Abstract

Abstract Hybrid stress blended eddy simulations (SBESs) are performed for the AFRL rotating detonation engine burner with detailed finite rate chemistry. The near wall boundary layer flow is modeled using URANS k-ω-SST model and the turbulent internal flow is modeled using Large Eddy Simulation (LES) dynamic Smagorinsky model. Using one-dimensional (1D) detonation tubes, the chemical mechanism and the chemistry integration algorithm are validated against the Chapman Jouguet (CJ) conditions and the Zel’dovich-von Neumann-DÖring (ZND) calculations with reasonably good accuracy. The impact of spatial and temporal resolutions on the wave propagation speed (WPS) and the von Neumann (VN) peak is studied using 1D detonation tube simulations. Using second order (SO) spatial accuracy, the SO temporal scheme over-predicts the VN peak with numerical dispersion near the pressure’s peak, while the first order (FO) temporal scheme is found to add the correct numerical dissipation. The SO and FO profiles show identical wave structure at different resolutions in the expansion region downstream of the compression shock’s peak. With FO temporal scheme, the VN peak was found to vary within 2%–6% band within a spatial resolution range of 0.5–0.001 mm and within 2%–20% band within a temporal resolution range of 1E−7 s – 1E−8 s. The spatial and temporal resolutions, however, are found to have a smaller impact on the WPS (varies within 2%–5%). For three-dimensional (3D) simulations the SO and FO comparisons with data were overall comparable. The results show very good agreement with the measurements mean static pressure data, WPS, specific thrust and specific impulse. The SO temporal scheme predicted a WPS with a mean error of approximately 11%–10% while the FO temporal scheme mean error is 6%. Back-flow due to the detonation wave in the air and fuel plenums are quantified to be 10% and 15%, respectively. The impact of wall heat transfer modeling on the detonation wave is studied. As the wall temperature increases more deflagration burning in the refill zone occurs ahead of the detonation wave. The fraction of heat released by deflagration and detonation is quantified for each case using cut-off values of 5 atm and 10 atm. Although the impact of wall heat transfer on WPS is found to be small, the fraction of heat released by deflagration increased by 10% for 600 K isothermal wall temperature compared to the 300 K isothermal wall temperature. For local static temperatures > 300 K, the fraction of heat released in the deflagration mode was 75%, 53%, and 40%, for the adiabatic walls, 600 K, and 300 K isothermal walls, respectively. The current numerical study shows that accurate wall heat transfer modeling is important for rotating detonation engine (RDE) numerical simulations.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call