AIAA 93-3955

Design of an Airborne Launch

Vehicle and an Air Launched Space
Booster
S. M. Poth, M. A. Fisher, D. W. Levy
and J. G. Eisley
University of Michigan
Ann Arbor, MI

AlAA
Aircraft Design, Systems
and Operations
Meeting
Auaust 11-13,1993 / Monterey, CA
For permlssbn to copy or republish, contact the Amerkan Instituteof Aemnautlcs and AatronautlCS
370 L'Enfant Promenade, S.W., Washlngton, D.C. 20024
 DESIGN OF AN AIRBORNE LAUNCH VEHICLE AND AN
AIR LAUNCHED SPACE BOOSTER

v Stefan M. Poth, Jr.'

Michael A. Fisher**
David W. Levy?
Joe G . Eisleyt
University of Michigan
Department of Aerospace Engineering
Ann Arbor, Michigan

Abstract In the future, there may be a market for an air launched

space booster an order of magnitude larger. A space
A conceptual design is presented for an air launched booster of this size would be capable of placing a heavy
space booster and an associated carrier vehicle. This payload into low earth orbit (LEO), say, for space station
system is capable of placing 17,000lbs in low earth orbit resupply or one or more satellites into geosynchronous
(LEO)and 7,900 lbs in geosynchronous transfer orbit transfer orbit (GTO).Such a large booster would require a
(GTO). The airplane features a twin fuselage new airplane to be purposely built for its launch since no
configuration for improved payload and landing gear current airplane could carry a space booster of this sue to
integration, a high aspect ratio wing for high lift and any significant altitude.
maneuverability at altitude, and is powered by six engines.
The aircraft weighs approximately 1.2 million pounds at Class History a n d Interactions
take off. It releases the booster at approximately 43,000
feet. The booster weighs just under 500,000Ibs and has In the fall of 1992,the University of Michigan AE 481
three stages. A primary design goal was to have launch Airplane Design class undertook the design of an airplane
costs well below those of competitive launch systems: to carry a large air launched space booster. A preliminary
thus, off the shelf technology has been used throughout. design was formulated by each of the 35 studencs in the
class. The space booster weight was set arbitrarily at
Introduction 250,000 pounds as a base for a feasibility study. The
class then split into two groups to continue the design
One way of reducing the total weight of a space booster process, with one group pursuing a conventional design
for a given payload weight is to launch the booster from and the other group pursuing a more unusual twin
an airplane. Among the advantages are that the kinetic fuselage design. The payload capacity was increased to
and potential energy of the airplane are added to that of the 500,000 pounds when the individual design projects
space booster and the launch takes place above a proved that to be a feasible goal.
substantial portion of the atmosphere so that aerodynamic
drag is reduced. One current design which takes advantage In the winter term, seven students elected to continue
of this approach is the Pegasus, manufactured by Orbital this design project under AE 490/590Advanced Airplane
Sciences Corporation. The Pegasus has a total weight of Design, a directed study class with the goal of presenting a
41.000 pounds and a payload weight of 900 pounds. It more fully defined design at the Universities Space
has been canied aloft and launched by a Bwing B-52and Research Association (USRA) Advanced Design Program
future launches are planned from a modified Lockheed (ADP) Annual Conference in Houston in June 1993.
L-1011. Simultaneously 40 students in AE 483 Space System
Design began the design of a 500,000 Ib booster to be
used with the evolving airplane. The two classes
* Undergraduate Student, Aerospace Engineering, AIAA coordinated their work through the end of the winter tern
Student Member After evaluating a conventional airplane design with the
** Undergraduate Student, Aerospace Engineering payload suspended beneath the fuselage and, briefly a
t Lecturer. Aerospace Engineering: currently flying wing, the final configuration chosen was a twin
Aerodynamics Engineer, Cessna Aircraft Company, fuselage airplane with the space booster suspended beneath
AIAA Senior Member the center wing. It proved by far the best for landing gear
J. Professor, Aerospace Engineering, AIAA Associate arrangement and payload integration with the booster that
+u Fellow was being developed in the AE 483 class.
 i

Fig. 1 . System Configuration.

