Non-premixed Rotating Detonation Engine Research Articles

Numerical Investigation on the Height of Detonation Wave in Rotating Detonation Engine

Three-dimensional numerical simulations were conducted on a non-premixed rotating detonation engine (RDE) with different exhaust nozzles (i.e. without a nozzle, with a divergent nozzle, with a convergent nozzle, and with a Laval nozzle). Aside from describing the flow-field structure inside the chamber, the detonation wave height of the detonation wave, and the outlet Mach number were also investigated. It is found that the convergent and Laval nozzles make the flow-fields more complex in chamber by the reflected shock waves. The height rises with the increase of air mass flow rate when detonation chamber exit is in choke state, but it decrease when detonation chamber exit is not in choke state.

The dynamics of a non-premixed rotating detonation engine from time-resolved temperature measurements

The structure and dynamics of a hydrogen-air rotating detonation engine (RDE) are described based on 100-kHz laser absorption spectroscopy measurements of water temperature at four simultaneous locations within the detonation channel. The analysis focuses on the evolution of the flowfield over a 200 ms period for three separate air mass flow rate cases. Two-dimensional unwrapped visualizations of the temperatures show a flowfield structure containing regions with the detonation front, combustion products, oblique shock, and refilling reactants, qualitatively agreeing with previous simulations and experiments. A major conclusion is that water from the combustion products is measured throughout all space and time in the RDE, including near the injector, implying the presence of performance loss processes such as burning upstream of the detonation wave or the back recirculation of combustion products with fresh fuel–air. By analyzing the elevated temperatures of the reactants during the refill process, one estimation for the mass fraction of combustion products in the reactants is as high as 20–30% on average. This product mass fraction is found to be inversely proportional to the bulk air mass flow rate and decreases as time progresses. This indicates these non-ideal processes are more significant closer to RDE ignition for poorer performing operating conditions. For the largest air mass flow case, water temperatures near the nominally cold plenum conditions likely corroborate the presence of a recirculation region on the RDE inner body. Analysis of inter- and intra-cycle temperature dynamics further support non-ideal processes occurring behind the detonation wave and during the refill process. As a whole, the data indicates that the RDE performance is better as time progresses away from ignition or for higher air mass flow rates. These data are also important for comparison with numerical models.

Detonation Structure in Ethylene/Air-Based Non-Premixed Rotating Detonation Engine

Rotating detonation engines (RDEs) are gaining significant interest as a practical approach to increasing the operating efficiency of propulsion systems through pressure gain combustion. To use RDEs in practical gas turbines, their stability and performance with regard to hydrocarbon fuels need to be tested. Prior experimental work has shown that ethylene/air systems exhibit suppressed wave propagation, with speeds that are nearly half the ideal wave speeds. The purpose of this study is to simulate realistic RDE configurations that have been experimentally studied to gain insight into this wave suppression. Detailed numerical simulations using a reduced 8-species 2-step and a 21-species 38-step chemistry model are conducted. When ethylene is diluted with hydrogen, the system reaches a stable mode comparable to detonation propagation after unsteady transition period between a weak deflagrative and a stronger detonation mode. It is further shown that pure ethylene cases stay in the weak regime with weak pressure propagation aided by deflagration along the wave path. Species, temperature, and pressure profiles normal to the wave front show a weak pressure rise followed by a broad reaction zone. As a result, heat release is not confined to the near-shock region but is distributed across the detonation chamber.

Novel Approach for Computational Modeling of a Non-Premixed Rotating Detonation Engine

Pressure gain combustion (PGC) has gained significant attention in airbreathing gas turbine applications due to its increased thermodynamic efficiency over a constant-pressure Brayton cycle. Rotating detonation engine (RDE) is a form of PGC, and in most RDEs, fuel and oxidizer are injected into the combustion chamber from separate plenums (for safety reasons), leading to an inhomogeneous mixture in the reaction zone. In this paper, a novel approach is developed to model inhomogeneous injection of the fuel–oxidizer mixture in a two-dimensional computational fluid dynamics analysis. A probability density function (PDF) of the fuel mass fraction from a converged three-dimensional, nonreacting simulation is extracted and is used as a spatially and temporally varying inlet boundary condition in the two-dimensional simulation. Using this method, a numerical analysis is carried out to predict the flow characteristics, flow structures, and detonation cell size of a non-premixed RDE, using -air as the fuel–oxidizer mixture. The simulation results are validated against experimental results in the literature, and these results from the non-premixed RDE simulations are compared against a perfectly premixed RDE, under same operating conditions. The present study provides a computationally inexpensive method to model RDE injector designs and helps elucidate the effects of inhomogeneous fuel–air mixing on RDEs.

