CONCEPT ASSESSMENT OF A HYDROCARBON FUELED RBCC-POWERED MILITARY SPACEPLANE

Transcription

1 CONCEPT ASSESSMENT OF A HYDROCARBON FUELED RBCC-POWERED MILITARY SPACEPLANE J. E. Bradford *, J. R. Olds ±, and J.G. Wallace SpaceWorks Engineering, Inc. (SEI) Atlanta, GA U.S.A. ABSTRACT The U.S. Air Force has long been interested in the next-generation space launch capabilities that might be enabled by high-speed air-breathing propulsion. Since the early 1960's, studies have examined the application of ramjet or dual-mode scramjet propulsion to potential reusable military spaceplane (MSP) concepts. Perceived advantages include flexible basing, flexible operations, hypersonic strike from the Continental U.S., and low cost space access. However, many of the requisite propulsion technologies for high-speed air-breathing flight remain ongoing research efforts still today. Vehicle studies and concept assessments are typically used to provide a guiding vision to propulsion research efforts. These studies help define requirements and goals while also serving to explore the potential advantages and disadvantages of air-breathing space launch vehicles relative to current alternatives. The paper will highlight one of the advanced vehicle concepts developed under a project entitled "Innovative Concept Development of Reusable Launch Vehicles Using Combined-Cycle Propulsion for Military Applications." SpaceWorks Engineering, Inc. (SEI) conducted the project under a multi-year award from the Air Force Research Laboratory's Propulsion Directorate, Wright-Patterson AFB, OH 1,2. The concept to be reported on is a two-stage, vertical takeoff/horizontal landing MSP concept referred to as the Sentinel. This vehicle uses RBCC engines as the primary propulsion system for the reusable booster stage. The booster is capable of supporting three different missions by configuring the expendable upperstage with missionspecific hardware. These missions include: a) space maneuvering vehicle (SMV) delivery to orbit, b) a hypersonic strike mission, and c) cargo delivery to due-east low Earth orbit (LEO) from Cape Canaveral Air Force Station (CCAFS) as well as polar orbits from Vandenberg Air Force Base (VAFB). Several trade studies were conducted on the Sentinel concept. Various flyback conditions, the sensitivity to installed engine weight, and different missions were each evaluated. These trade study results as well as the baseline configuration results will be reported along with summary conclusions in this paper. Results of the cost, safety/reliability, and operations assessment for the Sentinel will not be presented in this paper. AFRL ACS AOA CCAFS CONUS DMSJ ETO GLOW HTV Air Force Research Laboratory Attitude Control System Angle-of-Attack, degrees Cape Canaveral Air Force Station Continental United States Dual-Mode Scramjet Earth-To-Orbit Gross Lift-Off Weight, lbs Hypersonic Technology Vehicle NOMENCLATURE * - President. ± - CEO. - Senior Project Engineer. Copyright 2007 by SpaceWorks Engineering, Inc. (SEI). -1-

2 IRS IVHM LEO LOX MPS MSP POST RBCC RLV RTLS SEI SMV SOL SOV TBCC TPS T/W VTHL Independent Ramjet Stream Integrated Vehicle Health Monitoring Low Earth Orbit Liquid Oxygen Main Propulsion System Military Space Plane Program to Optimize Simulated Trajectories Rocket-Based Combined Cycle Reusable Launch Vehicle Return to Launch Site SpaceWorks Engineering, Inc. Space Maneuvering Vehicle Shock-On-Lip Space Operations Vehicle Turbine-Based Combined Cycle Thermal Protection System Thrust-to-Weight Vertical Takeoff, Horizontal Landing INTRODUCTION The Sentinel Space Operations Vehicle (SOV) is the first element of a two-stage MSP concept that uses a combinedcycle air-breathing and rocket propulsion system (see Figure 2). The nominal system takes off vertically and lands horizontally. Baseline propellants are JP-7 and LOX for both the booster and the upperstage. Additionally, this booster stage is capable of supporting three different missions by configuring the upperstage with mission-specific hardware. The primary missions supported by the Sentinel include: a) SMV delivery to orbit, b) a hypersonic strike mission, and c) cargo delivery to due-east LEO from CCAFS as well as polar orbits from VAFB. The various upperstage configurations will be detailed in a later section. The Sentinel concept is enabled through the use of numerous technology advances in the areas of propulsion, materials, structures, avionics, and integrated vehicle health monitoring (IVHM). Key design features and technologies for the booster include: (4) Rocket-Based Combined Cycle (RBCC) engines Graphite-Epoxy (Gr-Ep) airframe primary and secondary structures Titanium-Aluminide (Ti-Al) wings, verticals, and tail Cylindrical, non-integral Gr-Ep fuel tanks Cylindrical, non-integral Aluminum-Lithium oxidizer tanks AFRSI TPS blankets on fuselage leeward surfaces Advanced Carbon-Carbon (ACC) TPS for leading edges Conformal Reusable Insulation (CRI) TPS for fuselage windward surfaces, wings, and tails Electro-hydraulic actuators for control surfaces Extensive Integrated Vehicle Health Monitoring (IVHM) systems For primary propulsion, the booster uses a set of four rocket-based combined cycle (RBCC) JP-7/LOX engines that employ the Independent Ramjet Stream (IRS) cycle. These engines are fully reusable and require minimal maintenance between flights. The airframe is a novel winged-body design with two-dimensional side-mounted engine pods. Advanced composites and metal-alloy materials are used for a number of key structural elements to reduce system weight and protect against high temperatures. Structural designs that maximize airframe stiffness while minimizing weight and intrusions into the airframe are required due to the high loading conditions during various flight maneuvers (liftoff, pull-up, max-q, etc.). Multi-redundant avionics and flight systems capable of fully autonomous or virtual operations from liftoff to landing are used. Additionally, through the use of imbedded sensors and detection systems the IVHM provides constant assessment of the vehicle s subsystems as well as flight conditions. Allowing for immediate -2-

