<p>This paper shows the effect of the side clearance on the air-intake performance of a ramjet for the High Mach Integrated Control Experiment (HIMICO). As the result of the wind tunnel test, the mass capture ratio (MCR) of the original intake is 15-35% lower than the designed value. CFD analysis suggested that this is caused by the flow separation on the ramp surface due to flow leakage from a 1-mm side clearance located at the moving part. The mechanism of the side clearance effect is cleared by CFD, and the CFD results considering the side clearance almost agree with the experiment. Then, a new intake in which the side clearance is reduced to 0.25 mm was designed and tested. The maximum MCR of the new intake is increased approximately 14% compared to the original intake.</p>

The Japan Aerospace Exploration Agency, in partnership with academia and industry, are developing the Air Turbo Rocket for Innovative Unmanned Mission (ATRIUM) engine: an air turboramjet + rocket combine cycle propulsion system intended to replace conventional liquid rocket engines in Vertical Takeoff Vertical Landing applications, such as reusable sounding rockets. A subscale Flight Test Bed (FTB) vehicle is also being developed to demonstrate the ATRIUM engine in a flight environment. In this paper, the ATRIUM engine and FTB vehicle are introduced, and current progress in their development is summarized. Future test plans and practical applications are also discussed.

The void fraction and vapor quality are important parameters for characterizing the gas-liquid two-phase flow. However, neither an established void fraction measurement method nor a verified void fraction - vapor quality interconversion model is available for the two-phase hydrogen flow. The object of this study is the development of a void fraction measurement technique and the investigation of the void fraction-quality correlations. A capacitive void fraction sensor was developed using the electric field analysis (EFA) and design of experiment (DOE), and it was applied in a boiling hydrogen experimental facility. The experimental conditions were as follows: the inner diameter of the heating pipe was 15 mm, the mass flux was ranged from 50 to 110 kg/m(2)s, and the static pressure was ranged from 250 to 300 kPaA. Further, the correlation between the thermal equilibrium quality (chi(ac) = - 0.03-0.14) and void fraction (alpha = 0-70%) was compared with that obtained in previously proposed models, and the void fraction - actual quality - thermal equilibrium quality interconversion models applicable to the boiling hydrogen flow were investigated. It was observed that the combination of the Sekoguchi model for thermal equilibrium quality - actual quality conversion and the Steiner drift-flux model for actual quality - void fraction conversion agreed well with the experimental results. (C) 2019 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

The aim of this study is a characterization of boiling hydrogen flow in horizontal circular pipe flow. The most important parameters for boiling flow are a void fraction and flow quality. Although the void fraction is measurable in some way, there is no established method for cryogenic fluid. The authors developed a capacitive void fraction sensor and applied it for boiling hydrogen flow experimental facility. The correlations between the void fraction and flow qualities are investigated by comparing the previously proposed models. The conversion model of the combination of Sekoguchi simple model and the Steiner model agrees very well with the experimental result.

To realize high-performance cryogenic propulsion systems, the chilldown sequence has to be improved. Because the chilldown is carried out under low gravity, the effect of gravity on the two-phase flow, especially at low flow rate, should be investigated. To understand the physics under low gravity, an experiment was conducted using a sounding rocket. Two identical test sections with different mass flow rates simulated part of a turbopump, each of which has a complex flowpath including slits and a dead end. Using liquid nitrogen, the flight experiment obtained data of temperatures, pressures, void fractions, and video frames of liquid motion. Then, the flight experiment data were compared to the ground data taken under normal gravity, revealing that the slits played an important role in the chilldown process and that the test sections were quickly chilled down under low gravity. The slits of the test sections formed liquid jets, and their behaviors were different from those in the ground experiment. In the flight experiment, the jets easily reached the dead end of the test sections and cooled down the whole walls due to the increase in inertia and wettability; however, such behaviors were hardly observed in the ground experiment. The difference between the ground and flight is significant at lower flow rate.

This manuscript describes the work performed on void fraction measurements a cryogenic flow by means of a customized capacitive sensor. In a preceding activity, described in Part I, the instrument was developed and validated at room conditions. In the current study, the probe is exploited to detect the gaseous content during liquid nitrogen chilldown experiments. The sensor performances are evaluated both numerically and experimentally. The numerical simulations lead to the development of a new calibration formula improving the sensor measurement accuracy down to +/- 6.0%FS, within 99% confident interval. The experimental campaign mainly reveals a dependency of the sensor performance on the pressure and temperature variations during the cooldown of the test section. The so-called "thermal effect" therefore modeled and two compensation equations are derived. The void fraction results accordingly corrected, match the single-phase flows reference conditions within 2% discrepancy. Background light visualizations are also performed allowing the optical verification of the flow regimes. For a specific flow condition, a correlation between the recorded light intensity and the capacitive measurements is obtained. By means of the high-speed movies, the capacitive sensor response time is also evaluated to be 100 Hz.

This manuscript presents the design of a capacitive void fraction sensor for cryogenic LN2 two phase flows and its validation at room conditions. The capacitive void fraction sensor is first designed by means of Electric Field Analysis (EFA) simulations taking into account specific technical constraints coming from the test section in which it should be accommodated. Then it is manufactured and validated using a proper combination of fluids (Polydimethylsiloxane (PDMS) and air) having a dielectric constant ratio similar to the one encountered in LN2/GN2 two phase flows. The validation is performed through comparison with void fraction measured by means of optical visualizations and shows how the capacitive measurement technique robustness allows obtaining reasonable accurate values of void fraction also for the substitute fluid case. The sensor presented in this manuscript was used to evaluate the void fraction during LN2 chilldown of a rectangular cooling channel and the results are presented in the second part of this work.