Numerical investigation on H2/Air non-premixed rotating detonation engine under different equivalence ratios

In this paper, three-dimensional numerical simulations are performed to investigate the formation and propagation characteristics of rotating detonation wave in a non-premixed engine. By changing the mass flow rate of H2 and fixing air mass flow rate, the effects of equivalence ratio involving fuel lean and rich operating conditions are mainly discussed. Numerical results show that equivalence ratio plays a very critical role in the formation process and propagation mode, which further affects the propulsion performance of rotating detonation engine significantly. For current numerical geometry and operating conditions, the lean limit of equivalence ratio for formatting a stable RDW is about 0.4, dual-wave mode (at equivalence ratio of 0.6, 0.8, 1.0 and 1.4) and single-wave mode (at equivalence ratio of 1.2) are obtained, respectively. When equivalence ratio is 1.0, rotating detonation engine can exhibit excellent operating performance with the shortest formation time, best propagation stability, middling class thrust and specific impulse. Besides, the pressure contour analysis indicates that the effects of equivalence ratio and mass flow rate of H2 on the collision strength and times during the re-initiation process are the main mechanisms for determining the formation possibility and propagation mode of rotating detonation wave. Besides, the intensity of accumulated pressure wave and distributions of equivalence ratio are two important factors for the generation of new detonation wave front. Furthermore, it is also detected from the comparisons of the propulsion performance that the effects of equivalence ratio on thrust and specific impulse under fuel lean conditions are more significant than those under fuel-rich conditions.

Effects of air injection throat width on a non-premixed rotating detonation engine

Rotating detonation engines are widely studied because of their compact configurations and high thermal cycle efficiency. For simplification, most of the numerical simulations of rotating detonation engines used premixed reactant mixtures. The rotating detonation waves under non-premixed conditions are not studied enough. The width of air injection throat is an important design parameter for a non-premixed rotating detonation engine. Here, a series of three-dimensional numerical simulations of a non-premixed rotating detonation engine with different air injection throat widths are performed. Cases with three different throat widths of 0.4 mm, 1.0 mm and 2.0 mm are calculated. The transient explicit density-based solver in ANSYS Fluent is used to perform the simulations. Detonation combustion happens in different radial domains for different air injection throat widths. When the air injection throat width is 1.0 mm, there is only one rotating detonation wave in the combustion chamber when the total mass flow rate is 272.3 g/s or 500 g/s. When the total mass flow rate is further increased to 1000 g/s, there are two co-rotating detonation waves in the combustion chamber. While for air injection throat width 0.4 mm, the transition to two-wave mode happens at a smaller total mass flow rate of 500 g/s. When the air injection throat width is too large, the interaction between the injection process and the propagating of rotating detonation wave becomes very strong, and the detonation wave quenches due to the insufficient injected reactants. The engine requires a larger total mass flow rate to sustain the continuous propagation of the rotating detonation wave.

Investigation of the structure of detonation waves in a non-premixed hydrogen–air rotating detonation engine using mid-infrared imaging

The structure of detonation waves propagating through the annular channel of an optically accessible non-premixed rotating detonation engine (RDE) are investigated using mid-infrared imaging. The RDE is operated on hydrogen–air mixtures for a range of air mass flow rates and equivalence ratios. Instantaneous images of the radiation intensity from water vapor are acquired using a mid-infrared camera and a band-pass filter (2.890 ± 0.033 µm). The instantaneous mid-infrared images reveal the stochastic nature of the detonation wave structure, position and angle of oblique and reflected shock waves, presence of shear layer separating products from the previous and current cycles, and extent of mixing between the reactants and products in the reactant fill zone in front of the detonation wave. The images show negligible signal directly in front of the detonation waves suggesting that there is minimal mixing between the reactants and products from the previous cycle ahead of the detonation wave for most operating conditions. The mid-infrared images provide insights useful for improving fundamental understanding of the detonation structure in RDEs and benchmark data for evaluating modeling and simulation results of RDEs.

A non-premixed rotating detonation engine using ethylene and air

Only a few studies on ethylene-air rotating detonation engines were conducted due to the great difficulty of quick ethylene-air mixing. Hydrogen or oxygen was added into ethylene-air mixtures, or premixed ethylene-air mixtures were used in some previous studies, to obtain rotating detonation waves. Non-premixed rotating detonation waves of high speeds for ethylene and air sources at room temperatures were obtained in the present study, proving ethylene is feasible for air-breathing rotating detonation engines. The hollow combustor channel here had an outer diameter of 100 mm. Fuel and air were injected into the combustor from 150 cylindrical orifices of a diameter 0.8 mm and a circular gap with a throat width of 1 mm, respectively. The detonation speeds were 1256–1653 m/s for the air flow rates 465–689 g/s and equivalence ratios 0.47–1.06. Most of the speeds were above 80% of Chapman-Jouguet values. Both high-speed images and dynamic pressure sensors showed there was one rotating detonation wave in all the operating conditions.

Experimentation of Premixed Rotating Detonation Engine Using Variable Slot Feed Plenum

A premixed supply of fuel and oxidizer was fed through five concentric feed slots and detonated in a rotating detonation engine, representing the first known successful airbreathing premixed rotating detonation engine. The premixed feed system successfully arrested flashback. An operating map was constructed and compared to a nonpremixed rotating detonation engine. Mixing effects were shown to be important by comparing measured wave speeds to theoretical Chapman–Jouguet detonation velocities, published computational fluid dynamics results, and similar nonpremixed rotating detonation engines. High-speed imagery indicated that the wave speed measurements should account for the circumferential velocity of the gases relative to the detonation and that the Chapman–Jouguet calculation should account for partial mixing with combustion products.