3 identification of failing systems or violated operating conditions. This contributes greatly to the reduced ground processing times. Minimal technology advancements for the upperstage system are required. The primary technology challenges are the single JP-7/LOX liquid rocket engine and SMV (payload) integration and attachment to the expendable hardware elements. PROJECT BACKGROUND AND OBJECTIVES The United States Air Force has long been interested in the next-generation space launch capabilities that might be enabled by high-speed air-breathing propulsion. Since the early 1960 s, studies have examined the application of ramjet or dual-mode scramjet (DMSJ) propulsion to potential reusable MSP concepts. Perceived advantages include flexible basing, flexible operations, hypersonic strike from Continental United States (CONUS), and low cost space launch for cargo. However, many of the requisite propulsion technologies for high-speed air-breathing flight remain on-going research efforts still today. Previous studies have shown that both rocket-based combined-cycle (RBCC) and turbine-based combined-cycle (TBCC) propulsion approaches are strong candidates for application to a nextgeneration MSP. The rationale for this study was twofold. First, provide some guidance and definition for on-going AFRL research efforts toward the development of ramjet/scramjet propulsion with emphasis on combined-cycle propulsion systems. Second, compare and contrast alternate RBCC and TBCC propulsion approaches for next-generation military spaceplanes. To maintain consistency, all of the vehicles investigated during the study were designed with a common set of software analysis tools, design processes, and assumptions. Commensurate with study resources, all analysis was done at a level of fidelity appropriate for a conceptual or early preliminary design study. This equates to 3-degree of freedom, point mass trajectory optimization codes; panel and impact method aerodynamics codes (supported by limited CFD analyses, but no wind tunnel models); parametric weight estimating equations (no detailed structural design or finite element analysis); quasi-1-d analysis of dual-mode scramjet performance through the internal flowpath (2-D externally); and 1-D through-the-thickness thermal protection system analysis. The vehicle closure or synthesis process is an iterative process among the contributing disciplinary tools. For example, weight predictions from the mass estimating code must be fed into the trajectory optimization code. Propellant fractions from the trajectory code must be fed back into the mass estimating code to estimate new tank and fuselage weights. Depending on initial guesses, convergence amongst the disciplinary tools might take five to ten iterations to reach a required level of consistency between the coupling variables. SEI utilized a commercial design integration framework, Phoenix Integration ModelCenter, to manage the data exchange between the disciplinary analyses 3. Once each of the contributing disciplinary analyses was set up and verified for a particular configuration, the automated closure process could converge a design in a matter of hours over a distributed network of four or five computers. This automated convergence process was used to support design optimization, alternate mission profiles, and the various trade studies required by the study. Throughout the study, a goal was to avoid use or production of proprietary configurations or proprietary data. The Air Force customer desired open and exchangeable datasets and results, consistent with the Air Force s desire to release the data to other research partners and decision-makers in the field. This paper will report on the findings for one of the two MSP concept examined during this effort. -3-

4 SENTINEL MSP Mission Overview Current launch scenarios have space-access missions being conducted from a notional military spaceport facility at Cape Canaveral in Florida or Vandenberg Air Force Base in California. Flight operation scenarios for the spaceaccess SMV mission, hypersonic strike mission, and cargo missions will be presented next. Space Maneuver Vehicle (SMV) Delivery The nominal mission for the Sentinel is the delivery of a SMV to LEO. Figure 2 provides an ascent profile for this particular mission scenario. Specific details of the ascent profile for this mission are provided in the Trajectory Analysis section. The SMV payload is itself a reusable vehicle capable of remaining on-orbit for extended periods of time. The SMV has no capability to insert itself into orbit, thus it is carried as the payload of another earth-to-orbit (ETO) launcher. The SMV does feature a small, kerosene/h 2 O 2 liquid rocket propulsion system, the Rocketdyne AR2-3, for on-orbit maneuvering. The assumed vehicle weight for this analysis is 13,090 lb, which includes an experiment bay with a 500 lb of payload. The length from nose-to-tail is 27.5 feet and the wingspan is 15.0 feet. The SMV is deployed by an expendable rocket-powered upperstage. This stage consists of four propellant tanks (two fuel, two oxidizer), an attitude control system (ACS), flight controller, TPS blankets, and a single liquid rocket engine. The SMV is attached at its base to the fuel tank and engine thrust truss structure. The LOX oxidizer tanks are mounted over the wings on a secondary truss structure. The upperstage accelerates the SMV from the 9,000 fps staging condition to orbital velocities using JP-7 and LOX propellants (consistent with the booster stage propellants). Figure 3 shows an external view of the expendable upperstage system and the SMV payload (without the TPS shroud). Hypersonic Strike Mission with HTV Delivery The Sentinel booster is capable of acting as a long-range strike system. To perform this type of mission, the Sentinel booster can be outfitted with an external conformal fuel tank and two banks of two Hypersonic Technology Vehicles (HTVs) in place of the upperstage. The HTVs are unpowered hypersonic gliders with a nominal gross weight of 2,000 lb each. Figure 4 provides an aft-section view of the Sentinel integrated with a conformal fuel tank and four HTVs. Note that the HTVs are notionally represented as pyramidal volumes and are not representative of any specific design concept. Cargo Payload Delivery The Sentinel vehicle can also be configured to support general cargo delivery missions to LEO. In this configuration, the SMV is replaced with a similarly shaped payload fairing. This fairing, which has an internal volume of 950 ft 3 and weighs 2,500 lb, allows for up to 10,020 lb of payload to be transported to LEO due-east from CCAFS. Summary Concept Closure Results The length of the Sentinel fuselage is feet, with a wingspan of 75.8 feet and tail height of 26.5 feet. The weight of the booster, without propulsive fluids, is 158,060 lb. Without an upperstage module, the weight of the fully loaded booster with the JP-7 fuel and LOX oxidizer propellants is 677,810 lb, yielding a total gross liftoff weight (GLOW) of 756,545 lb (less startup losses) for the SMV deployment mission. Figure 5 provides external top, side, and frontal views of the mated vehicle with system dimensions. Table 1 provides a summary of key performance metrics for the system. Table 2 provides a summary of key upperstage performance metrics. The upperstage weighs 78,735 lb when loaded with propellants and the SMV. The total stage length is 46.7 feet when integrated with the SMV payload. -4-