The convergence and accuracy of gradient values on high aspect ratio grids remain problems in CFD. One of the methods solving these problems is to use a hyperbolic system. In this study, we investigated time evolution methods for hyperbolic systems and compare the hyperbolic method with a traditional method. We solve the following test cases: one and two dimensional advection-diffusion problems, Navier-Stokes problems such as laminar flow on a flat plate and laminar flow around a cylinder. We confirmed that the convergence in hyperbolic systems was much more rapid and the accuracy of gradient values was higher than that of traditional system. The hyperbolic system takes almost the same time or shorter time than traditional system on same grids. In the case of Navier-Stokes problems such as high Reynolds number boundary flow, on grids achieving the same accuracy, it takes less time in hyperbolic systems than in traditional systems. One of the major findings is that using approximate Jacobian gives the same order accuracy as using exact Jacobian and reduces calculation time remarkably in hyperbolic system. Calculation time was 19% shorter in 1D advection-diffusion problem, 9% in 2D advection-diffusion problem, and more than 74% in Navier-Stokes systems.

A capacitance-based void fraction sensor has been developed for the rocket or airbreathing engines, which is simple and do not disturb the flow. Typical conventional sensors usually have two concave electrodes mounted on the outer wall of the dielectric tube. They are relatively low accuracy if they have a noise shield; the maximum measurement error is over 30% in our research. The aim of this study is to improve the measurement accuracy while keeping the advantage of simplicity, mountability and non-intrusive characteristics. A theoretical formulae and electromagnetic field analysis, EFA, are used to design the sensors and are compared to an experiment using air/silicon-oil mixture flow. As the result, a newly developed asymmetrical type sensor which consists of asymmetric flat electrodes with side walls shows good performance; the inaccuracy between true void fraction and measured void fraction is 6% for the stratified flow.

The hypersonic air-breathing engine, which is currently under development by Japan Aerospace Exploration Agency (JAXA), uses liquid hydrogen as the fuel. In order to accurately control the fuel flow rate during the start-up, it is essential to measure the heat transfer and pressure drop of the two phase flow. These two characteristics depend on void fraction, flow velocity and flow regime; thus, measurement methods for these values are required to be established. In this study, the void fraction measurement method by the high-speed image analyses has been developed. Two images taken from top and side directions by high-speed cameras of 1000 fps are used for "two-direction semi-automatic analysis". We also develop "One-direction full automatic analysis" which uses only the side view as a full-automatic and high sampling rate method with little accuracy deterioration. The preliminary verification test using vertical pipe and acrylic ball shows favorable results within 2.2% error against the theoretical value. A cryogenic experiment using two-phase nitrogen flow was also conducted. Sampling rate of "One-direction full automatic analysis" can be up to 1000 Hz. The difference between the results of two methods was as minor as 5% when the void fraction was below 30%.

Capacitance type void fraction sensors for cryogenic fluid has been developed by our research group. The sensors are roughly categorized into three types; arc type, small-sized arc type, and asymmetric type. In this paper, the three type sensors are compared, and each sensor's merits and demerits are indicated. The arc type sensor was first developed. The sensor is simple, but S/N ration is low and influence of "temperature drift" is large. The small-sized arc type sensor was developed for the sounding rocket S310 test No.43, so the sensor was needed to make as small as possible. The sensor succeeded to measure the void fraction of the liquid nitrogen flow even under microgravity. The asymmetric type sensor was deviced to improve the measurement accuracy. Inaccuracy of the sensor between true void fraction and measured void fraction was only 3% for stratified flow, whereas that of the arc type sensor 30%.

The Japan Aerospace Exploration Agency launched the S-310-43 sounding rocket from the Uchinoura Space Center on Aug.04, 2014 for the purpose of investigating such behavior as boiling and flow of cryogenic liquid rocket propellant in an environment simulating coasting flight on orbit by using the sounding rocket's sub-orbital ballistic flight. In the low-gravity state, the cryogenic fluid (liquid nitrogen) was introduced into the test sections of similar shapes to the flow channels in the cryogenic propulsion systems. The boiling of liquid nitrogen inside the test-sections and the transition of flow regimes from gas/liquid two-phase flow to liquid mono-phase flow were visualized. The temperatures, pressures and void fractions of each channels were measured as well. Development of the experimental equipment for S-310-43 sounding rocket is described in this paper.

A numerical simulation of a mixed-flow compressor under windmill conditions is conducted to investigate the cause of stagnation pressure loss captured in the previous experiment, and to clarify the qualitative flow structure under windmill conditions. As a result of the simulation, the compressor rotor causes a stronger load on the airflow even under windmill conditions, because the compressor is connected to the turbine. It turns out that the mass flow rate increases in the order of rated operating point: windmill conditions with turbine and windmill conditions without turbine. In addition, a negative angle of attack is confirmed for the first stator under windmill conditions, which leads to a large separation on the pressure-side of the blade. This tendency increases when the rotational speed decreases. By a simple analysis, it is clarified that the lower the pressure ratio and rotational speed, such as under windmill conditions, the further the rotor and stator angle of incidence become separated from the optimal degree.

