Published papers

A cylindrical rotating detonation engine (24 mm diameter) with multiple injector holes on the combustor side wall for both propellant injection and cooling by injector gas flow was designed and tested. As each set of injectors created its own detonations, the synchronized detonation waves propagated along the three injector rows. From the axial view, those detonation combustion regions stood off from the injector surface as the non-well-mixed propellant existed between the wall and the combustion zone. This combustion region tended to broaden in lateral and radial directions as the mass flow rate increased by pushing its region with the injected propellant. Wall heat flux and heat balance were evaluated by a one-dimensional unsteady heat transfer model with propellant injection cooling. Even when the flow rate was doubled, the increase in the wall heat flux was only 18–25%. This heat trend and the image of standoff self-chemiluminescence from the injector surface implied that a non-well-mixed unburned propellant acted as a heat-reduction layer to ease heat load into the combustor. Measurements and thermal analysis verified the flow structure near the injector and heat-exchange mechanism due to the propellant gas flow, which has a potential for thermal steady operation.

This paper presents the results of an S-shaped pulse detonation engine (PDE) ground firing test in the form of a detonation engine system. The world’s first technology demonstration of PDE in space using a sounding rocket is planned, and the aim is to control the rocket spin rate in the axial direction using pulsed detonation. The PDE operation at full sequence was successful. The despin rate change of the rocket between continuous oxygen supply and successful PDE operation is expected to be 0.95 deg ∕s per run. This change in despin rate can be measured by an onboard gyro sensor, making the system flyable. The test results were compared with data from thrust measurement tests conducted in a laboratory, the results of which confirmed the thrust generation under an ambient pressure of 0.5 0.1 kPa. The average thrust values in the thrust measurement experiments showed good agreement of 101 3% with a quasi-steady-state model introduced to predict the PDE thrust. These results demonstrate the feasibility of the newly developed PDE and its system as the world’s first technology demonstration of detonation propulsion in space.

A rotating detonation engine (RDE) generates a continuous thrust with one or more rotating detonation waves. Because of the velocity on the order of kilometers/second, the reaction zone is relatively small. Therefore, the RDE realizes a short combustion chamber length. However, the detonation waves induce an azimuthal motion of propellant, resulting in torque around the thrust axis. Because the motion does not contribute to the thrust, the torque is important in terms of performance loss. Herein, we conducted combustion tests with a six-axis force sensor to simultaneously measure 0.149 0.009 Nm torque and 48.1 0.9 N thrust. A comparison of detonation waves captured by high-speed camera revealed that the torque followed the direction and was offset when the waves existed in both of two directions simultaneously, which indicates the possibility of controlling the torque. Under a mass flow rate at 87 9 g∕s and an equivalence ratio at 1.43 0.28, when the azimuthal component of shear force was 8.8 0.6% of the thrust, 0.77 0.10% of the total kinetic energy of the exit flow was distributed to the azimuthal component of velocity and did not contribute to the thrust. We therefore concluded that the effect of the azimuthal motion on the RDE’s performance was small.

The blast-mitigation effect produced by water droplets sprayed into a confined geometry was studied. When a one-dimensional blast wave interacts with a layer of water droplets, blast mitigation occurs that can be expected to reduce the damage from an accidental explosion. Focus was on the blast-mitigation effect produced by the quasi-steady drag force between the shocked air and the stationary water droplets. Momentum transfer due to the drag mitigated the blast wave, and the maximum blast-wave mitigation occurred just behind the layer of droplets. The study parameters (the volume fraction of water droplets, the location and length of the region with sprayed water droplets, and the mass of the high explosive) demonstrated that higher momentum loss from the air resulted in greater mitigation. The momentum loss was evaluated using the volume fraction and length of the layer of water droplets and the quasi-steady drag force when the shock wave reached the layer. Relationships were formulated between the initial conditions, the momentum loss, and the blast-mitigation effect, which made it possible to evaluate quantitatively the mitigation effect on the blast wave caused by the water droplets.