5 Design Process and Methodology The complex vehicle architecture was designed and analyzed using a collaborative, distributed framework ModelCenter available from Phoenix Integration, Inc 3. Figure 6 provides a screen view of the performance closure model for this vehicle concept. A number of industry standard analysis tools were used in concert with numerous in-house codes developed at SEI. Each icon along the diagonal of the image shown in Figure represents an instance of an analysis tool that is being hosted and executed on a PC, Macintosh, or SGI platform. Lines connecting one icon to another in the model represent interdisciplinary data transfer. Additionally, connecting lines above the diagonal are for data flows from the upper left corner to the bottom right corner (i.e. feed-forward). Connecting lines below the diagonal represent data flows from the bottom right to the upper left (i.e. feed-backward). Ideally, all data flows are feed-forward. Anytime a feed-backward data flow occurs, an iterative process results due to disciplinary coupling. The coupling data is resolved through a simple fixed-point iteration (FPI) method, which generally requires 5-15 iterations to achieve a root mean-squared error convergence of or less (sum square of the relative difference between guessed values and calculated values). Specific industry codes used in the design of the Sentinel included: POST 2 3-degree of freedom trajectory simulation code 4 APAS - S/HABP aerodynamic and aeroheating load predictions 5 NASCART-GT 3-D Euler with external flowfield analysis 6 SRGULL scramjet engine performance analysis 7 Solid Edge configuration layout and center-of-gravity assessments 8 Tools developed by SEI in-house and used in the design of the Sentinel included: REDTOP attitude control system (ACS) performance predictions 9 REDTOP-2 liquid rocket engine design (booster tail-rockets and upperstage engine) 10 PARADIGM RBCC engine performance prediction tool SEI-Sizer airframe, tank, and subsystem weight modeling and system sizing Sentry transient, 1-D aeroheating analysis and TPS design/sizing 11 SESAW avionics subsystems 12 Disciplinary tools with tightly coupled variables (between multiple disciplines) that were capable of functioning in an automated environment were integrated together to create the Sentinel system closure model. Non-dimensional parameters like the aerodynamic coefficients and air-breathing propulsion system thrust coefficient permit scaling within reasonable ranges. These values were each normalized by a reference area that scales linearly with overall vehicle planform area as the vehicle is resized to meet mission requirements. Computing platforms used within the distributed framework included a SGI Octane Workstation, a dual 1.8Ghz 64-bit G5 Macintosh, and two 3.2Ghz Pentium-4 PCs. Trajectory Analysis DISCIPLINARY ANALYSIS RESULTS A typical space access mission for the Sentinel is comprised of a number of key events that must occur throughout the flight profile. For the SMV-deployment mission, the system nominally launches from a notional military spaceport facility at Cape Canaveral, located at approximately 28.5 N latitude. After the vehicle has been attached to the launch tower and erected vertically, the LOX/JP-7 rocket thrusters of the RBCC engines are brought up to full throttle. This provides a liftoff thrust-to-weight ratio of 1.25 for this system, with each engine providing 236,420 lbf of sea-level thrust at an Isp of 330 seconds. At full throttle, the hold clamps on the launch tower release and the Sentinel begins to accelerate upward, reaching 1,000 feet in 15 seconds. The autonomous flight system commands the vehicle along a flight path that puts the vehicle through the transonic flight regime at an altitude of 29,000 feet and -6 angle-of-attack. From liftoff up to Mach 3.5, the RBCC engine is operating in its IRS mode. At liftoff, the engine fuel injectors are shut off completely and are not injecting any additional JP-7 fuel into the air-breathing sections of the engine flowpath. As the vehicle accelerates, the rockets remain at full power and the equivalence -5-