© 2015, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. In the present experiment, by using the sounding rocket’s sub-orbital ballistic flight, realized the gravitational environment similar to that of liquid-fueled rockets during its coasting flight. In the low-gravity state, the cryogenic test fluid, liquid nitrogen, was introduced into the test sections which had similar shapes to the flow channels in the cryogenic propulsion systems. The boiling of liquid nitrogen inside the test-sections and the transition of flow regimes from gas/liquid two-phase flow to liquid mono-phase flow were successfully visualized. The temperatures, pressures and void fractions in each channel were measured as well. The mechanisms enhancing heat transfer were discussed based on the visualization. In the present case, compared with the corresponding ground test, it was confirmed that the two-phase flow in the complex channel could wet the heat transfer surfaces more easily due to the absence of gravity, and that more uniform chill-down effect could been obtained.

The flow around an ONERA-M6 wing, including the effect of wind tunnel wall interference, is computed using CFD analysis with a porous wall model. The computational domain sets porous walls at the top and bottom of the wing section similar to actual wind tunnel experiments. The computational result captures almost the same shock wave shape as the wind tunnel. This could not be computed exactly in previous works that did not include wall effects. The interference of the porous walls reduces the Mach number and attack angle of the flow, and these effects alter the swept angle of front shock wave and the location of rear shock wave. The aerodynamics coefficients are also affected by the wall interference. The lift coefficient becomes smaller due to the reduction in attack angle. The lower Mach number decreases the drag coefficient, while the reduction in attack angle causes additional drag by a mechanism similar to induced drag.

In this experimental study, the combined effect of spatial and temporal variations of fuel-air mixture on self-excited combustion instabilities in a gas-turbine model combustor (similar to 60 kW) with a low-swirl injector is reported. Detailed measurements were performed in 4 fuel split (to upstream/downstream injections) conditions while keeping the total equivalence ratio constant. The combustion stability was found to be very sensitive to the fuel split parameter which determined the local equivalence ratio distribution. The majority of the heat-release oscillations was generated in the flame-to-wall impingement region in a manner that satisfied the Rayleigh criterion. The driving force of the instability was considered the periodic interaction between the traveling vortex filled with fresh mixtures and the flame in the near-wall region as reported in the previous study on homogeneous mixture flame. However, the strength of the instability was sensitively modified by the change in the local equivalence ratio distribution. In the strongest oscillation case with inhomogeneous mixture, temporal variations of equivalence ratio exhibited a positive contribution to the thermoacoustic coupling. This suggested that temporal variations in equivalence ratio were enhancing the driving factor of the thermoacoustic instability in addition to the vortex-flame interaction mechanism. (C) 2014 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

A new capacitance type void fraction sensor was designed, produced, and tested. This sensor applies the difference between the relative permittivity ε of gaseous hydrogen (ε= 1.0) and that of liquid hydrogen (ε = 1.2). Following the sensor verification test using light diesel oil and air, a cryogenic experiment using liquid nitrogen was conducted. As a result, the void fraction measured by the sensor showed good agreement with the result obtained by an optical analysis using a high speed camera. One of the key problems on the sensor is an existence of the temperature drift caused by the change of the relative permittivity of the glass tube. In order to reduce the temperature drift, the length of electrodes and material of tubes were changed. Combination of short arc length electrodes and iupilon tube is ideal for reducing the unwanted temperature drift. The sensor which has shorter electrodes reduces the quantity of the temperature drift by 63% compared to the original sensor.

Laser Induced Plasma Spectroscopy (LIPS) is a technique aiming at measuring local equivalence ratio inside gas turbine combustors without disturbing the combustion fields. In this study, LIPS is applied to well controlled mixtures of gases at the pressure of 1.5 MPa to obtain plasma spectra as calibration. Two types of estimation methods are examined. One is the peak area ratio method in which ratios between particular atomic emissions in the plasma spectra are used. The other is the correlation method where correlation coefficients between the objective and the database spectra are used. The correlation method showed the best performance in terms of precision in the calibration test. Finally, the technique is applied to the measurement of 2.5 MPa combustion gas. The peak area ratio of H656/ N746 provided a good agreement with the gas sampling data, while the peak area ratio of H656/O777 and the correlation data shifted toward higher equivalence ratios. Oxygen quenching is considered as a dominant cause of the shift. In summary, the results showed promising features of the LIPS technique under high pressure conditions. For improving the accuracy further, it is required to take into account the effect of oxygen quenching.

To model porous walls used in transonic wind tunnels, flow through a hole is investigated using Computational Fluid Dynamics (CFD). First, we analyze the relation between flow rate and differential pressure across the hole. At low differential pressures, such as for wind tunnel porous walls, the flow rate is found to increase linearly with differential pressure. We therefore propose a new model based on a linear relationship between flow rate and differential pressure. The effects of hole shape and boundary layer conditions near the hole are then investigated. In the outflow case (i.e., wind tunnel to plenum chamber), the flow rate increases as the ratio of hole depth to diameter becomes large due to variation of the flow separation area at the hole exit. Boundary layer thickness also affects the flow field: when the ratio of boundary layer thickness to hole diameter becomes small, the flow rate decreases, because the flow along wind tunnel side wall interacts more strongly with the flow through the hole.

