Behavior of the thrust vector in the NSTAR ion thruster

The magnitudes and rates of offsets and changes in the thrust vector of an engineering model NSTAR thruster (EMT2) used in an 8000 hour wear test and the flight thruster (FTl) for the DS-1 spacecraft were characterized with an innovative thrust vector probe. Offsets from the ion optics axis were less than 2° in the horizontal and vertical axes immediately after ignition and less than 1.5° during steady-state operation. Sensitivity testing revealed that the thrust vector is responsive to the beam current, beam voltage and engine flow rates. Changes in these parameters during throttling after ignition result in rapid changes in the thrust vector orientation which are followed by thermal transients with timescales of hundreds to thousands of seconds. Throttling from one power level to another results in changes of less then 0.5° with very little long-term thermal component. Higher frequency fluctuations of less than 0.05° amplitude and periods of tens to hundreds of seconds are thought to be induced by the laboratory xenon feed system and are not expected to be observed in flight. A longterm drift of less than 0.3° in each axis was observed during the 8000 hour test. Significant differences in behavior were observed between the engineering model and flight thrusters which are likely due to changes in the engine structural and thermal design. 'Copyright 1998 by the American Institute of Aeronautics and Astronautics, Inc. No copyright is asserted in the United States under Title 17, U.S. Code. The U.S. Government has a royalty-free license to exercise all rights under the copyright claimed herein for Governmental purposes. All other rights are reserved by the copyright owner. Introduction Xenon ion propulsion is finally entering an age of application in NASA's planetary program. A xenon ion primary propulsion system is one of the key technologies to be demonstrated on Deep Space 1, the first of the New Millenium missions. This spacecraft will be launched in 1998 and fly by an asteroid and a comet in 1999. Ion propulsion was considered an enabling technology for the DS4/Champollion comet sample return mission, which is slated for launch in 2003. Ion thrusters are also being studied for use in a number of proposed Discovery missions and outer planet missions. One issue that spacecraft designers must consider in the integration of electric propulsion on planetary spacecraft is how the thrust vector of the ion engine varies with time and throttle level. The gimbal design must have sufficient range and rate to null disturbances produced by initial offsets in the thrust vector from the spacecraft center-of-gravity as well as shortand long-term transients. The magnitudes and rates of these offsets and transients must therefore be characterized to support gimbal and control system design efforts. The behavior of the thrust vector has been studied as a part of the NASA Solar Electric Propulsion Technology Application Readiness (NSTAR) program, which is validating NASA's 30 cm xenon ion thruster technology for use in near-Earth and planetary missions. This program is designed to develop the industrial capability to produce flight engine, power processor and propellant feed system hardware and demonstrate that the technology is mature enough for flight applications. The technology validation portion of the program is focussed largely on providing flight program managers with sufficient information on performance, reliability and spacecraft interactions to allow them to use the technology. This paper presents data obtained on the variation of the thrust vector measured in an 8000 hour wear test of an engineering model NSTAR thruster and a functional test of the flight thruster for the New Millenium DS1 spacecraft. Initial offset of the thrust vector from the thruster axis, changes in the thrust vector during engine startup, variations while throttling the engine from one power level to another, and drift over very long periods of operation were characterized. The implications of the measured magnitudes and rates of thrust vector drift on engine gimbal design are discussed. Apparatus and Procedures