This Paper presents the results of a rotating detonation engine rocket sled test demonstration on a 127 mm gauge rail at 100 m long high-speed running track test equipment to evaluate the thrust performance of a stand-alone rotating detonation engine propulsion system. The developed sled test demonstrator contained all onboard subsystems, including engine, gas cylinders, valves, controller, and a data logger. The sled test demonstrator successfully generated thrust, and three or four detonation waves of 1460 to 1800 m∕s were observed during the 2.0 s burn time. The measured specific impulse in the sled test was 144 s as estimated from the acceleration of the demonstrator and the mass decrease in the onboard cylinders. The test results were compared with the data from a laboratory-stand firing test using a load cell, and the specific impulses of the sled test were found to be comparable to the results of the stand firing test.

Geometric throats are commonly applied to rocket combustors to increase pressure and specific impulse. This paper presents the results from thrust measurements of an ethylene/gas-oxygen rotating detonation engine with various throat geometries in a vacuum chamber to simulate varied backpressure conditions in a range of 1.1–104 kPa. For the throatless case, the detonation channel area was regarded to be equivalent the throat area, and three throat-contraction ratios were tested: 1, 2.5, and 8. Results revealed that combustor pressure was approximately proportional to equivalent throat mass flux for all test cases. Specific impulse was measured for a wide range of pressure ratios, defined as the ratio of the combustor pressure to the backpressure in the vacuum chamber. The rotating detonation engine could achieve almost the same level of optimum specific impulse for each backpressure, whether or not flow was squeezed by a geometric throat. In addition, heat-flux measurements using heat-resistant material are summarized. Temporally and spatially averaged heat flux in the engine were roughly proportional to channel mass flux. Heat-resistant material wall compatibility with two injector shapes of doublet and triplet injection is also discussed.

Supersonic accelerators use sabots to suppress bore balloting. However, sabot separation induces flight deviation in projectiles owing to their interactions with fluids and shock waves. A seamless calculation of the acceleration phase in the tube and the full separation process using 3-D computational fluid dynamics coupled with rigid body dynamics is presented in this study to investigate the shock-wave interactions around a railgun-launched projectile. The railgun acceleration induced a normal shock wave propagating ahead of the projectile; when the projectile reached the tube end, a substantial expansion at the muzzle generated two shock waves, namely, a spherical precursor shock wave and a Mach disk, outside the tube. Aerodynamic forces on the projectile/sabot decreased to almost zero just after the tube exit, with the ambient fluid around the tube being faster than the projectile by the substantial expansion. After exiting, the projectile penetrated the precursor shock wave at t 0.15 ms; this increased the aerodynamic forces acting on the projectile, initiated sabot separation, and generated multiple shock-wave interactions, which induce unsteady aerodynamic loads on the projectile.

© 2018 Elsevier Ltd. A two-dimensional (2D) numerical simulation is conducted to clarify the propagation behavior of gaseous detonation in a water spray and its structure. The computational target refers to the experiment conducted by G. Jarsalé et al., and C2H4-air gaseous detonation propagates where the water droplets (WDs) are sprayed. The parameters used are the C2H4-air equivalence ratio and WD mass fraction. The flow field, Favre-averaged one-dimensional profile, and cellular structure are revealed in 2D simulations. Stable propagation of gaseous detonation is observed in the water spray, and the decrease in velocity relative to the Chapman-Jouguet velocity without WDs is as much as 3.2%. Adding WDs changes the cellular pattern, especially for leaner mixtures. The weak triple point decays, and the cell width increases because of the longer induction length due to decreased velocity. The WD presence changes the detonation flow field substantially, and evaporation occurs primarily at 10 mm behind the shock wave. The high-evaporation region propagates at the detonation speed, and the compression wave formed when the detonation reflects from the two-phase medium propagates backward. Furthermore, WD evaporation suppresses the velocity, vorticity, and temperature fluctuations. Rapid evaporation with WDs leads to lower hydrodynamic thickness than that without WDs or in the Zel'dovich-von Neumann-Döring model.

© 2018, Springer-Verlag GmbH Germany, part of Springer Nature. Two-dimensional explosion experiments were conducted to discuss the interaction between a blast wave and sand and show the mitigation effect of the sand on the blast wave. The explosive used was a detonating cord 1.0 m in length, which was initiated in a sand hill shaped like a triangular prism and whose cross section was an isosceles triangle with base angles of 30∘. Sand-hill heights of 30 and 60 mm were used as parameters to discuss the effect of sand mass upon blast-wave strength. The interaction of the blast wave with the sand/air interface causes multiple peaks in the blast wave, which are induced by successive transmissions at the interface. The increase in the sand mass further mitigates the blast parameters of peak overpressure and positive impulse. The results of this experiment can be utilized to validate the numerical method of solving the problem of interaction between a compressible fluid and a particle layer.