6 ratio of the engine is increased, providing additional thrust and Isp. The augmented thrust from the IRS mode is most dominant and beneficial at supersonic flight conditions, with no augmentation occurring at subsonic speeds. Upon reaching a flight Mach number of 3.5, the RBCC engines transition from operating in the IRS mode to the DMSJ mode. To accomplish this, the rocket thrusters are quickly throttled down. The DMSJ engines operate in a fuel-lean mode until the flight speed increases enough to maintain the thermal choke at the nominal fuel-rich (phi=1.1) condition. Around Mach 6, pure supersonic combustion is achieved in the engine and the thermal choke is not required. The vehicle operates under DMSJ power and along a dynamic pressure boundary of 2,100 psf up to Mach 8. Upon reaching Mach 8, the LOX/JP-7 RBCC rocket thrusters are reignited and the vehicle departs from the high dynamic pressure path, beginning a sharp pull-up maneuver to setup the staging event. Within a few seconds the dynamic pressure drops considerably and the RBCC engines performance approaches that of a traditional rocket engine with a high expansion ratio. Note that the Mach number at this point is still approximately 8 due to atmospheric impacts on the speed of sound. After about 75 seconds, the RBCC engines on the booster stage have accelerated the system to the 9,000 fps staging condition at an altitude of 196,000 feet (q=25 psf). At the staging point, the attach clamps for the upperstage are released and the single upperstage rocket engine is ignited. Simultaneously, the Sentinel reduces its speed and flight path angle, initiating the separation of the two stages. Under all-rocket propulsion, the upperstage continues to the insertion velocity at a perigee altitude of 70 nmi. This results in the SMV deployment to an elliptic 70 by 197 nmi orbit at a 28.5 o inclination. The SMV is released from the expendable upperstage hardware elements (tanks, avionics, engine, etc.) and proceeds to conduct its primary mission. The remaining upperstage elements then use the single main engine to pitch over and perform a small delta-v burn of 100 fps for de-orbit and atmospheric disposal. After staging, the Sentinel continues to reduce its speed and performs an unpowered turn towards the launch site. When the vehicle has reduced its speed and altitude to Mach 5.0 at an altitude of 80,000 feet (q=1,000 psf), the DMSJ engines are restarted and used to fly the vehicle back to the launch site. The total flyback distance of 335 nmi requires just over 7 minutes to complete. Figure 7 provide time histories for the vehicle flight Mach number and altitude during the mission. The total ascent time-to-orbit for the SMV is just under 10 minutes from launch, with the staging maneuver occurring at 6 minutes into the mission. In the event that the vehicle is launching from an alternate base that has nearby downrange bases for service and propellant resupply, a partial amount of the flyback propellants can be used to either loiter or phaseup with a desired orbital plane. Figure 8 provides the relative flight velocity versus time profile for the nominal mission. Note that while in Figure 7 it appears the vehicle has a constant Mach number and is not accelerating during the pullup maneuver, the velocity gradient is positive over the same time period in Figure 8. Figure 9 provides the freestream dynamic pressure and altitude versus flight time profile for the nominal mission. Main Propulsion The Sentinel uses an RBCC propulsion system featuring an IRS integration approach 13. Instead of high-isp turbine engines for low-speed acceleration, the concept uses rocket thrusters embedded in the flowpath of an air-breathing engine. These rocket thrusters provide the liftoff, low-speed, and exo-atmospheric acceleration capability for the vehicle. During high-speed, atmospheric flight, these engines operate as a pure air-breathing system, with the rockets turned off. This engine requires a number of advanced technologies (to either improve performance or reduce weight) that are being researched and/or developed commercially at various funding levels within the U.S. government. Some of these technical challenges include: Rocket powerpack and nozzle integration with main flowpath (cooling issues at small radii locations) Reusability of hydrocarbon regenerative cooling channels High-temperature seals for variable-geometry inlet ramps Lightweight engine structures requiring long life (>500 flights) Non-intrusive inspection and part removal for replacement -6-

7 The four RBCC engines operate from liftoff to Mach 8 during the ascent phase and then provide the flyback propulsion after staging. The engines, which use JP-7 fuel and LOX oxidizer, are grouped in pairs on each side of the booster airframe. Each engine has its own, independent turbomachinery powerpack. During liftoff, each engine can generate 236,420 lb of thrust using the afterburner with a corresponding Isp of seconds. The engines have an uninstalled, sea-level-static thrust-to-weight ratio of 27:1, resulting in an installed T/W ratio of 23.5:1 (23.5 lb of SLS thrust per lb of engine weight, including inlet, powerpacks, actuators, engine cooling, etc. but prior to the addition of growth margin). The nominal engine T/W value used for the Sentinel is an optimistic, aggressive value. This number was not the result of a detailed analysis of the engine component weights, but was arrived at from a qualitative assessment. Previous RBCC engine programs funded by NASA have included engine T/W estimates as low as approximately 10, for a high-mach, hydrogen fueled engine with extensive variable geometry, and T/W values as high as 35, for a lower-mach number design with less variable geometry. Factors that support the assumed Sentinel RBCC engine T/W value included: 1) relatively low maximum operational Mach number of 8, 2) use of hydrocarbon fuel versus cryogenic hydrogen, and 3) the low level of variable geometry. Table 3 provides a summary of key RBCC engine parameters for the Sentinel and Figure 10 shows the engines integrated into the booster figure. Table 4 provides the design specifications for one of the four RBCC rocket thrusters. Note that the quoted thrust and Isp values are those that would be obtained from the thruster without integration in the RBCC engine. The RBCC engine effectively provides a larger expansion ratio for the thruster and hence installed performance is higher than that of the small thruster area ratio. Table 5 provides a summary of the Sentinel s RBCC IRS-mode engine performance data at various flight conditions. Note that these flight conditions do not necessarily correspond to actual conditions experienced by the vehicle. At hypersonic speeds in the atmosphere, the engines operate to Mach 8 as dual-mode scramjets (DMSJs). The engines use regenerative cooling with the JP-7 fuel and feature variable geometry inlets with a thermal-choke in the nozzle system. Achieving and maintaining the thermal choke enables both operations at lower flight Mach number conditions and keeps the engine weight to a minimum by eliminating the requirement for a mechanical-choke nozzle. The engine controller requires input from numerous engine sensors to assess the correct fuel flowrate and injection location to maximize operating performance. Figure 11 shows the reference centerline geometry used in the performance analysis of the DMSJ engines. The internal flowpath is shown in the minimum contraction configuration (low Mach number DMSJ operation). The forebody provides a significant amount of external compression and is designed for a Mach 8 shock-on-lip (SOL) condition. There are three, 2-D external compression ramps prior to the inlet ramp at the cowl leading edge. The forebody angles relative to the vehicle centerline are as follows: initial 5 turn from freestream, 9 intermediate turn, and final 12 turn to the inlet at 771 inches. The cowl is oriented at 0 incidence with the waterline and does not translate. There is sidewall contraction of 15% from the cowl leading edge to the inlet throat. The step or base area, which is evident at an axial location of 1,050 inches, is the rocket thruster nozzle exit plane. Combustion products from the rocket thrusters are injected into the engine at this location. Table 6 provides a summary of the engine performance in terms of thrust coefficient (Ct) and Isp versus Mach number. All data shown corresponds to a dynamic pressure of 2,000 psf. A fixed reference area (Aref) is used with q to non-dimensionalize thrust to obtain Ct. Here, Aref is the maximum cross-sectional area at the plane of the cowl entrance, 97.8 ft 2. The Sentinel reignites its rocket thrusters for the final pull-up maneuver before staging. The rocket-mode operation is needed to facilitate a low-q staging exo-atmospherically. Vacuum performance in this engine mode is predicted to be 355 seconds of Isp with a 35% throttled thrust level of 78,200 lbf (per engine). The Sentinel upperstage features a single liquid rocket propulsion system. These liquid rocket main propulsion systems do not require significant technology advancements in the area of performance, but they are new engine development efforts and manufacturing methods should be introduced that minimize costs. -7-