The booster size and configuration were driven by the for the launch system is anticipated as this would reach
performance and size of existing rocket engines. In early until the introduction of the next planned launch systems.
attempts the AE 483 class fried for an all solid fuel Only two airplanes would be built. The first would be a
system; however, it proved impossible to achieve. The fully functional aimaft while the second would act as a
final booster design uses threedifferent rocket engines and sauctural span.
three different fuels.
The final configuration of the system is presented in
Overview Figure 1. W
Mission requirements include: Some of the imponant parameters are tabulated in Table
1. Thetwin fuselage configuration of the airplane, from
Airmli now on called the Eclipse. allows for sm'ghtfonvard
mission radius of 750 statute miles payload and landing gear integration. The high aspect
launch altitude of at least 40,000 feet ratio wing was necessitated by the need for conhulled
the ability to use existing airpon facilities flight at high altitude. Thm an six GE-90engines to
the ability to perform 2.2g post-launch maneuver at provide the thrust meded to attain altitude. Two vertical
altitude tails are positioned with the horizontal tail between them
in a cruciform layout, This configuration keeps the tails
Eixtsta out of the wake of the wing, payload, and engines while
17,000 Ibs in LEO consistent with the requirements keeping the horizontal tail below an area susceptible to

. for space station resupply

7,900 Ibs in GTO consistent with the requirements
of typical communicariom satellites
deep stall.

The booster, from now on called the Gryphon, consists

of three stages for the GTO configuration. For the LEO
configuration the thud stage is removed and replaced with
. &kal
reliance on components, subsystems, and processes
currently used in production or which will be in the
p u n payload, The major components are shown in
Figure 2. The first stage consists of two Castor 120
very near fume solid fueled engines attached to the sides and one LR-91in
commercially competitive launch costs the main body using storable liquid fuel. The second
* safety in assembly, handling, and launching stage has two LR-91 engines and the third stage one
near term delivery of complete system RLlOA-4 using cryogenic fuel.

FuU details on che Eclipse and the G

Market studies concluded that six launch missions per
year would be a realistic design goal. A ten year life span in the two USRA ADP final repom. IYhon are provided L
 Gryphon pitches back down before the second stage
ignition to an angle of 70 degrees from the vertical.

The fmt stage engines bumout at 130,397 feet and a

velocity of 7,297feet per second. Second stage ignition
accelerates the Gryphon to a circular parking orbit at
574,240 feet altitude and a velocity of 24,864 feet per
second. The payload shroud is jettisoned along the way at
an aItitude of 200,ooO feet. Once the parking orbit is
reached the second stage is ejected and the Gryphon o h i u
until it reaches the proper position for insertion in GTO.
Finally, the third stage ignites and GTO insertion is
completed.

Table 1. Eclipse Paramem.

Take off weight 1,227,000 Ibf

Payload weight (v. of Mich. Gryphon) 479,000Ibf
Fuel weight (including ramp fuel) 241,ooO Ibf
Operational weight empty 541,000 Ibf
wing area 11,750 ftz
wing span 368 li
Overall length 188.6 ft
Overall height 62.4 ft
Take off thrust (wlo exhactions) 600.000 Ibf
Thrust-to-weight ratio 0.489
Take off field length (to 35 ft) 4,300ft
Landing field length (from 35 ft) 3.400 ft
wing aspect ratio 11.53
Wing quarter chord sweep angle 22.10
wing loading 104.4 IbfIft2
Vertical taiI area (each) 950 ft2
Horizontal tail area 3700 ft2
Cruise velocity 515 mph
Launch altitude (design mission) 43,300ft

A
Fig. 2. Booster InternaI Components.