© 2019, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. In this paper, the effects of projectile acceleration profile on interior and transition ballistics are investigated in order to design HVP (Hyper-Velocity projectile) for railgun. Our railgun facilities accelerate projectile to 600 – 2501 m/s in order to evaluate the effect of exit velocity on sabot separation. However, sabot melt prevents us from observing detail sabot separation and shock wave interactions. Thus, in this paper, CFD-RBD method is applied to calculate the projectile acceleration and sabot separation motion due to unsteady aerodynamic force. In the interior calculation, acceleration behavior is approximated by the cubic spline interpolation using our experimental result in Vexit = 658, 1268, 1875, 2048, 2501 m/s. Firstly, in the interior ballistic calculation, the projectile and sabot induce compression waves due to acceleration. In 0.6 ms, the single normal shock wave stands ahead of projectile. The precursor shock wave reaches the end of tube and propagates outside of tube. When the precursor shockwave propagates outside of tube, Mach disk stands just before the precursor shock wave due to expansion. Especially, the cases of Vexit ≥ 1875 m/s show similar pressure distribution and precursor shock wave propagation outside of tube. On the other hand, supersonic exit velocity, in Vexit = 1268 m/s, increases distance between precursor shock wave and Mach disk. Also, in Vexit = 658 m/s, although Mach disk is close to the muzzle, precursor shock wave propagates far away from Mach disk, which generates completely different pressure distribution. Secondary, in transition ballistic calculation, sabot separation motion does not depend on projectile exit velocity in our railgun configuration. From this result, the design of HVP need not to consider sabot trajectory variation even if exit velocity changes from supersonic (Vexit = 658 m/s) to hypersonic (Vexit = 2501 m/s) regime.

© 2019 by Nagoya University. This study evaluated the propulsion performance of a nozzleless, cylindrical rotating detonation engine (RDE). Using a C2 H4 –O2 mixture, the RDE was tested in a low-back-pressure environment at propellant mass flow rates of 8–45 g/s. In high-speed imaging of the self-luminescence within the combustor, rotating luminous regions were observed at mass flow rates above 22 g/s. Measured pressure distributions suggest that burned gas reached sonic velocity at the combustion chamber outlet. This paper proposes the structure of internal flow in the RDE and confirms that calculated pressure distribution based on the structure was close to the experimental distribution. This study also estimated the RDE’s thrust by pressure and momentum exchange and confirmed it by experimental measurement. Moreover, the theoretical thrust calculated under the assumption that exhaust is a sonic flow agreed with the load cell thrusts, suggesting that RDE combustion is perfectly completed inside the chamber. Specific impulses are 80–90% of specific impulses for ideal correct expanded flow for all mass flow rates, and its value was close to that of an annular RDE. In addition, RDE performance will increase by about 20% if the RDE is equipped with a divergent nozzle and the gas is correctly expanded to back pressure.

© 2018 The Combustion Institute. Published by Elsevier Inc. All rights reserved. We describe the critical condition necessary for the inner cylinder radius of a rotating detonation engine (RDE) used for in-space rocket propulsion to sustain adequate thruster performance. Using gaseous C2H4 and O2 as the propellant, we measured thrust and impulse of the RDE experimentally, varying in the inner cylinder radius ri from 31 mm (typical annular configuration) to 0 (no-inner-cylinder configuration), while keeping the outer cylinder radius (ro = 39 mm) and propellant injector position (rinj = 35 mm) constant. In the experiments, we also performed high-speed imaging of self-luminescence in the combustion chamber and engine plume. In the case of relatively large inner cylinder radii (ri = 23 and 31 mm), rotating detonation waves in the combustion chamber attached to the inner cylinder surface, whereas for relatively small inner cylinder radii (ri = 0, 9, and 15 mm), rotating detonation waves were observed to detach from the inner cylinder surface. In these small inner radii cases, strong chemical luminescence was observed in the plume, probably due to the existence of soot. On the other hand, for cases where ri = 15, 23, and 31 mm, the specific impulses were greater than 80% of the ideal value at correct expansion. Meanwhile, for cases ri = 0 and 9 mm, the specific impulses were below 80% of the ideal expansion value. This was considered to be due to the imperfect detonation combustion (deflagration combustion) observed in small inner cylinder radius cases. Our results suggest that in our experimental conditions, ri = 15 mm was close to the critical condition for sustaining rotating detonation in a suitable state for efficient thrust generation. This condition in the inner cylinder radius corresponds to a condition in the reduced unburned layer height of 4.5-6.5.

