Search

The microwave arcjet thruster uses microwave energy to create a free-floating plasma discharge within a microwave resonant cavity. This discharge typically absorbs 99% of the input power and converts it to thermal energy which is then transferred to the flowing propellant gas. Recent modifications have allowed the thruster to be operated in a fixed configuration where neither the cavity geometry nor the tuning mechanisms are adjusted. The prototype has demonstrated its ability to operate in this fixed configuration using a variety of propellant gases, i.e., nitrogen, helium, ammonia, and hydrogen. The current design is capable of efficient operation over a wide range of power levels (250 W to over 6000 W). Current work is focused on obtaining LIF velocimetry data of the velocity profile at the exit plane of the nozzle.

There is no definitive knowledge of which of several concurrent processes ultimately results in unstable combustion within liquid rocket chambers employing shear coaxial injectors. Possible explanations are a detrimental change in the atomization characteristics due to a decrease in the gas-to-liquid velocity ratio, a change in the gas side injector pressure drop allowing acoustic coupling to the propellant feed system or the disappearance of a stabilizing recirculation region at the base of the LOX post. The aim of this research effort is to investigate these proposed mechanisms under conditions comparable to actual engine operation. Spray characterization was accomplished with flash photography and planar laser imaging to examine the overall spray morphology and liquid jet breakup processes and with a PDPA to quantify the spatial distribution of droplet size and mean axial velocity. A simplified stability model based on the Rayleigh criterion was constructed for the flow dynamics occurring within the chamber and injector to evaluate the potential coupling between the chamber and injector acoustic modes and was supported by high frequency measurements of chamber and injector pressure oscillations. To examine recirculation within the LOX post recess, velocity measurements were performed in the recess region by means of LDV. Present experiments were performed under noncombusting conditions using LOX/GH2 stimulants at pressures up to 4 MPa.

Results are presented from preliminary tests conducted to evaluate the performance of a prototype microwave electrothermal thruster. The primary component of the device is a microwave resonant cavity. The device produces stable axial plasmas within a pressurized section of the cavity with the plasma positioned in the inlet region of the nozzle. Plasma stability is enhanced by axial power coupling, an optimal distribution of electric power density within the cavity, and a propellant gas flow which has a large vortical velocity component. The thruster has been operated with a number of propellant gases: helium, nitrogen, ammonia, and hydrogen. Plasmas can be formed in a reliable manner at cavity pressures of 1 kPa and incident power levels ranging from 50 W to 350 W, depending on the gas used, and can be operated at pressures up to 300 kPa at power levels up to 2200 W. Ideal performance results of vacuum Isp and thermal efficiency vs. specific power are presented for each gas. Representative results of this preliminary work are: He - Isp = 625 s, eta-thermal = 90 percent; N2 - Isp = 270 s, eta-thermal = 41 percent; NH3 - Isp = 475 s, eta-thermal= 55 percent; H2 - Isp = 1040 s, eta-thermal = 53 percent.

Although stable operating regimes for cryogenic coaxial injectors have been empirically determined, there is no knowledge of the spray characteristics corresponding to stable operation, or the physical processes which produce the atomization patterns that result in stable or unstable operation. The current engineering method for determining the stable operating regime of a cryogenic coaxial injector is the 'hydrogen temperature ramping' method, however there is no definitive knowledge of whether the hydrogen temperature influences the chamber stability by decreasing the injected gas velocity, by affecting a recirculation region at the base of the LOX Post, or by changing the pressure drop across the injector, allowing chamber pressure oscillations to couple to the fuel feed system. Results for the injector response from a linearized lumped-element model are presented as a function of temperature and frequency. LDV measurements in the recess region at the base of the LOX post show reverse flow indicative of a recirculation region. Finally, Phase Doppler Particle Analyzer (PDPA) measurements of droplet size and velocity distributions are discussed for a coaxial injector element similar to the SSME preburner element operating with water and air at atmospheric pressure and liquid and gaseous nitrogen at 20 bars.

A numerical analysis of low Reynolds number nozzle flows was performed to investigate the loss mechanisms involved and to determine the nozzle wall contour that minimizes these losses. DSMC was used to simulate flows through three different nozzle configurations at two different stagnation chamber temperatures so that the heat transfer losses could be separated from the wall contour effects on performance. A trumpet-shaped nozzle had 5 percent higher efficiency than a conical nozzle and a 3 percent higher efficiency than a bell-shaped nozzle with the unheated flow. With heated flow both the trumpet and bell-shaped nozzles had a 6.5 percent higher efficiency than the conical nozzle. The conical nozzle had the highest discharge coefficient of the three configurations, 0.92, and the trumpet-shaped nozzle had the lowest, 0.82. The discharge coefficient of each nozzle was unaffected by the change in stagnation temperature; however the increase in stagnation temperature increased the heat transfer and viscous losses in the boundary layer. These results suggest that the trumpet-shaped wall contour performed most efficiently except near the throat region, where it incurred large viscous losses. However, the bell-shaped nozzle may increase its overall performance with an increase in stagnation temperature.

A new approach to numerically solving the problem of the constricted electric arcjet is presented. An Euler Implicit finite difference scheme is used to solve the full compressible Navier Stokes equations in two dimensions. The boundary and initial conditions represent the constrictor section of the arcjet, and hydrogen is used as a propellant. The arc is modeled as a Gaussian distribution across the centerline of the constrictor. Temperature, pressure and velocity profiles for steady state converged solutions show both axial and radial changes in distributions resulting from their interaction with the arc energy source for specific input conditions. The temperature rise is largest at the centerline where there is a the greatest concentration arc energy. The solution does not converge for all initial inputs and the limitations in the range of obtainable solutions are discussed.