct exhaust nozzle was also new to commercial transport engines, although again more
than ten years military experience had been accumulated to date. The variable area nozzle was required for cycle
optimization and airflow scheduling to meet the airflow demands of the supersonic inlet to minimize spillage and
installation drag. As in the duct heater, relatively cool fan duct air was used for cooling the nozzle flaps. Actuator
cylinder cooling was accomplished by the fuel used as the actuating fluid.
The reverser design achieved relatively modest loads by pressure balancing the aerodynamic forces and relieved
the temperature durability requirements by stowing the reverser in a cool environment when not is use.
The design of the JTF17A-21 engine drew heavily on the experience gained in the J58 engine development and
from the Mach 3.2 cruise flight experience in the YF-12A and SR-71 aircraft. Operation in these aircraft provided
considerable insight into the effects of high Mach number operation. The design and materials selection for the SST
powerplant, and in particular for the hot section components, benefited from the experience gained in the J58
program.
Certainly the most powerful factor affecting development risk was the high level of turbine inlet temperature
required. In view of the lack of operating experience at the temperature levels which the Phase IIA aircraft studies
had shown to be required for the SST, it was concluded that a major portion of the engine development program
focus had to be in the area of attaining commercial durability and reliability at temperatures where current
experience had been limited to short runs on test stands in a simulated high Mach environment. The 2200 F cruise
temperature requirement for the SST was 150 F hotter than the 2050 F cruise temperature of the military J58. Fig.
12 shows the projected improvement in supersonic cruise TSFC as a function of years from initial commercial
operation as turbine temperature was increased from 1900 F to 2200 F and component performance was improved.
Figure 12. Projected Improvement in Supersonic Cruise TSFC
Phase IIC tests and analysis indicated that the following FAA noise objectives could be met:
1. Airport take-off at 1500 feet from runway centerline
116 PNdb
2. Community flyover 3 statute miles from start of take-off 105 PNdb
3. Approach 1 statute mile from runway threshhold
109 PNdb
Prediction of the unsuppressed and suppressed engine noise at the three FAA noise certification points is shown in
Figures 13, 14 and 15.
13
American Institute of Aeronautics and Astronautics
Figure 13. Predicted Noise: Airport Take-off at 1500 ft from Runway Centerline
Figure 14. Predicted Noise: Community Flyover 3 Statute Miles from Start of Take-off
14
American Institute of Aeronautics and Astronautics
Figure 15. Predicted Noise: Approach 1 Statute Mile from Runway Threshold
An interesting operational advantage of the duct-heating turbofan concept was the ability to tailor the noise
sources at the community flyover condition to minimize the noise signature. Specifically, the operation of the duct
heater offered three potential matching advantages. First, by providing more thrust from the fan duct exhaust,
primary thrust could be reduced and the required overall engine thrust still maintained. Jet noise produced by the
primary and fan duct jets can be matched closely by this method. Second, by reducing the primary thrust output, fan
speed was also reduced with attendant reduction in fan noise. Third, by applying this technique during power
cutback after take-off, the range between fan speeds at this flight condition and at airport approach will be
minimized. Since acoustical liners must be designed for a limited range of fan speeds to provide optimum noise
attenuation, this approach permits the design of one, more optimum liner for both flight conditions. The results of
an analysis of the noise attenuation value of this technique for typical flight conditions following power cutback at
take-off are shown in Fig. 16.
15
American Institute of Aeronautics and Astronautics
Figure 16. Noise Attenuation Value of Optimum Liner Design
Astroloy turbine and compressor disks performed most successfully in the J58 engine. This experience
demonstrated the outstanding capability of the alloy and substantiated its suitability for turbine disk application for
the JTF17 engine. Structurally, the first stage turbine disk was almost identical to the J58 engine first stage turbine
disk in size, shape and manufacturing technique. A comparison of the J58 and JTF17 turbine disks is shown in Fig.
17.
Figure 17. Comparison of J58 and JTF17 Turbine Disks
16
American Institute of Aeronautics and Astronautics
Similarly, as shown in Fig. 18, the JTF17 first stage turbine blade bears a close resemblance to the J58.