© 2018 The Combustion Institute This article reports on an experimental investigation into dynamical behaviours of detonation in non-uniform mixtures, generated from stoichiometric propane–oxygen, oxygen and ethane, with initial temperature and pressure 290 K and 20 kPa, respectively. Composition gradients are parallel to the direction of detonation propagation, with an equivalence ratio (ER) that first decreases from lean values and then increases to rich ones. Composition distributions are characterized according to the depth of the ER sink. Gradients are generated in a 50 × 50-mm2-square cross-section and a 665-mm total length chamber. The mixture components are injected separately in the pre-evacuated chamber in their order of decreasing density through porous plates at the chamber top-end to ensure planar filling of the chamber. Non-uniform distributions are then precisely controlled as a function of time by means of optical oxygen sensors. A Chapman–Jouguet (CJ) detonation is transmitted at the chamber bottom-end from a 3.6-m-long driver tube. Fast pressure transducers, sooted plates and Schlieren visualizations coupled with high-speed cameras are used to characterize the longitudinal velocity, cellular structure and transmission, failure and re-initiation mechanisms of the detonation front. Shallowest ER sinks produce the supercritical transmission mode of the CJ detonation with continuous adaptation of velocity and multicellular structure to local composition. Deepest sinks lead to the subcritical behaviour characterized by sudden detonation failure from shock-flame decoupling when ER decreases, and without detonation re-initiation when ER increases again. Intermediate sink depths generate critical behaviour with detonation re-initiat ion at chamber walls from expanding combustion kernels and reflected transverse Mach waves and then from SWACER retro-active mechanism. An elaboration of the failure criterion used in a previous study is found to well predict conditions for shock-flame decoupling.

© Copyright 2017 by Nagoya University. A pulse-detonation thruster can generate a high-repeatability small impulse at a high operating frequency. To operate a pulse-detonation cycle in a vacuum environment without a purging material, a liquid-purge method proposed by Matsuoka et al. ("Development of a Liquid-Purge Method for High-Frequency Operation of Pulse Detonation Combustor," Combustion Science and Technology, Vol. 187, No. 5, 2015, pp. 747-764) and throat at the exit of the combustor were introduced. In the demonstration experiment, gaseous ethylene-liquid nitrous oxide, as detonable mixture, and a throat having an inner diameter of 3.6 mm (blockage ratio = 87%) were used. The measured cyclic flame-propagation speeds were 2005 ± 90 m/s and 92 ± 4% of the estimated Chapman-Jouguet detonation speed. Consequently, a 50 Hz pulse-detonation operation without a purging material in the ambientpressure range of 0.035-1.5 kPa was confirmed. A quasi-steady model, in which gases in the combustor are in the stationary state, was newly developed to investigate the operating characteristic of the pulse-detonation thruster. The experimental pressure history in the combustor during the burned-gas blowdown process was in good agreement with that of the model. Moreover, using the model, the thrust performance of the pulse-detonation thruster with a converging-diverging nozzle was investigated. It was found that the estimated specific impulse was comparable with that of the theoretical steady-state rocket engine.

© 2018 Elsevier Ltd. We used the computational fluid dynamics-discrete element method (CFD-DEM) model of compressible flow to numerically investigate the flame structure during shock-wave-induced layered coal-dust combustion, which poses a significant risk in coal mines. This represents the first attempt to apply a CFD-DEM model to compressible reactive gas-particle flow. The Eulerian-Lagrangian model for compressible gas-particle flow was extended to simulate shock-particle interactions in a reactive flow field. The particle interactions were predicted by the DEM, and source terms for homogeneous and heterogeneous reactions were included in the governing equations. The calculations were validated for inert particle dispersion from shock-particle interactions and flame propagation velocity in layered coal-dust combustion. The results were consistent with those from previous experiments. Furthermore, the flame structure in layered coal-dust combustion was revealed using the CFD-DEM approach. The simulation of the layered coal-dust combustion indicated that the shock wave was initially generated by gas detonation in a narrow channel with coal-dust particles at the bottom. The predicted propagation mechanism during layered coal-dust combustion was consistent with that reported in a previous numerical study based on the Eulerian-Eulerian approach (Hoium et al., 2015). Moreover, the flame comprised a leading shock wave and diffusion flame; the coal-dust particles were dispersed and heated by the shock wave and combustion products, respectively. The diffusion flame structure propagated at 350-500 m/s, resulting in devolatilization behind the reaction front. However, the CFD-DEM results indicated that the particle dispersion heights were higher than those predicted by the Eulerian-Eulerian approach (despite similarities in the inert particle dispersion results of these methods), attenuating the compression wave from the reaction front and slowing the leading shock wave.