Mission Profile a n d Trajectory

The design mission profile for the system is shown in

2
/
/--7wF, 4
7
\ a
Figure 3. The aircraft takes off and climbs to an efficient J
1
cruise altitude and cruises for up to 750 miles to an
1. Take off 6.Cruise
predetermined launch location. Upon approaching the
2 Climb 7. Descent
launch site after a short loiter it drops the booster at 3.750micruise 8. Prelanding loiter
43,300feet. It immediately turns away to clear the space
4.Prelaunch loiter 9.Landing
for the booster, returns to cruise altitude, and returns to 5.Launch maneuver
base.
A. Fitststageignition C. LEO
The Gryphon is dropped from the Eclipse at 43,300feet B. Second stage ignition
and during an ensuing 10 second drop the aerodynamic
shape cause a 20 degree pitchup. At this time the fmt Fig. 3. Airborne Mission Profde.
stage engines ignite and a pitch rate of 6.25 degrees per
'bd' second is obtained by gimballing the engine nozzles until
the pitch angle reaches 10 degrees from the vertical. me
 Propulsion a leading edge sweep of 25'. This raises the critical Mach .'
number to 0.78, or roughly the cruise speed of the
The engine selected for the Eclipse is a planned growth airplane. The vertical tails are NACA 0012 airfoils while
version of the General Eleceric GE-90 which will be rated the horizontal tail is a NACA 0008 airfoil for the inboard
at 100,ooO pounds take off thrust. Six of these engines section, to raise critical Mach number where there is no .J
'
will be used to provide the thrust needed to attain altitude. leading edge sweep, and a NACA 0010 airfoil outboard of
The growth version of the GE-90 was selected because it the vertical tails. The outboard sections have a leading
is the only engine which will be commercially available edge sweep of 40°. This results in a higher critical Mach
in the near future which meets the thrust requirements for number for the horizontal tail than the wing, as required
this airplane. The high lapse rate of thrust with altitude for stability and control purposes.
makes the airplane highly overpowered at take off and
during a large portion of the climb. Thrust specific fuel Drag polars for the airplane were found using semi-
consumption and uninstalled thrust available (includin empirical method^.^ Along with this, the interference
ram drag) data was supplied by the manufacturer. f drag from the payload, i.e. the space booster, was
Installed thrust levels were obtained using empirical accounted for by adding ten pemnt of the wing drag to the
models for estimating extractions.4 area which is directly affected by the flow around the
payload. An additional five percent is added to the final
Nacelles were sized using semi-empirical methods based drag number to account for interference in the rest of the
on experience and wind tunnel t e ~ t i n g .The
~ engines are airp1ane.7
located with a diameter and a quaner between each engine
and from the fuselage. They are mounted below and in A single slotted Fowler flap with an area of 1346.4 ft2,
front of the wing. This location was chosen to minimize covering 30 percent of the chord and from 23 to 32 and 38
the losses associated with installation and therefore to 70 percent of the half span, is utilized to lower the take
improve altitude performance. off and landing speed to acceptable levels. No leading
edge devices are employed. For take off, CLmU = 2.06
Various combinations of existing rocket engines and with a flap deflection of 20' and for landing, C L ~ S=
staging were studied. Those identified in Table 2 were 2.65 with a flap deflection of 45'.
rejected for the reasons shown.
Structures & Weights
Table 2. Rejected Configurations.
A V-n diagram was constructed to determine the loads
Configuration Drawback acting on the aircraft during flight Between the maneuver
All solid fuel Not enough payload, Too heavy load and gust induced load plots, the maximum loads on
No cryogenic fuels Not enough payload, Too heavy the airplane can be deduced. These showed a limit load of
Cryogenic Stage 2 Safety concern 2.5g which implies, with a factor of safety of 1.5, a
Extra Stage Too exwnsive ultimate load of 3.758.