8 Weights and Sizing (W&S) As previously detailed in the Disciplinary Analysis Tools section, the Weights and Sizing (W&S) model is the primary synthesis tool for obtaining closure of the vehicle. The general procedure for the W&S discipline during the design cycle iteration is as follows: 1) Update the upperstage performance metrics (e.g. mass ratio to main engine cutoff, etc.) 2) Resize upperstage to achieve latest required mass ratio and compute a new gross weight, dry weight, and size for system. 3) Update booster stage performance metrics (e.g. mass ratio to main engine cutoff, mixture ratio, wing loading, etc.) 4) Update booster payload from new upperstage analysis 5) Resize booster to achieve latest required mass ratio and compute a new gross weight, dry weight, and size for the vehicle. Once a closed vehicle configuration has been achieved, the W&S model contains all of the subsystem size and weight details for the concept. W&S results for both the booster stage and upperstage elements will be presented next. Booster Sizing and Mass Properties The booster system entails the first stage of the Sentinel MSP (i.e. the SOV) and all the fully reusable hardware. This includes all the hardware necessary to accelerate the vehicle from the ground up to the staging condition at 9,000 fps. Table 7 provides a top-level listing of the main hardware elements and associated weights for the Sentinel. In the hardware elements subtable of Table 7, it is evident that a large portion of the vehicle s dry weight is due to the RBCC main propulsion system (approx. 26%). The airframe structure, which includes the main propellant tanks, aeroshell, and wing carry-through, is the next heaviest weight item, as to be expected. Note that a 15% dry weight margin, or manager s reserve, of 20,615 lb is also carried for this system. This dry weight margin is over and above any materials strength margin and factors of safety carried in the individual hardware weight estimates. The fluids weights table, on the right side of Table 7, shows the contributors to the 523,860 lb of propellant being carried onboard the system. In addition to the ascent propellants used to reach the staging condition, reserves, residuals, and unusable fluids are accounted for. The reserve and residual propellants for use by the main propulsion systems are estimated as 0.75% and 0.2% of the ascent propellants respectively. For the ACS system, the reserve and residuals are estimated as 10% and 3% (respectively) of the ACS nominal propellant load. It should be noted that in addition to the required propellant volumes needed to contain all the fluids, an ullage volume of 3% per tank is also included. Upperstage Sizing and Mass Properties The expendable upperstage hardware elements are designed with limited reliance on new technology developments to keep the recurring system costs to a minimum. The propellant tanks are all constructed from aluminum-lithium with cryogenic insulation on the oxidizer tanks. The ACS is a simple blow-down, pressure-fed, bipropellant arrangement using gaseous oxygen and the JP-7 fuel. The upper half of the body is covered in thermal blankets to protect the hardware during the ascent, prior to booster release. Since the system is unmanned, minimal system redundancy has been used. Table 8 provides a top-level listing of the main hardware elements and fluid weights for the upperstage. The integrated system has a combined weight of 78,735 lb when loaded with the SMV and all fluids contained in the right-most column of Table