© 2018 The Combustion Institute. As a novel detonation combustor that differs from a pulse and a rotating detonation engine, a reflective shuttling detonation combustor (RSDC), in which detonation waves shuttle repeatedly, was proposed. In a fan-shaped two-dimensional combustor, detonation waves propagate, repeating attenuation and re-ignition by a shock reflection at the side wall. In the demonstration experiment, chemiluminescence visualization and pressure measurement with ethylene-oxygen mixture were conducted at the same time. As the result, a single shuttling wave coupled with pressure rise was observed in the combustor. The tangential velocity of the wave was 1526 ± 12 m/s and approximately 60% of the estimated Chapman-Jouguet velocity of 2513 m/s. The ratio of pressure in front of the wave to one behind the primary wave or the reflected wave was in good agreement with one-dimensional shock theory, and it was suggested that the rapid reaction behind the reflected shock wave sustained the continuous propagation of the shock wave.

© 2018, Springer-Verlag GmbH Germany, part of Springer Nature. This paper conducts a numerical study of particle dispersion by vertical shock waves using a combination of computational fluid dynamics and the discrete element method (DEM). The Magnus force is important for particle dispersion, and the DEM approach can consider the rotation of individual particles and provide a detailed analysis of particle–particle and particle–wall interactions. Simulations are conducted and the results are compared with those of previous experiments, showing that the models can capture the particle dispersion process and shock wave geometry inside the mixed gas–particle region. During the particle dispersion process, the contact force due to the particle–particle interactions causes the particles to move upward after the shock wave passes and then decreases rapidly due to the low collision frequency in the particle cloud. The Magnus force is initially lower and has little effect until the contact force decreases, when it becomes dominant and maintains the particles’ upward movement. It is mainly driven by the gas rotation above the dust layer, and the effect of particle rotation is relatively small in comparison. The gas velocity gradient above the dust layer in the particle dispersion region is caused by gas–particle interactions. The particle dispersion region becomes wider and thinner over time, meaning that the gas velocity gradient above the dust layer becomes small and hence that the Magnus force (which is driven by the gas rotation) also becomes small. In contrast, the downward drag force remains constant during the vertical motion because it is primarily affected by the velocity difference between the particles and the gas. This means that the drag and Magnus forces eventually balance, and the particles stop rising.

© 2018, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. The numerical investigation of the gas-particle flow in the shock tube is conducted using the continuum model (Two-Fluid model) and the discrete particle model (CFD-DEM). From calculated results of both methods, particles are concentrated at downstream edge of particle bed and dispersed at upstream edge of particle bed with the passage of time. The incident shock wave is divided into the reflected shock and the transmitted shock because of interactions with particles. These flow field can be observed in the experiment conducted by Rogue et al., and Two-Fluid model and CFD-DEM can qualitatively reproduce the interaction phenomena between shock wave and particles. In the experiment, the pressure oscillations behind the reflected and the transmitted shock wave are observed as well as CFD-DEM, but no oscillation is seen in Two-Fluid model. Therefore, Two-Fluid model can qualitatively reproduce the interaction phenomena of shock wave and particles qualitatively, although the behavior of individual particle is neglected in the model.

© 2018, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. Rotating detonation engine (RDE) is a promising candidate for future upper stage motor because of its high theoretical efficiency and short combustor length. Technological demonstration in space will be necessary to evaluate the performance and flight dynamics of RDE as a space propulsion system. Toward a flight test, we built an open-structure, rocket RDE-powered vehicle to measure thrust to validate a feasibility of an onboard propulsion system. We carried out 4.4 s combustion test, and achieved over 73% of ideal specific impulse above 1.1 MPa of static pressure in the RDE combustor.