A precooled turbojet engine has been developed by JAXA used for the hypersonic airplane and spaceplane. The subscale engine named "S-engine" whose thrust and weight are about 1.2 kN and 100 kg was designed, fabricated and tested. The components and the system firing tests under the sea-level-static condition were successfully conducted. In the next phase, a flight test of the S-engine is planned using a stratospheric balloon in 2010 called balloon-based operation vehicle (BOV). The vehicle is dropped from an altitude of 40 km by a high altitude balloon. After 40 s free-fall, the vehicle is pulled up and the S-engine operates for 30 s at about Mach 2. High-altitude tests of the core-engine verified the performance and healthiness of the engine under the condition corresponding to the BOV flight trajectory. (C) 2009 Elsevier Ltd. All rights reserved.

The aim of this paper is to apply a multi-objective optimization generic algorithm (MOGA) to the conceptual design of the hypersonic/supersonic vehicles with different cruise Mach number. The pre-cooled turbojet engine is employed as a propulsion system and some engine parameters such as the precooler size, compressor size, compression ratio and fuel type are varied in the analysis. The result shows that the optimum cruise Mach number is about 4 if hydrogen fuel is used. Methane fuel instead of hydrogen reduces the vehicle gross weight by 33% in case of the Mach 2 vehicle.

Precooled turbojet engines are effective propulsion systems for hypersonic aircraft. However, a serious problem is that frost forms on the cooling tubes of the precooler, thereby decreasing the engine performance. This paper presents a new method for defrosting the precooler using jet impingement. The validity of the proposed defrosting method was investigated through fundamental experiments. In the experiments, the frost formed on the cooling tubes of the single-row heat exchanger was removed by jet impingement. We used the jet periodically. The jet interval is 10-50 s. In addition, the jet duration is short (about 0.1 s). Therefore, the consumption of the high-pressure air to make the jet flow is small. The coolant temperature and main How speed influences on the defrosting method effectiveness were assessed. Results show that this defrosting method is effective, especially when the coolant temperature and the main flow speed are low.

An innovative defrosting method for precooled turbojet engines are presented, and validated in this study using experimental methods. High speed gas jet was impinged on the cooling tubes of a heat exchanger for the purpose of defrosting. The coolant of the heat exchanger was liquid nitrogen, and whose temperature was 83K. The air flow speed, the air temperature and the air humidity were 1.0m/s, 23ºC and 59%, respectively. The effects of the jet duration, jet intervals and humidity of the jet gas on the heat exchange were assessed. As a result, we found that the presenting defrosting method is valid for the defrosting of the precooler.

A fundamental study of frost formation around a single cold cylinder was conducted using both experimental and numerical methods. We specifically examined the mass transfer around the cylinder under conditions in which a phase change of the vapor occurs in the flow. Through the experimental study, the mass flux to the cold surface of the cylinder was measured at a constant surface temperature (200-250 K). The results show that the mass flux decreases according to the decrease of the wall temperature below 230 K, although it increases above 230 K. This phenomenon cannot be expressed using the common equation with the Sherwood number, which excludes the vapor's phase change (condensation). Numerical studies calculated the flow over the cylinder, including the vapor's phase change. The scheme for compressible, flow was modified to solve lower speed flow. Results of calculations show that we obtained the same tendency as that of the experiment: the mass flux decreases at low temperatures where the phase change occurs.

Expansion Deflection nozzles present an attractive proposition as a replacement for conventional nozzles on launch vehicles, due to their reduced length, and altitude compensating capability. However, it has long been speculated that they suffer in the latter regard due to aspiration of the low speed flow region inside the nozzle by the supersonic jet surrounding it. This effect is investigated in this paper by direct experimental measurement of base pressures, and found to have little effect on the base pressure of the nozzle within the range of operating conditions investigated. Wall pressures were also used to calculate the efficiency of the altitude compensation within the nozzle, which was found to be between 87 and 100% for the three operating pressure ratios examined. This represents a significant improvement over conventional nozzle performance, and further conformation that wake pressures are indeed close to ambient.

This paper reports experimental studies on telescopic aerospikes with multiple disks. The telescopic aerospike is useful as an aerodynamic control device; however, changing its length causes a buzz phenomenon, which many researchers have reported. The occurrence of buzzing might be critical to the vehicle because it brings about severe pressure oscillations on the surface. Disks on the shaft produce stable recirculation regions by dividing the single separation flow into several conical cavity flows. The telescopic aerospikes with stabilizer disks are useful without any length constraints. Aerodynamic characteristics of the telescopic aerospikes were investigated through a series of wind tunnel tests. Transition of recirculation/reattachment flow modes of a plain spike causes a large change in the drag coefficient. Because of this hysteresis phenomenon and the buzzing, the plain spike is unsuitable for fine aerodynamic control devices. Adding stabilizer disks is effective for the improved control of aerospikes.

In this paper, a concept of a new variable-geometry aerodynamics device, which is designated "Multiple-Row-Disk (MRD) device," is introduced. The MRD device divides large separation region around the shaft of an aerospike into several small cavity flows with multiple disks arranged on the shaft. Experimental studies on aerodynamic characteristics of conical nose with axisymmetric cavities were conducted in order to evaluate a feasibility and a fundamental characteristics of the MRD device. It was found that the MRD device could improve not only drag characteristics compared to the conventional aerospikes, but also static longitudinal stability characteristics compared to the conical nose.