The configuration finally selected is identified in Table 3. The majority of component weights for the airplane
were found using semi-empirical methods.8 These values
Table 3. Propulsion Configuration were calibrated using a production aircraft part of similar
dimensions and configuration. Engine and nacelle weights
System Fuel were supplied by General E l e ~ t r i c . ~Wing weight was
Stage
calculated using an analytical method developed
1 2 Castor 120 Solid
specifically for twin fuselage a i r ~ r a f t . ~This method
1 1I%l-A.J-11 Storable liquid
calculates the structure required to resist the shear and
2 2 LR91-AI-11 Storable liquid
bending moments associated with the diseibuted and point
3 1 RLlOA-4 Cryogenic liquid
loads acting on the wing. Semi-empirical models are then
used to give weights for non-load bearing surfaces and
This final design meets the payload and cost goals trailing edge devices. The analytical method results in a
originally selected. wing group weight percentage of gross take off weight
which is in line with that of current production airplanes.
Airplane Aerodynamics
The longitudinal center of gravity was calculated by
The wing airfoil is a NASA 14-percent thick assigning each component a center of gravity and then
supercritical airfoiL6 A supercritical airfoil was chosen to calculating the weighted average. All measurements are
minimize wave drag and therefore increase cruise speed. ma& from a reference datum 50 feet in front of the forward
To further increase the critical Mach number, the wing has most point of the wing. The payload and fuel centers of L
gravity are collocated with the airplane center of gravity to the large tails needed for an acceptable static margin at
minimize center of gravity shifts during the flight and cruise.
launch maneuver.
The Critical condition for lateral and directional stability
The Gryphon uses conventional booster stuctural design and control is a one engine inoperative condition on take
condepts. In general each stage has the following: off. This criteria sized the vertical tails. The two engine
inoperative condition was also analyzed. Meeting the hu0
Enginemounts engine out criteria required either a very high speed or
* Propellant tank supports excessively large vertical tails to maintain control. Since
Interstage connections meeting this criteria would be detrimental to airplane
External skin with reinforcements performance, the one engine out condition was selected as
the design condition.
Ease of manufacture and assembly were primary concern
throughout. Booster Guidance, Navigation, and Control