However, to ensure the turbine life necessary for a commercial engine, the JTF17 engine blade metal temperature
was initially designed to be 150 F lower at cruise conditions than that for the J58 engine. At the time of the SST
proposal, the development engine test time on the J58 at maximum rated turbine inlet temperature was over 10,000
hours with only 500 hours at or above 2100 F. This was in addition to substantial in-service experience.
Figure 18. Comparison of J58 and JTF17 First Stage Turbine Blades
Significant achievements that were demonstrated on the JTF17A-21 experimental engines:
•
•
•
•
•
•
•
•
•
Maximum thrust during a transient run
Maximum steady state thrust
Take-off TSFC (4% below spec)
Total engine time
Total duct-heating time
Time at 2000 F and above
Time at 2200 F and above (M 2.7, 65,000 ft.)
Heated inlet time
Time at cruise conditions (M 2.7, 65,000 ft.)
58,800 lbs.
57,830 lbs
1.74
154.55 hours
27.88 hours
56.65 hours
14.27 hours
32.34 hours
28.59 hours
In addition, engine noise testing was accomplished with the ejector-suppressor installed and a reduction of 5
PNdb below that predicted by the SAE method was demonstrated which confirmed the level used for the
specification values.
Fig. 19 is a photo of the J58 engine on the left next to the JTF17 demonstrator engine.
17
American Institute of Aeronautics and Astronautics
Figure 19. J58 and JTF17 Demonstrator Engine
Fig. 20 is a photo of the JTF 17 demonstrator engine with the ejector/suppressor nozzle installed.
Figure 20. JTF17 Demonstrator Engine with Ejector/Suppressor Nozzle Installed
18
American Institute of Aeronautics and Astronautics
Fig. 21 is a photo of the JTF17 at maximum duct-heating without the ejector/suppressor installed.
Figure 21. JTF17 Demonstrator Engine at Maximum Duct Heating without the Ejector/Suppressor Nozzle
E. Epilogue
At the end of Phase IIC, Lockheed selected the P&W duct-heating turbofan for their fixed wing design. Boeing
selected the General Electric afterburning turbojet for their variable sweep configuration. The FAA stated that the
engine competition would be based solely on the initial certification engines offered. Therefore, only the 2000
F/1900 F performance for the P&W duct-heating turbofan could be considered. GE’s initial certification engine was
close to the 2300F max turbine temperature that P&W was promising only years later. In December, 1966, the FAA
announced that Boeing’s variable sweep wing design had won the aircraft competition and that the General Electric
afterburning turbojet had won the engine competition. Twenty-two months later, in October, 1968, Boeing
announced that critical design issues related to the variable sweep wing could not be solved and that they were
abandoning the variable sweep concept and switching to a fixed wing design. By then, it was too late for the ductheating turbofan.
As a result of the SST competition, all future gas turbines would have cooled turbines and the figure of merit for
risk had changed from the turbine entry gas temperature to the turbine blade metal temperature.
19
American Institute of Aeronautics and Astronautics
IV. General Electric
The General Electric Company traces its initial involvement in jet engines to the development work on turbosuperchargers by Dr. Sanford A. Moss, who joined GE in 1903. GE became a major producer of turbosuperchargers, delivering over 370,000 units in a period of over 30 years through the 1940’s, winning a Collier
Trophy in 1938 and a spot in the National Aviation Hall of Fame for Dr. Moss. GE was selected by the U.S. Army
Air Corps in 1941 to develop the first U.S. jet engine based upon the British W1 design of Frank Whittle. This led
to the GE I-A 1250 lb-thrust engine, which powered the Bell XP-59A aircraft on the first U.S. jet-powered flight on
02 October 1942. GE has been a major developer and producer of jet engines ever since, for both military and
commercial aviation.
The advent of jet propulsion enabled aircraft to achieve flight speeds well in excess of the speed of sound. In the
early 1950’s GE developed the landmark J79 turbojet, the first Mach 2 jet engine. The high thrust-weight ratio of
this engine resulted from the successful implementation of variable stator vanes (VSVs) in the compressor. The
16000 lb-thrust J79 powered a number of well-known high-speed aircraft including the B-58, the F-104, and the F-4.
Over 17000 J79 engines were produced, and it won GE another Collier Trophy in 1958.