This paper describes a development study of a precooled-cycle hypersonic turbojet engine for the first stage of TSTO space plane and hypersonic airplane. With reflecting the key technologies accumulated from ATREX (expander cycle ATR engine) ground tests, the next flyable subscale engine "S-engine" is now developed. S-engine has 23 cm x 23 cm of rectangular cross-section, 2.2 in of the overall length and about 100 kg of the weight employing a variable-geometry rectangular inlet and nozzle. It produces 1.2 kN of thrust at SLS, which corresponds to (1)/(4) of the ATREX engine. Design of the hypersonic components such as the inlet, precooler and nozzle has been finished and their aerodynamic performances were verified by wind tunnel tests and CFD analyses. A prototype model of the diagonal-flow compressor whose pressure ratio is 6 was manufactured. Its rotating tests under the very-low pressure conditions are now in progress. The reverse-flow annular combustion chamber was successfully tested.
The first flight test of the S-engine is to be conducted in 2008 by the balloon-based operation vehicle (BOV) which is about 5 m in length, 0.55 m in diameter and 500 kg in weight. The vehicle is dropped from an altitude of 40 km by a high altitude balloon. After 40-s free-fall, the vehicle pulls up and S-engine operates for 30s at about Mach 2. High altitude tests of the engine components corresponding to the BOV's flight condition have been conducted. (c) 2007 Elsevier Ltd. All rights reserved.

This paper describes design study of the gas turbine engine using PDE as a main combustor of which hydrogen is used as a fuel. In the present analysis, exhaust gas condition of the simple PDE in which no turbine or other obstacles are installed which is estimated by CFD is referred as a turbine inlet condition. About 65 % of enthalpy drop at the exit of the PDE occurs during the exhaust of high temperature combustion gas behind the detonation wave. Furthermore, turbine adiabatic efficiency map of a conventional turbine is taken into account. Approximately, 51-80% of enthalpy drop through the turbine can be extracted as a turbine power. Partial filling of the PDE tube has an effect of smoothing the peak of U/C0. Copyright © 2007 by the American Institute of Aeronautics and Astronautics Inc. All rights reserved.

Development of the Balloon-based Operation Vehicle (BOV) is currently in progress for the first flight scheduled in 2006. In a series of BOV experiments, a vehicle in a wing-body configuration is lifted by a high-altitude balloon and dropped, after which the microgravity experiments will be performed onboard the vehicle under favor of the quasi-free-fall environments. Although the BOV is originally designed for the microgravity experiments, various types of experiments can also be performed in a hypersonic flight at lower altitudes. One candidate currently under review is a flight experiment of a precooled turbojet engine in reduced dimension. In this article, an overview of the BOV experiment is introduced, and the current development status of the BOV and a flight model of the precooled turbojet engine is presented. The aerodynamic load and the aerodynamic characteristics of the BOV are obtained by computational fluid-dynamic analyses and wind-tunnel experiments. (C) 2006 Elsevier Ltd. All rights reserved.

A study on mass flux of water vapor and mist around a cold cylinder (120–250K) was conducted by means of both experimental and numerical methods. The cylinder was placed in a forced convection air flow at a speed of 1m/sec. The experimental study revealed that the mass flux of the cylinder decreases rapidly under the temperature of about 200K. The mass flux at the cylinder temperature of 120K is one-sixth of that of 240K. The numerical study could simulate the mass flux around the cylinder in which we used a new phase change model with considering the transfer of the particles of mist. By this calculation we found some characteristics of the mass transfer of the cold cylinder which is unique when the temperature of the cylinder becomes cold enough to occur condensation.

A fundamental study for frost formation around a single cold cylinder was conducted using an experimental and numerical method. We focused on the mass transfer around the cylinder under the condition where phase change of the vapor in the flow occurs. By the experimental study, the mass transfer rate on the cold surface of the cylinder at a constant surface temperature (200–250K) was measured. The results show that the mass transfer rate decreases according to the decrease of the wall temperature below 230K, while it increases above 230K. This phenomenon can not be expressed by the common equation of Sherwood number in which the phase change of the vapor (condensation) is excluded. In the numerical study, we calculated the flow around the cylinder including the phase change of the vapor. The scheme for compressible flow was modified to be able to solve lower speed flow. As a result of the calculation we obtain same tendency as that of the experiment that the mass flux decreases at low temperatures where the phase change occurs.

Aerodynamic performances of a rectangular intake were investigated experimentally. After a tradeoff study of rectangular intakes whose operative Mach number is from 0 to 6, 20% external compression intake is selected as the best intake from the viewpoint of low number of actuators. Intake performances such as total pressure recovery and mass flow ratio are evaluated by wind tunnel tests. The free stream Mach number of the wind tunnel was M5.1. The size of the intake was 75mm in cowl capture height. Low ramp driving force was achieved by connecting links of the second ramp and third ramp. After the first wind tunnel test that is performed to evaluate the basic performance of the intake, the configuration of the intake is modified. Ramp length of the first ramp and the second ramp were changed to improve the total pressure recovery. Bleed from the second ramp is added. Seal mechanism between the variable ramps and the sidewall is modified. Total pressure recovery is improved from 9.9% to 21.7% by the modifications.