Actual weights of engines, propellant tanks and other The launch team must be able to keep track of the
components from existing systems were used. Detailed Gryphon's position, velocity and acceleration in order to
design and structural analysis of engine mounts, interstage determine what attitude controls need to be implemented.
connections, payload shroud, and external body were The following choices have been made:
carried out to determine both strength and weights
requirements. Details may be found in the fmal report2 Mission Control. The Gryphon will utilize the existing
ground support infrastructure of Pegasus.
Airplane Stability & Control
Telemetry, Tracking, and Command. The telemetry
Stability and control calculations are based on semi- tracking and command ( l T C ) services from the Eastern
empirical method^.^ Longitudinal stability calculations and Westem Space and Missile Centers with be used.
focused on three areas. First, the airplane must be able to
rotate about its main gear for take off. This did not size Inertial Measurement Unit. Inertial reference is supplied
the horizontal tail. It did, however, size the elevators. by the strapdown inertial measurement unit (MU),
Second, the airplane must be trimmable during cruise. consisting of integration gyroscopes, linear
Last, the airplane must meet a static margin requirement accelerometers, and sensor electronics. The Litton LR-81
during cruise. The third criteria sized the horizontal tail. system will be used.
Ideally, the tail would be positioned such that all three
criteria would be met simultaneously. This is not the Global Positioning System Receiver. The Trimple
case with the Eclipse and the tails are therefore larger than Quadrex GPS Receiver was chosen.
they might otherwise be. This is acceptable since the
shorter fuselages which are associated with the larger tail Launch Panel Operator. One additional crew member
allow for a shorter landing gear to meet the tip over and aboard the Eclipse will be needed to monitor the
rotation requirements. Also, fuselage weight and tail Gryphon's system before and immediately after launch.
weight are almost an equal trade off so there is a minimal
weight penalty associated with the larger tail surface with On Board Computer. The onboard computer system
sholter fuselages. interfaces with the subsystems and determines the course
of action that they should take.
Take off rotation and trim calculations were estimated
by summing the moments acting on the airplane. In the Gimbaled engines are the main source of attitude
fust case, by summing around the median point of the conwl; however, passive aerodynamic characteristicswere
main gear contact points and in the second case by used during the drop phase
summing around the center of gravity. The landing trim
requirement was not evaluated. Static margin calculations Airplane Performance
are done through the use of an x-plot. The design is for
five percent static margin. This value was chosen as a Airplane performance was calculated for three mission
point where there is sufficient margin for safe pilot scenarios: the normal launch mission, the minimum fuel
control without making the tails extremely large. With mission, and the ferry mission. The minimum fuel
only a five percent static margin, a stability augmentation mission consists of a climb to maximum altitude, loiter,
system will be necessary to ensure controlled operations launch, and descent. The lack of fuel for the cruise
of the airplane. Static margin at take off is high due to portions results in a higher launch altitude. The ferry
mission consists of flying the maximum distance possible For safety, four levels of redundancy are employed with f
with the empty space booster. each of the primary flight control surfaces (aileron, rudder,
and elevator) with each section of the conaol surface
The calculations for the two launch missions assume a powered by three separate systems. Secondary control
launch abort at the last possible moment, which is the systems employ only two levels of redundancy. The high 'J
worst case scenario due to the extra drag and weight of the level of redundancy is called for since only two airframes
booster which is attached for the return portion of the are being built and any hull loss would be devastating to
mission. Calculations were based on analytical models of the program.
airplane performance.l0 Upon launch, a minimum radius
turn is conducted at constant altitude. This maneuver ElecUical System. The elechical system for the Eclipse is
should give m i m u m separation between the vehicles in sized using the requirements of an airplane of similar
a given amount of time. dimensions, the Boeing 747. This is done because the
actual elecuical loads of the systems are not known until a
Take off and landing distances are the distance to clear a more detailed design for the airplane systems is canied
35 foot obstacle from or to a complete stop, respectively. out.
The final values for this analysis are presented in Table 4.
Three engine driven generators are used for normal
Table 4 Design mission performance airplane operations. There is an extra generator connected
to the auxiliary power unit and an emergency generator
Take offdistance 4.300 ft connected to a free-fall ram air turbine. Batteries are
Landing distance 3,400 ft incorporated into the system to maintain DC system
Launch altitude 43,300 ft voltage under transient conditions, to supply power for
Launch turn load factor 2.2g short term heavy loads, and to supply power in an
Launch turn radius 9200 ft emergency. In addition, there is a ground connection
Mission time 6 hr 55 min point for ground operations.

Airplane Subsystems Flight Control System. A mechanical signaling system

is used for the flight contmls in this aircraft. AU flight
Various subsystems of the airplane are now briefly conaols are irreversible. A mechanical signaling system
&scribed was chosen, despite the weight penalty and higher
maintenance, to avoid the cost and difficulty associated W
Landing Gear. The landing gear is a quadracycle with the development of either a fly-by-wire or flyby-
configuration. This layout allows for all six gear mcks light control system.
to fully retract into a fuselage without interfering with the
payload. The main gear consists of four shuts, each with Two separate mechanical systems are employed for
an eight wheel landing gear truck in a dual-twin-tandem redundancy with the primary flight controls, while the
configuration. Each nose gear is steerable and has three secondary controls use only one level of redundancy.
wheels in a triple configuration. The main gear retract
fonvard and aft and the nose g e a r r e m a aft. Fuel System. The fuel system has a capacity of 260,000
pounds in the primary tanks. This value is a direct result
The main gear are located such that the airplane can of the performance calculations for the design mission. In
rotate about the median point between the m c k s by addition, there are auxiliary tanks with a capacity of
extending one shock absorber and compressing the other. 90.000 pounds which brings total fuel capacity to
The struts are long enough to ensure that the airplane 350,000 pounds. This extra fuel volume is used when
meets all clearance and tip-over requirements. The nose fenying the airplane, either with or without the payload,
gear are located so that a sufficient amount of the airplane from place to place. All tanks are located in the outboard
weight rests upon it to allow for effective steering. section of the wing between the fust and third spars.