9 Aerodynamics The Sentinel booster is a unique winged-body design with only a fraction of the lift being generated by the fuselage. The forebody is designed to minimize flow non-uniformity and provide a high mass capture and total pressure recovery to the engines. The external compression ramp angles and lengths result in a shock-on-lip (SOL) condition at Mach 8, with an initial turn angle of 5 and a maximum turning angle of 12. The leeward surface of the vehicle is designed to nominally provide a zero-incidence surface during flight for minimum drag. It is not practical to completely shield the upperstage from the flow, thus a small drag penalty is always incurred when the upperstage is mated to the booster. The aft, underside section of the vehicle is designed to provide additional compressive lift from the airframe. The wing size and incidence angle is determined by a desire to keep the angle-of-attack during DMSJ operation below 5. This will minimize crossflow and flow non-uniformity at the engine inlet face. The wing characteristics include a sweep angle of 52.3, a taper ratio of 0.16, a span of 75.8 feet, and a thickness-to-chord ratio of 0.04 at the root and 0.02 at the tip. The wing is mounted to the fuselage at a positive incidence angle of 4. The wing structure is designed to sustain a load factor of 1.75 times the GLOW of the system. Table 9 provides a summary of the wing, tail, and wingtip verticals design specifications. Once supersonic, the vehicle experiences an average AOA of Aerodynamic data for Sentinel was obtained from the APAS code. The geometry used for the analysis is shown in Figure 12 and contains approximately 2,568 surface nodes. Note that the apparent gap between the wings and the fuselage is a modeling artifact of APAS. The RBCC engines are nominally located in this gap but are not included as part of the aerodynamic modeling. Due to the analysis methods being used by APAS and S/HABP, this gap does not impact the ability of the wings to generate lift, etc. Aeroheating and Thermal Protection As the vehicle accelerates into the high Mach number flight regime, the flowfield temperatures around the Sentinel begin to increase and would quickly exceed the material limits of the aluminum and Gr/Ep airframe structures. All external surfaces are required to be covered with high-temperature materials to protect the vehicle. Fortunately, due to the relatively low hypersonic Mach number envelope, all thermal protection systems are passive (i.e. no active cooling required on external surfaces). The Sentry thermal analysis and design tool assessed each stackup material on a vehicle analysis grid that consisted of 2,188 nodes. The grid was similar to the grid used for the APAS analysis (Figure 12) and possessed a smooth upper surface (no upperstage). The aeroheating analysis started at Mach 1.25, with a uniform structure temperature of 600 R. The analysis modeled the flyout, pullup and staging maneuver and included a two-hour flyback and cool down (thermal soak) period. A constant emissivity of 0.8 was used for all surfaces and the vehicle was assumed to have a leading edge radius of one inch. The backface temperature for the airframe was required to not exceed 760 R, while the backface temperature for the wings and tails was limited to 950 R. Also, the maximum material temperature for the AFRSI blankets, TUFI AETB-8 tiles and Advanced Carbon-Carbon (ACC) are 1660 R, 2860 R and 3360 R, respectively. The Conformal Reusable Insulation (CRI), and Ultra-High Temperature Ceramic (UHTC) SHARP candidate materials have a maximum reuse temperature of 2000 R and 4460 R, respectively. The optimal TPS material stackup for the Sentinel is determined at each grid node by considering the thickness manufacturing constraints, maximum reuse surface temperatures, and the minimum weight. The resulting materials and their weights selected by the analysis are listed in Tabl 10. The average weight for all the TPS materials is 1.23 psf. The majority of the windward surfaces of the airframe are covered with CRI, although the nozzle employs TUFI ceramic tiles. The leeward surfaces of the airframe are covered with AFRSI blankets. The vehicle has fairly sharp leading edge radius of 1.0 inch, which results in a stagnation point temperature exceeding 2,980 R and requires the use of ACC. The wings, tail, and verticals experienced maximum leading edge (stagnation) temperatures of 3,350 R, 3,310 R, and 3,340 R respectively. The majority of the acreage for the upper and lower wing, verticals, and tail surfaces are covered with a thin layer of CRI over a Ti-Al structure. The wing control surfaces and leading edges are covered with ACC. Some of the lower wing surfaces, away from the leading edge, required ACC due to the flow incidence angle and resultant heat rate being experienced during the pullup maneuver. This resulted in a noticeable -9-

10 increase in the wing TPS weights. With further trajectory optimization of the pullup maneuver, this weight penalty could be eliminated. The maximum temperature calculated for each vehicle panel is shown in Figure 13. CONCEPT TRADE STUDIES A series of trade studies have been conducted to better understand the concept and ascertain the additional capabilities of the Sentinel. Specifically, two studies were conducted, a thrust-to-weight sensitivity on the booster s main propulsion system and an analysis of the baseline system s ability to conduct hypersonic strike missions. A brief description and summary results for each trade conducted will be presented next. RBCC Engine T/W Sensitivity At the start of the concept study, the uninstalled, sea-level static thrust-to-weight (T/W) of the booster s RBCC propulsion system was assumed to be 27:1. This assumption was based on common values quoted in the literature for these types of engines. A parametric study was conducted to determine the Sentinel s sensitivity to this assumption. Specifically, a more aggressive T/W value of 35:1 was examined along with a more conservative, less optimistic value of 20:1. Note that these T/W values are prior to the addition of growth margin, which is book kept separately. The impact of the T/W changes was assessed in terms of the engine weight, not engine thrust levels (an alternative approach that will yield different results). For both of these cases, the entire vehicle closure process was repeated to arrive at two new closed vehicle designs. Table 11 presents the top-level results from this activity. As expected, the lower T/W engines resulted in an increase in system size and weight when compared to the nominal, higher T/W scenario. Of real interest though is the magnitude of the changes and sensitivity to the T/W assumption. For the T/W of 20:1 (an increase of 26% in engine weight), a net increase of 23% on the booster gross weight and 29% increase on the booster dry weight was experienced. For the higher T/W of 35:1 (a decrease of 29% in engine weight), a net decrease of 12% on the booster gross weight and 16% decrease on the booster dry weight was experienced. This is particularly interesting because the percentage change in engine weights for the T/W values investigated were nearly identical (26% versus 29%). It should also be noted that the upperstage gross and dry weights were basically unaffected by the increase in booster size. They were unaffected because all three cases provided staging at a flight velocity of 9,000 fps for the upperstage. Thus, the total delta-v required from the upperstage remained constant. The small differences in gross weight were due to small variations in the staging conditions due to solution convergence within the POST tool. Hypersonic Strike Mission The capability of the baseline Sentinel booster to perform a hypersonic strike mission (HSM) was assessed. As noted previously and shown in Figure 4, the vehicle s upperstage system is replaced with a conformal propellant tank and up to four hypersonic technology vehicles (HTVs). Each HTV nominally weighs 2,000 lb and is assumed to travel an additional 1,500 nmi when released at the Sentinel s staging condition 9,000 fps. Replacing the upperstage with an equivalent gross weight of 78,735 lb and subtracting the four HTV payloads that weigh 8,000 lb, their attachment structure with a weight of 800 lb, and a conformal fuel tank with a structural weight of 1,630 lb, and an 11.1% maneuvering margin leaves a useable JP-7 propellant load of 61,475 lb in the top-mounted conformal tank. It is assumed for the maximum strike range case that down-range basing and/or aerial refueling will be available. Therefore, two-thirds of the nominal flyback propellant (contained internally in the booster) can also be used for extended range. This results in a total of 78,940 lb of JP-7 available for the HSM. The remaining flyback propellant, 8,730 lb of JP-7, is reserved to allow the booster an additional powered flight range of 100 nmi for refueling and/or landing at a remote base after the HTVs are released. Using the maximum available propellant load, the extended range capability was calculated at various flight conditions ranging from Mach 4 to Mach 6. These ranges can be added to the nominal booster range of 335 nmi and HTV range of 1,500 nmi to determine the total hypersonic strike range capability. The results of this analysis are presented in Table