In the late 1950’s GE was selected by the U.S. Air Force to develop a Mach 3 capable engine for the B-70
program. This engine was designated the J93 turbojet and incorporated a number of new innovations, including
actively-cooled turbine blades with STEM-drilled blades and an exhaust nozzle with independently-variable throat
and exit areas. Six of the 28000 lb-thrust J93 engines powered the XB-70. Although only two prototype aircraft
were eventually built and flown, the flight test program demonstrated Mach 3 flight beginning in Oct 1965.
SST engine development at GE began formally in 1958 in the form of a study engine designated the X279M, a
version of the highly successful J93 turbojet. The X279M was an engine in supersonic military transport studies
performed by a number of airframers in the late 1950’s and early 1960’s.
President John F. Kennedy announced that the US would pursue the development of a supersonic transport (SST)
effort in a speech at the USAF Academy on 5 June 1963. The program was managed by the US Federal Aviation
Administration (FAA), led by Najeeb Halaby. In 1967, responsibility for the SST program transferred to the newlyformed Department of Transportation. The SST special project office (SPO) was managed by Gordon Bain, Bill
Maxwell, and finally Bill Magrauder. Two airframers (Boeing and Lockheed) and two engine companies (GE and
Pratt Whitney) competed for the SST contract award.
GE named Ed Hood (Fig. 22) General Manager of the SST project. The engine was designated the GE4. It
would be a pure turbojet engine leveraging many of the advanced technologies developed for the J93 engine, which
was by that time in flight test. Key among these technologies was techniques for cooling the turbine gas path
components to permit turbine inlet temperatures of approximately 2300 deg F, or about 500 degrees higher than the
current practice. GE submitted its formal proposal to the FAA SPO in September 1966 and on December 31 it was
announced that the GE-powered Boeing 2707 (see Fig 23) would receive the contract to develop the US SST. The
GE4 engineering team was led by John Pirtle (Fig 24).
20
American Institute of Aeronautics and Astronautics
Figure 22. Ed Hood, Manager of GE’s
SST Project. Photo courtesy of the
General Electric Company.
Figure 23. Boeing 2707-300 SST Mockup. Photo courtesy of the Boeing Company.
21
American Institute of Aeronautics and Astronautics
Figure 24. John Pirtle with GE4. Photo courtesy of the General Electric Company.
In order to meet the ground test needs of the GE4 program an altitude test cell was constructed at the Evendale,
Ohio engine plant. This cell was capable of simulating flight conditions of 80,000 ft altitude and Mach 3 flight
speed (Fig 25). The cell is still in use today.
Figure 25. GE4 Installed in the Altitude Test Cell. Photo courtesy of the General
Electric Company.
22
American Institute of Aeronautics and Astronautics
Some of the other features of the GE4 engine are:
− A low-aspect ratio, distortion-tolerant compressor with conservative stage loading and variable stators
throughout, and featuring hollow blades and vanes.. The front and rear sections were ganged and actuated
separately as demanded by the engine control.
− A conservative compressor exit diffuser.
− A full-annular combustor to provide a more-uniform exit temperature distribution, and eliminating crossfire
and transition components.
− A two-stage turbine with advanced air-cooling and with a choked first-stage nozzle to trap/attenuate transient
pulses.
− A conventional afterburner (the largest ever built) and an advanced exhaust nozzle with independent primary
and secondary nozzle area control and an integrated thrust reverser.
− A conventional three-bearing, three-frame support system for precise clearance control.
− A fast-response control system to manage the complex interactions between the airframe, inlet, engine, fuel
system, and exhaust nozzle.
− Control and accessories components enclosed in a capsule cooled by the circulation of fuel.
The GE4 was (and is) the most powerful turbojet engine ever built. Four of the giant engines were to power the
Boeing 2707. On test it produced 69,900 lb of sea level static thrust. Figures 26 through 31 show additional details
and historical images.
Figure 26. J93 (Top) with GE4. Photo courtesy of the General Electric Company.
23
American Institute of Aeronautics and Astronautics
Figure 27. Two J79s Used as Gas Generators for Early GE4 Afterburner Test.
Photo courtesy of the General Electric Company.
Figure 28. GE4 Exhaust Nozzle at Full Afterburner. Photo courtesy
of the General Electric Company.
24
American Institute of Aeronautics and Astronautics
Figure 29. GE4 Exhaust Nozzle. Photo courtesy of the General Electric Company.