Hydraulic System. A conventional hydraulic system is Pumps are sized for one and a half times the maximum
employed. It operates at a pressure of 5,000 psi. This fuel flow. The pumps have a through flow of 473 pounds
was chosen as a compromise between volume and weight per minute. The lines are two inches in diameter. Surge
requirements of a lower pressure system and the sealing tanks are located outboard of the main tanks.
requirements of a higher pressure system. In addition, the
5,000 psi system is the standard used on large commercial In order to keep the center of gravity of the fuel
transports today. constant, several fuel cells are used with flow valves and
pumps which are computer controlled to maintain a
constant center of gravity.
 Crew Issues. The crew of the Eclipse consists of thrre Gryphon Assembly and Integration
people, the pilot, copilot, and a launch officer. The crew
are in the forward section of the left fuselage in a The System used to put the Gryphon together and attach
pressurized compartment. The pilot and copilot are on a it to the Eclipse include:
v raised platform to enhance visibility with the launch
officer behind them. A galley and lavatory are also Gryphon assembly building (GAB)
provided for crew comfort during the mission. Transportation and aetachment of completed booster
* Attachment of Gryphon to Eclipse
A ground based simulator is needed for this airplane to
maintain pilot proficiency. Since this airplane will only Provision is ma& for two parallel assembly lines. The
be used once every other month for missions, the pilots schedule calls for up to one vehicle every two weeks, or
need other flight time to remain proficient. This could be two vehicles in close succession if launch windows require
attained through use of the airplane itself for training. but it.
this risks loss of the airplane. Instead a six degree of
freedom trainer could be used to train the pilots for flight Once assembled and check out a transportation trailer
and to keep them proficient in the airplanes flight takes the Gryphon to the Eclipse. It is capable of moving
qualities. One other option is available for training, the the Gryphon so that attachment points line up. Once
flying simulator. This concept was rejected due to the correct alignment is achieved the hydraulic interface
cost associated with maintaining the aircraft. mechanism will close and securely attach the Gryphon to
the Eclipse.
Payloads
The location of the facility was based on the availability
The system was designed to meet the following criteria: of rocket fuels on site, proximity to the equator for GTO
launches, distance from large population Centers, and the
Delivery of 7,900 lb, including the payload support availability of a 10,ooO foot runway.
structures, to GTO
Delivery of 17,000 Ib, including the payload support The interface mechanism has eight attachmentment
structures, to LEO points located on the second stage of the booster,
Maximizing of usable payload envelope symmetric a b u t the center of gravity.
W
Capability of multiple satellite deplooyments to both
GTO and LEO Cost Analysis
Compatibility of delivering Space Station Freedom
(SSF)resupply packages The airplane cost analysis was performed using
empirical methods.ll The price was determined in 1993
The GTO delivery limit will allow the booster to carry dollars. Since only two airplanes are being built and the
a large majority of existing commercial communication weight of the Eclipse is so large, the precision of these
satellites. The LEO capability will allow for the delivery values remains uncertain.
of scientific satellites and resupply packages to a space
station. The payload bay dimensions are compatible with The cost to acquire the two airplanes is estimated at
currently planned SSF payload packages which are IS feet $1.72 billion. This includes all costs associated with the
in diameter and IO to IS feet in length. Communications design, testing, manufacturing, flight testing, profir, and
satellites cover a large range but many are 7 to 10 feet in financing. This cost is much higher than buying an
diameter and 8 to 12 feet in length. Single, double, and, airplane such as the Boeing 747, but this is due to the
possibly, aiple stackedpayloads can be accommodated. small number of planes over which the development cost
is spread.
Thermal Control
The cost of the Gryphon was determined by the known
Thermal control of the Gryphon is concerned primarily cost of existing components such as the engines and
with the external structure and the avionics bay. The propellant tanks and the estimated cost of standard
external structure uses ablative coating during the ascent consauction techniques used for new components such as
of the booster. The avionics bay uses a multi-component the payload shroud and main body. The cost of the
system including a helium purge, a heat sink radiator, Gryphon is estimated at $22.1 million per copy.
enamel coatings, and multilayer insulation.
Throughout this design study a primary goals was to
develop a system that could provide investors with a 15%
r e m on their money. Every decision, consistent with
the safety and integrity of the project, was made to try to
achieve a launch cost 50% below the chief competitors. used in designing this system have been outlined. The 3'