11 The optimal, feasible conditions for achieving maximum range occur at Mach 5 (with DMSJ propulsion) at an altitude of 76,000 feet (q~1,400 psf). These conditions result in a total range of 2,540 nmi from the launch site to the ground target. The optimal altitude was determined at each cruise Mach number using an optimizer to maximize the range subject to a minimum dynamic pressure of 450 psf as a lower limit for DMSJ operation. Additionally, a maximum AOA of 5 degrees was imposed. It is an interesting finding that there is a peak in the cruise range for the strike mission that occurs at a Mach number of 5.0. While there are many competing factors that impact the strike range of the vehicle, a simplified simulation at a constant dynamic pressure for all conditions also showed this same effect. Between Mach 4 and 5, the engine Isp is relatively constant. While the lift-to-drag is slightly lower at Mach 5 than at Mach 4, the impact of the reduced starting weight at Mach 5 offsets this negative effect on range. Thus, for very similar Isp values and a fixed propellant load at 78,940 lbs, the Sentinel can travel farther at the higher Mach number. Above Mach 5, the engine Isp and aerodynamic lift-to-drag ratio begin to decrease significantly. Again, for the same fixed propellant load, this results in a net decrease in range even though the starting weight is slightly less. This range is less than a third of the desired global-reach capability for the DARPA/Air Force FALCON program at 9,000 nmi. Note that the Sentinel capability for performing this mission is lower because of the relatively short flyback/rtls distance of 335 nmi required for its nominal mission. With modifications to the booster to increase internal tank volume, this range could be greatly expanded. However, this would result in higher gross and dry weights for the baseline system. CONCLUSIONS Throughout the course of this effort, a number of observations regarding the Sentinel system configuration were made. Some of these are configuration specific and the result of design decisions made early in the program. Others are more broadly applicable to this class of launch system (i.e. a two-stage rocket-based combined cycle booster with an expendable upperstage). Some of the key lessons learned are: 1) The Sentinel s RBCC Independent Ramjet Stream (IRS) operational mode did not appear to offer any significant thrust or Isp augmentation until flight conditions exceeded Mach 2. Given that the rocket thrusters were shut down at Mach 3.5, the usefulness of integrating the rockets in the flowpath is questionable. A non-integrated propulsion system could achieve similar performance up to Mach 3.5 without the added complexity of the rocket integration. Additionally, the DMSJ performance could likely be improved by removal of the base area containing the rockets, resulting in a net gain for the entire system. The recommendation for this configuration is to examine alternative RBCC cycles, such as the Simultaneous Mixing and Combusting (SMC) or Diffusion-then-Afterburning (DAB), which offer true augmentation at lower flight speeds. 2) The flyback, or return-to-launch site (RTLS), requirement for the booster is a significant impactor on the vehicle s size and weight. A quick trade analysis (not included in detail in this paper) that eliminated the flyback requirement for the booster resulted in an 18% reduction in gross weight and a 14% reduction in dry weight. While it is likely infeasible and undesirable to eliminate the flyback capability, it is useful to understand the system sensitivity to this often neglected part of the mission. Unique methods of retrieving the booster including aerial refueling, dual-location basing, and rocket boost-back should all be investigated in future studies. 3) The high thrust levels required by the main engines to support a vertical takeoff launch result in excessive thrust margin during the final pullup maneuver prior to staging. For the Sentinel, the main engines were throttled back to 35% of their maximum thrust for this maneuver. It is noteworthy that this is also a common problem in all-rocket single-stage to orbit (SSTO) designs as they near their final orbital insertion conditions. While throttling to 35% is not unrealistic, it is near the low end of normal rocket engine operating conditions and forces the turbomachinery to operate significantly off-design. This capability will increase the development costs of the engines. -11-