Figure 30. GE4 at Full Thrust on Sea Level Test Stand. Photo courtesy of the General
Electric Company.
25
American Institute of Aeronautics and Astronautics
Figure 31. GE4 at Full Thrust on Outdoor Test Stand. Photo courtesy of the General
Electric Company.
Table III summarizes the design specifications of the GE4.
Table III. Design Characteristics of the GE4
Comments
The GE4 is an axial-flow, augmented turbojet for the Boeing 2707 Supersonic Transport (SST). It
incorporates a nine-stage, variable-stator, axial compressor directly coupled to a two-stage, axial, air-cooled
turbine. The design objectives are to provide a workhorse engine based on proven technology, long life,
reliability, and ease of maintenance consistent with the high standards established by the airlines. The
engine is optimized for Mach 2.7 cruise.
Features
Compressor
axial, 9-stg, VSVs Combustor
annular
Max press ratio SLS
12.7 Turbine
2-stage, axial
Max air bleed
2 per cent Turbine Cooling
Air
Max airflow SLS
633 lb/sec Max rated TIT SLS
2280 F
Performance
SFC
TIT/TET (deg
AIRFLOW
Thrust
RPM
Rating
(lbm/hr/lbf)
F)
(lb/sec)
(lbf)
Max (wet)
68,600
5200
1.770
2280/1700
633
Max (dry)
51,500
5200
1.040
2280/1700
633
Idle
3170
Size and Weight
Length, overall
296 in.
Weight (dry)
13,243 lb
Diameter, max
90 in.
26
American Institute of Aeronautics and Astronautics
On 29 March 1971 the US Senate, by a vote of 49 to 48, cancelled the US SST program. Many blame concerns
about the environmental impact of the SST, but there were also serious concerns about the economic viability of the
program as well. The effect upon GE was quite significant, resulting in the loss of about 1600 jobs.
A total of three GE4 test engines were built prior to the cancellation. One of these engines is currently stored at
the GE factory in Evendale OH, and the other two are in the collection of the Smithsonian National Air and Space
Museum in Washington DC.
V. Conclusion
The international effort to develop engines which would enable a supersonic transport was fruitful, even if the
intended aircraft never flew or entered revenue service. Technologies developed in the SST engine programs
included cooled turbines, advanced materials, integral afterburners, thermal management, high-temperature fuels
and oils, advanced controls, engine/aircraft integration and noise suppression.
These technologies, which were developed by all three manufacturers, have been further developed and
implemented into a variety of both commercial and military aircraft engines. The experience gained by the
operation of the Olympus in the Concorde and subsequent service of P&W and GE powerplants incorporating SSTdeveloped technology lay the groundwork for the next generation of supersonic commercial flight.
References
A. Rolls-Royce
Baxter, A., Olympus – the First Forty Years, Rolls-Royce Heritage Trust, Derby, England, 1990.
John, D. J., “The Olympus 593 Powerplant for Concorde,” Aircraft Engineering, April 1967.
B. Pratt & Whitney
PWA FR-1855, 10 May 1966. “Development of a Supersonic Transport Engine, Phase IIC”
“The Pratt & Whitney Aircraft JTF17 Augmented Turbofan Engine for the SST” (Brochure)
PWA 64-544, 1 November 1964, Volume IE Summary, Phase IIA
PWA 66-100, August, 1966, Volume I, Design and Development, Phase IIC
PWA FR 2239, 10 January 1967, Monthly Progress Report, Phase IIC
Boeing Model 2707, Volume II-14, Propulsion Report – Part C, Engine Evaluation, 6 September 1966
C. General Electric
Blanton, J. C. and D. C. Wisler, “A Brief History of Aircraft Engine Development at General Electric”, AIAA
Paper 2007-5339 (2007).
Eight Decades of Progress: A Heritage of Aircraft Turbine Technology, General Electric Co., (1990).
Pirtle, John C. and B. W. Bruckmann, “Requirements of a Commercial Supersonic Aircraft and its Engines”,
AIAA Paper 67-376.
St. Peter, James, “The History of Aircraft Gas Turbine Engine Development in the United States: A Tradition of
Excellence”, ASME (2000).
27
American Institute of Aeronautics and Astronautics