This paper discusses the current R&D status and plans concerning a Mach 6 turbojet engine with an air-precooling system for the first stage of a two-stage-to-orbit space plane (TSTO). An air-turbo ramjet engine with the expander-cycle (ATREX) has been designed and tested at sea level static conditions, which demonstrated the engine system and component performance. Experimental and numerical research of the components have also been conducted to build the fundamental technologies and to improve the engine performance. Some innovative ideas were proposed such as a multi-row-disk inlet and a defrosting method on precooler tubes using methanol. As the next step, the development of a subscale flight-type engine (S-engine) has started. The partial expander cycle was selected instead of the full expander cycle as the prototype engine cycle as a result of optimizing analyses. Total length and weight of S-engine are about 2.2 m and 100 kg respectively including a variable air-inlet and nozzle. Because S-engine will be tested in a flight demonstration program after ground tests, it must be designed with consideration of weight reduction as well as keeping of the high performance. The first engine flight test will be around Mach 2, using a balloon dropped test vehicle. This is scheduled in 2007 FY.

Six Sigma is the management strategy developed by Motorola to reduce defects in products. Design for Six Sigma (DFSS) is a methodology for determining the values of the design parameters, which maximize the performance of some system without tightening the material, manufacturing or environmental tolerances. This paper presents the implementation of DFSS for redesign of the LE-7 engine. Uncertainties with design parameters and operational conditions are considered in evaluating thrust performance, thrust chamber life, turbo-pump cavitation, and combustion stability. Traditional deterministic optimization results and probabilistic optimization results are compared. It is found that robustness of rocket engine is not always consistent with the extension of thrust chamber life.

An air-precooling system before compression is indispensable to extend the flight envelope and the improvement of the performance of turbo-based air breathing engines for the space plane. One of the critical problems on a shell-and-tube-type precooler is a deterioration of its heat exchange and pressure recovery performance due to the thick frost formation on its tube surface. An innovative method is proposed to mix a condensable additive like ethanol, methanol, etc. in the airflow as a defrosting system. The defrosting effectiveness and essential factors on the additive were investigated by using a small heat exchanger under two different cooling temperature conditions, that is, lower and higher cooling wall temperatures than the melting point of mixture of the water vapor and the additive. It was cleared in the test that most of the alcohols bad good effectiveness with methanol the best. This methanol addition concept was applied in the precooler of the practical ATREX engine. A methanol injection system worked well and the thick frost layer formed on the tube surface at the entrance side of precooler could be eliminated. The required methanol mass along the ATREX engine flight path is estimated to be less than 3% of fuel hydrogen consumption. (C) 2003 Elsevier Ltd. All rights reserved.

To study the dynamic response of a supersonic airbreathing engine and establish control logic for intake unstart, restart control tests were conducted at Mach 3 using a subscale engine model consisting of an axisymmetric intake (inlet) and a turbojet. Assuming that the combustion flame is blown out by intake unstart, restart control follows a sequence. First, after a flow is started the turbojet engine is ignited. Second, the intake is started while the rotational speed and the combustion gas temperature of the core engine are controlled. Third, the intake spike position and the terminal shock position are controlled, and the intake total pressure recovery achieves the design value (60%). The tests were successful, and engine thrust was recovered at approximately 30-40 s after engine start-up. A sudden increase in combustion gas temperature and rotational speed occurred after intake unstart. To reduce the sudden increase in the gas temperature, a new sequence that involved closing a fuel control valve after detection of intake unstart was implemented, and the increase in gas temperature was reduced. To avoid intake buzz, buzz margin control using a bypass door was successfully implemented.

The effectiveness of a method to improve the precooler performance under frosting condition was investigated by experiments on a sub-scale heat exchanger model. Addition of a methanol proved to be most effective compared with other possible substances in both cases of low and high cooling wall temperature. Then the effectiveness of the methanol addition was ascertained for the practical condition that means the same tube configuration and flow velocity as the precooler designed for the ATREX engine firing test model. The result showed that the addition of the same quantity as the water vapor could restrain the frost layer from choking the flow in the duration of 300 seconds, which is sufficient time for precooler operation. The required methanol mass along the ATREX engine flight path was estimated to be less than 3% of fuel hydrogen on board. Accordingly, the method came to be promising candidate for practical application.

A control system of variable geometry mixed compression axisymmetric intake is experimentally studied at ONERA S3 supersonic wind tunnel. The acceleration/deceleration of the space plane is simulated by changing the free stream velocity. The intake is successfully controlled with 90% of the maximum total pressure recovery and mass capture ratio. In this experiment, two subjects about control of axisymmetric intake are also cleared. First, the effect of the trapping of the terminal shock by bleed holes causes the disturbances in the terminal shock control system. Second, a special compression form change operation is necessary when the intake compression form change from all external compression to mixed compression.

Feasibility studies were carried out aiming at the application of carbon/carbon (C/C) composites to a turbine disk, heat exchangers, and a plug nozzle for an engine intended for use in a future reusable space vehicle. In these applications, the maximum temperature was estimated to be about 1500degreesC. In order to withstand this high temperature, attempts were made to utilize three-dimensionally reinforced C/C composites. The most serious problem encountered in the application of C/Cs to the turbine disk was the loss of fragments of the composite located near the outer periphery due to strong centrifugal force, which resulted in severe vibration due to rotational imbalance. The heat exchangers and plug nozzle have complex shapes in order to realize a large heat exchanging area. Joined structures were explored for these components. The principal effort in these applications has been placed on finding structures requiring low joining strength and developing materials with low gas leakage.