12 4) The unique wedge-shaped fuselage configuration of Sentinel increases its engine capture area relative to the vehicle s wetted area. This increases overall vehicle acceleration and is generally beneficial for a space launch mission. However, care must be taken to limit the fuselage angle-of-attack during DMSJ operation. Excessive AOAs lead to cross-flow at the cowl entrance and degrade overall performance. 5) During this study, several candidate wing incidence angles were investigated between 0 and 4. A wing incidence angle of 4 was thought to produce the best compromise between aerodynamic performance, propulsion performance, and aeroheating impacts, but this value was not necessarily optimized rigorously. At this wing incidence angle, the fuselage sees an average AOA of only 2.25 during DMSJ operation. The wing flies at an average net AOA of Future work might investigate this variable in more detail and further refine the results. ACKNOWLEDGEMENTS SEI and the authors would like to express their sincere appreciation to Dr. Dean Eklund, Mr. Albert Al Boudreau and Mr. Glenn Liston from the Air Force Research Laboratory (AFRL) at Wright-Patterson Air Force Base for leading and managing this important project. Technical review and support was provided by a number of talented engineers at NASA field centers and Air Force support contractors. Specifically, SEI and the authors would like to thank Dr. Chuck Trefny at NASA Glenn Research Center and Mr. Jeff Robinson at NASA Langley Research Center. Dr. Trefny provided technical guidance for the Sentinel s Independent Ramjet Stream (IRS) Rocket-Based Combined Cycle (RBCC) propulsion system. Mr. Robinson was able to attend a number of milestone reviews during the course of the activity. Additional technical contributions and continuous support via monthly telecoms and attendance at milestone reviews were made by Mr. Chan Cho of Spiral Technology, Inc. at Edwards Air Force Base in California and Mr. Chuck Bauer of Universal Technology Corporation at Wright-Patterson Air Force Base in Ohio. The authors would also like to acknowledge Phoenix Integration, Inc. for providing assistance in the development of the collaborative environment through access to the latest ModelCenter and Analysis Server products. REFERENCES 1. Bradford, J. E., Eklund, D. R., Charania, A., Wallace, J. G., "Quicksat: A Two-Stage to Orbit Reusable Launch Vehicle Utilizing Air-Breathing Propulsion for Responsive Space Access," AIAA , Space 2004 Conference and Exhibit, San Diego, California, September 28-30, Wallace, J. G., Bradford, J. E., Charania, A., Escher, W. J. D., Eklund, D., "Concept Study of an ARES Hybrid-OS Launch System," AIAA , 14th AIAA/AHI Space Planes and Hypersonic Systems and Technologies Conference, Canberra, Australia, November 6-9, ModelCenter 6.1, Phoenix Integration, 26 April 2006, 4. Striepe, S.A., et al, Program to Optimize Simulated Trajectories (POST II), Volume II, Version G, NASA Langley Research Center, January Sova, G, P. Divan, Aerodynamic Preliminary Analysis System II: Part II User s Manual, NASA CR , Numerical Aerodynamic Simulation via CARTesian Grid Techniques, Department of Aerospace Engineering, Georgia Institute of Technology, Atlanta, Georgia, 26 April 2006, 7. Pinckney, S. Z., J.T. Walton, Program SRGULL: An Advanced Engineering Model for the Prediction of Airframe- Integrated Subsonic/Supersonic Hydrogen Combustion Ramjet Cycle Performance, NASP TM 1120, NASA, January Solid Edge Evolve to 3D: 3D CAD for Engineers, UGS, 26 April 2006, 9. Rocket Engine Design Tool for Optimal Performance (REDTOP), 26 April 2006, Bradford, J.E., A. Charania, B.D. St. Germain, REDTOP-2: Rocket Engine Design Tool Featuring Engine Performance, Weight, Cost, and Reliability, AIAA , 40 th AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit, Fort Lauderdale, Florida, July 11-14, Bradford, J. E., Olds, J. R., "Thermal Protection System Sizing and Selection for RLVs Using the Sentry Code," AIAA , 42nd AIAA/ASME/SAE/ASEE Joint Propulsion Conference, Sacramento, California, July 9-12, Simple Electrical Systems and Avionics Wizard (SESAW), April 2006, Yungster, S., Trefny, C.J., Analysis of a New Rocket-Based Combined-Cycle Engine Concept at Low Speed. AIAA

13 FIGURES Figure 2. Sentinel Space Operations Vehicle (SOV). Figure 1. Sentinel SMV Delivery Mission Profile. -13-

14 Figure 3. Sentinel Upperstage and Space Maneuvering Vehicle (SMV) Elements. Figure 4. Sentinel Hypersonic Strike Mission (HSM) Configuration. Figure 5. Baseline Sentinel Outer Moldline Layout and Geometry. -14-

15 Figure 6. Sentinel Performance Closure Model in ModelCenter. -15-

16 Mach Number Altitude 450, , , , ,000 Mach Number , , ,500 Altitude (ft) , , , , Time (seconds) Figure 7. Sentinel Flight Mach Number and Altitude vs. Time. 25,000 22,500 20,000 17,500 Velocity (fps) 15,000 12,500 10,000 7,500 5,000 2, Time (seconds) Figure 8. Sentinel Relative Flight Velocity vs. Time. -16-

17 2, ,000 2, ,000 2, ,000 Dynamic Pressure (psf.) 1,750 1,500 1,250 1, , , , , ,000 Altitude (ft) Dynamic Pressure Altitude 90,000 45, Time (seconds) Figure 9. Sentinel Freestream Dynamic Pressure (q) and Altitude vs. Time. Figure 10. Sentinel Booster RBCC Engine Integration. Units: inches Figure 11. Sentinel Booster Centerline Geometry (top view). -17-

18 Figure 12. APAS Aerodynamic Analysis Grid for Sentinel Lift and Drag Predictions. Windward Leeward Figure 13. Sentinel Maximum Surface Temperatures. -18-