This would leave 50% for financing, insurance, profits payload integration, landing gear, hydraulic, electrical,
and competitive pricing. flight control, fuel, and crew systems designed for the

The direct cost per mission for a 60 mission life time is

Eclipse have also been outlined. In addition. the results of
wind tunnel testing have been presented and discussed ,
'v

an estimated $28 - 30 million. This includes the

expendable booster and the operating costs. When the It is technologically feasible to design an aircraft which
initial cost of the Eclipse is spread over 60 missions the can cany an 500,000 pound air launched space booster 10
total launch cost rises to over $50 million. A higher an altitude over 40,000 feet and launch it. The stumbling
number of missions could lower this cost substantially. block, however, is the initial cost associated with
This increased number of missions could be accomplished acquiring the airplanes. The per mission cost can be ma&
in several ways: by increasing the number of missions per reasonable, if a large enough number of missions is
year, by lengthening the life span, or by fmding alternate flown.
uses for the airplane. Finding alternate uses would be a
last resort since it risks loss of the airframe. Acknowledgments

The cost estimates in the two final reports show some Following the references is a complete list of students,
differences; they had not yet converged when the deadline teaching assistants, and faculty who contributed to this
(end of the term) was reached; however, by all estimates project Other persons who contributed are acknowledged
the cost per pound to orbit, which is $6200 for the 7,9W in the two final reports.
lb payload to GTO, is competitive with existing launch
systems and has sufficient margin to cover intangibles and References
make a profit for investors. Unfortunately, the initial cost
of producing the airplanes requires such a large up front Poth, S.M., Jr., et al., Desisn of an Airborne Lnunch
investment that it may make the program infeasible. Vehiclefor an Air Launched Space Boosfer,University of
Michigan, Ann Arbor, Mich., July 1993.
Wind Tunnel Testing Fisher, MA., et al., Projecf Gryphon: Air Launched
Space Boosfer, University of Michigan, Ann Arbor,
A 1:120 scale model of the Eclipse was tested in the Mich., June 1993.
University of Michigan Sx7' subsonic wind tunnel. The
model was tested with and without the Gryphon attached.
3 Lewis, R.J., Sr., Big Engine Data, General Electric
Aircraft Engines, Evendale, Ohio, Unpublished Nov. 16,
w
The tunnel was run at two speeds, 120 and 150 miles per 1992.
hour. This data was then scaled to Re = 3 9 . 4 ~ 1 0for
~ Roskam, J.. Airplane Design Pari VI: Preliminary
comparison with the analytical data Calculation of Aerodynamic, Thrusf, and Power
Characferisfics,Roskam Aviation and Engineering Corp.,
The clean predicted and test drag polars are similar. Ottawa, Kansas, 1989.
Unfortunately, due to the scale of the model, there was 5 Berry, D.L., Civil Aircrafi Propulsion Infegrafion:
very bad interference between the Gryphon and the Currenf and Fufure, Boeing Commercial Airplane
Eclipse. This led to a much worse zero Lift drag for the Company, Seattle, Wash., Unpublished, Nov. 12, 1992.
model than was anticipated. Due to the low speed of the 6 Harris,C.D., "Aerodynamic Characteristics of a 14-
test, the assumption of zero angle of attack for the payload Percent-Thick NASA Supercritical Airfoil Designed for a
was not valid. The model lift curve slope, however, is in Normal-Force Coefficient of 0.7," NASA-Langley
good agreement with the analytical value. Research Center, Hampton, Virginia, NASA TM X-
72712, July 1975.
Conclusion 7 Bengelink, R., Boeing Commercial Airplane Co.,
Seattle, Wash., Personal Conversation, Feb. 1993.
A design has been presented outlining an airborne 8 Roskam, J., Airplane Design Part V: Component
launch vehicle and associated large air launched space Weight Esfimafion, Roskam Aviation and Engineering
booster. The airplane has a twin fuselage layout for Corp., Ottawa, Kansas, 1989.
payload and landing gear integration. The wing has a high 9 Udin, S.V., and Anderson, W.J., "Analytical Method
aspect ratio for increased altitude performance. The three for Wing Mass Calculation of Twin Fuselage Aircraft,"
stage booster is of conventional configuration and design. Journal of Aircrafi, Vol. 29, Sept.-Ocr. 1992, pp. 907-
This design meets or exceeds all of the goals which were 914.
set at the stan of the design process. l o Vinh, N.X., Aerospace 440 Vehicle Systems
Performance Unpublished Lecture Notes, University of
The propulsion, aerodynamics, smctures and weights, Michigan, Ann Arbor, Mich., Jan.-May 1993. L
stability and control, and performance analysis methods
 _-