A feasibility study of three-dimensionally fiber-reinforced carbon-carbon composites (3DC/Cs) for application to a turbine disk of ATREX (Air turbo ramjet engine with expander cycle) was carried out. Spin burst tests at room temperature were conducted using 3D-C/C disks, and the fracture behaviors were characterized. A 3D-C/C disk was totally fractured at a peripheral speed of 516 m/s (r = 150 mm), which is sufficient for the ATREX application. However, fiber bundles at the disk periphery prematurely suffered micro-scale damage, and fragments of the fiber bundle unit flew out before total fracture occurred. In order to prevent the fly-out behavior, the disk was impregnated with Si only near its periphery. Although this treatment increased the initiation speed of the fly-out behavior, this improvement was considered insufficient for purposes of the ATREX application. Next, a simplified analysis was conducted to characterize the fly-out behavior. Based on this analysis, the following three measures were discussed: (1) decreasing bundle thickness (i.e. using fine fiber texture), (2) increasing toughness of the fiber bundle interface, and (3) minimizing local curvature in waviness of the fiber bundles in the circumferential direction.

The flight of Spaceplane is always under accelarating in the assent way and always under decelarating in the desent way and yet cruising in the return way. Besides, its flight envelope is considerably wider than that of airplane. Thus the integrated design method is required to build the best transportation system optimized taking into account the propulsion system and the airframe under the entire flight conditions. In this paper it is shown an optimization method on TSTO spaceplane system. Genetic algorithm (GA) was applied to optimize design parameters of engine, airframe, and trajectory simultaneously. Several types of engine were quantitatively compared using payload ratio as an evaluating function. It was concluded that precooled turbojets is the most promising engine for TSTO among Turbine Based Combined Cycle (TBCC) engines.

A feasibility of ATREX (Air-Turbo-Ram Expander cycle) engine with conventional aft-turbine configuration has been studied to be developed in about 10 years, if the development project has started under enough resources. The novel tip-turbine of the original ATREX engine is replaced by a conventional aft-turbine, and the maximum turbine inlet temperature (TIT) is reduced to 1200K, to realize the engine by only using approved metal technologies of modem jet engines. The capability of the performance has been shown by parametric studies by changing components' design parameters. The study shows that the performance of the ATREX engine is not less than that of pre-cooled turbo jet.
Some technical issues on developing the new ATREX engine have been addressed. The most important issue would come from the transient total temperature change due to the rapid acceleration from sea level static (SLS) condition (288K) to Mach 6 at 30km of altitude (1680K) in 6 minutes. The deformation due to transient thermal expansion has to be controlled. Especially, the change of the tip clearance and the clearance between rotors and stators are pointed out to be important design issues. The ATREX engine, which has shorter axial length and simpler rotor, has structural advantage over turbo jet. (C) 2002 International Astronautical Federation. Published by Elsevier Science Ltd. All rights reserved.

Here is presented an experimental and analytical study on a precooler for hypersonic air-breathing engines. Precooling of the incoming air breathed by an air-inlet gives extension of the flight envelope and improvement of the thrust and specific impulse. Three precooler models were installed into an air-turbo ramjet engine and tested under the sea level static condition. When the fan inlet temperature was down to 180K, the engine thrust and specific impulse increased by 2.0 and 1.2 times respectively. Thick frost formed on the tube surfaces at the entrance part of the precooler blocked the air-flow passage. On the other hand, very thin frost formed at the exit part because the water vapor included in the air was changed to mist particles due to the low temperature of the air in this part. Parametric studies on the precooler design values and a sizing analysis were also performed. Decrease of tube outer diameters on the precooler is only way to increase heat exchange rates without increase of its weight and pressure loss.

A review of the development of a precooler for the air-turboramjet expander-cycle (ATREX) engine is given. Three types of precooler for the ATREX engine ground-test model were designed, manufactured, and tested under sea-level static conditions. The results suggested two problems affecting the precooler performance, heat transfer rate and airflow pressure drop. One is nonuniformity of the airflow through the tube banks. The other problem is frost formation on the heat transfer surfaces. Concerning nonuniformity of airflow, the shell configuration was modified based on analysis by computational fluid dynamics calculation. To improve the precooler performance under frosting condition, a new method to add a condensable gas into the airflow was proposed and examined by experiments on a subscale heat exchanger model. Addition of a small quantity of ethanol can effectively restrain the decline of the precooler performance due to frost formation.

The study on ATREX engine (Air-Turbo Ramjet engine) development is being been conducted in ISAS since 1986 as a candidate fur the propulsion system of the fly-back booster up to Mach 6 on the reusable TSTO space plane. ATREX is a fan-boosted ramjet engine using liquid hydrogen as a fuel and coolant. Sea-level static firing tests of ATREX with precooling were carried out in parallel with the wind tunnel tests on the aerodynamic components such as air intake and plug nozzle. Further studies on precooler have been conducted not only for performance improvement but also for weight reduction to meet flight requirements.
In 1998, a new type precooler (Type-III) was designed and tested. Its heat transfer performance could be improved by increasing its compactness using tubes of smaller in diameter as well as its weight could be reduced. On the other hand, some difficulties increased in its manufacturing due to larger number of tubes required. The new precooler performance on heat transfer and pressure loss compared with the old type of precoolers are presented here. The test results of ATREX engine installed with new precooler is presented.
Some progress in aerodynamic studies on ATREX engine is also presented here. The control of the axisymmetric air inlet by capturing the terminal shock waves acid the reduction on boat tail drag of the plug nozzle were conducted in the wind tunnel. (C) 2001 Elsevier Science Ltd. All rights reserved.