l 1 Roskam, J., Airplane Design Parr VIII: Airplane systems Scott Cohen
Cosf Esfimafion:Design, Development, Manufacfuring. Integration group: Brett Deutscher
and Operaling, Roskam Aviation and Engineering Corp., Andrew Rotten
Ottawa, Kansas, 1989. Mauricius Gibin
L/ ". I Brian Nehez
Class Roles
AE 490/590. Advanced Airplane Design
AE 481. Airplane Design
Faculty: David W. Levy
Faculty: David W. Levy Graduate Teaching Assistant: John Blow
Graduate Teaching Assistant: John Blow Students:
Students: Chin Chao
Rich Choi
MaizeTeam Scoa Cohen
Brian Dumont
TeamLeader Rich Choi Mauricius Gibin
Assistant: Michael Fisher Robert Jorden
Stefan Poth'
Aerodynamics group: Kevin McGuirk
Robert Mitchell' AE 483. Space System Design
Thuc Nguyen
Faculty: Joe G.Eisley
Suuctures and Stephen Battle* Graduate Teaching Assistant James A k a
Weights group: Guan Soh Students:
Mauhew Wagner
Project leader: Lyon King
Stability and Benjamin Place Assistant project leader: Michael Fisher
Control group: Tracy Scott*
Payload team: Christopher Bernard
U
Propulsion Frank Gulczinskil James Dice
Integration group: Robert Jorden Karina Jacobson'
Charla Widmer Craig Litherland
Kevin Whalen
Systems Chin Chao
Integration group: Enrique Garcia Mission Analysis: Richard Draper
Scot May Sean Fifeld
Matthew Taylor' Scott Mullison
Ryan Waddington Alan Ristow,
Vincent Wiltse
Blue Team
Propulsion: Krista Campbell*
Team Leader: Stefan Poth ChadHoggard
Assistant: Jack Kobus Adam Nagaj
Bilal Rathur
Aerodynamics group: Brian Dumont* John Vandenberg
Jeffrey Feutowski
Cynthia Tan suuctures: Scoa Huggins
Joseph Ruddy*
Structures and Rolf Kappe* Wolfgang Schubert
Weights group: Kevin Kilbum Ronald Shimshock
Ran- Stehouwer Philip Wojcik

Stability and Michael 01* Mission Control Kah-Wai Aw

Control group: Sarah Lewis and Communications: Scott Egbert'
Brian Smith
Propulsion Kevin Williamson* Christopher Yee
Integration group: Bilal Rathur
Power and Thermal: Joshua Baron
Thomas Godfroy
Kevin Kilbum
Josef Regnu*
Romy Sharieff

Spacecraft Integration: Timothy Ballew

David Coxtwright
Elizabeth Hilbert'
Todd MueUer
Daniel Potter

&raft Interfa: Jonathan Albert

Michael Hmdley'
Adam Koziel
Christopher Vegter

Teamleader