Air intake is one of the most important components for an airbreathing propulsion system of supersonic and hypersonic vehicles. Air intake can be evaluated by air mass capture ratio and total pressure recovery ratio. In higher Mach number flight condition, larger total pressure losses occurs in the compression processes of air intake and reduces the propulsion performance. By utilizing the precompression coming from oblique shocks generated underneath vehicle forebody, a part of functions loaded in air intake can be substituted by the forebody precompression, thereby overall propulsive performance is able to be improved effectively. In the present paper, the precompression effects given by nose shape of forebody and geometrical arrangement of air intake underneath fuselage were analyzed by CFD calculation using 3-dimensional compressible Navier-Stokes equations.

This is the status report of the development study on ATREX engine (Air Turbo Ramjet) that is now under way in the Instutute of Space and Astronautical Science (ISAS) in cooperation with the Ishikawajima Harima Heavy Industries (IHI), the Kawasaki Heavy Industries (KHI), the Mitsubishi Heavy Industries (MHI). ATREX engine will be applied for the propulsion system of fly-back booster of TSTO space plane. ATREX is the fan-boosted ramjet producing the effective thrust from sea level static to flight Mach number 6. ATREX is worked on the expander cycle with precooling incoming air as shown in Fig.1. ATREX employs the tip turbine configuration which allows compactness and light weight of turbo machinery and the variable geometry airintake and plug nozzle which allows the wide range flight conditions.
ATREX development study has been conducted with the sea level static tests since 1990. ATREX engine for tests is the scaled model designated by "ATREX-500" of which fan inlet diameter is 300 mm and overall length 2,200 mm. From 1992 have been performed the wind tunnel tests on the primary aerodynamic components such as the axisymmetric variable geometry air intakes, the precoolers and the variable geometry plug nozzles. The application study on advanced carbon-carbon composite for the tip-turbine and fan assembly has been conducted. This study status is presented in the another session of this IAF congress(1).
The flight test of ATREX is now planning to verify the engine performance and functions in the practical flight conditions by using an unmanned flying test bench.
In 1995 was tested ATREX-500 installing the precooler under the sea level static conditions to examine the engine performance and the icing problem on the precooler.
The present paper addresses primarily the test results of precooled ATREX engine. (C) 1997 International Astronautical Federation. Published by Elsevier Science Ltd.

This is the status report of the development study on ATREX engine (Air Turbo Ramjet) that is now under way in the Institute of Space and Astronautical Science (ISAS) cooperation with the Ishikawajima Harima Heavy Industries (IHI), the Kawasaki Heavy Industries (KHI), the Mitsubishi Heavy Industries (MHI). ATREX engine will be applied for the propulsion system of fly-back booster of TSTO space plane. ATREX is the combined cycle (a fan-boosted ramjet) engine providing the effective thrust from sea level static to flight Mach number 6. ATREX is worked on the expander cycle with precooling the incoming air as shown in Fig. 1. ATREX employs the tip turbine configuration which allows the compactness and the light weight of turbo machinery and the variable geometry airintake and plugnozzle which allow the wide range operation conditions. From 1990 to 1992, "ATREX-500" has been tested at the sea level static conditions. ATREX-500 is the 1/4-scale model of which fan inlet diameter is 300 mm and overall length 2,200 mm. From 1992 have been performed the wind tunnel tests on the primary components of ATREX, the axisymmetric variable geometry airintakes, the precoolers and the variable geometry plug nozzles. In parallel to the windtunnel tests, the ram combusters have been tested simulating the hypersonic flight conditions and the application studies on advanced carbon-carbon composite for the tip-turbine and fan assembly has been proceeded. In 1994 initiated the flight test plan in which ATREX will be verified in the practical flight conditions by using an unmanned flying test bench. In 1995 will be tested ATREX-500 installing the precooler under the sea level static conditions to examine the engine performance and the icing on the precooler. The present paper addresses the high loading ram combuster experiment using the mixer with skewed lobes to generate swirl flow and the analytical studies and the designs on the precooler and the precooled ATREX engine and the flight test plan. © 1998 Published by Elsevier Science Ltd. All rights reserved.

Unstart phenomena due to compound choking of airframe-integrated scramjet engine inlets were investigated numerically. The compound choking is a phenomenon caused by interaction between the main flow and the boundary layer flow ingested into engines. The numerical analysis was performed by using a one-dimensional two-stream-tube model and a quasi-three dimensional MacCormack differential model. The two-stream-tube model allowed to predict the choking condition at the throat from the given inlet conditions of stream tubes. The results of both analyses coincide qualitatively, and it is shown that the boundary layer flow spreads and pushes out the main flow in engine inlets, and that engines designed assuming a uniform inlet flow can be made unstart easily because of the existence of a low flow speed region like a boundary layer. When side walls of engines sweep backward, the transition from start to unstart is rather continuous and the inlet performance at unstart conditions is not so badly deteriorated as the performance of inlets without sweep.