After having been in development for many years at the Glenn Research Cente
r (formerly the Lewis Research Center), the national aeronautics and space
administration-designed, 30 cm, ring-cusp, xenon ion engine was launched on
the Deep Space 1 (DS1) spacecraft on 24 Oct. 1998 from the Kennedy Space C
enter in Florida. It has since accumulated 2200 h of in-space thrusting at
input power levels ranging from 0.52 to 1.96 kW, has successfully enabled t
he spacecraft to fly by the asteroid Braille in July 1999, and is now thrus
ting DS1 along a trajectory towards its comet destinations in 2001. The des
ign, assembly, test, integration, and operation of this thruster comprise a
unique path of technical determination, artful design choices, persistent
engineering and analysis, and mastery of vacuum chamber operations. The tes
ting program over the development years, the assembly and integration perio
ds, and the flight operational period thus far have shown that the project
test philosophy of segregating effects against unique causes proved itself
most useful. The 8000 h life test, the culmination of the pre-launch ground
test plan, not only met its goals but surpassed them with margin. This art
icle will explain the thruster test program from beginning to end, illustra
ting the technical and programmatic decision making along the way. It will
justify the use of engineering models as an inexpensive method of determini
ng answers to key design questions and will explain why testing of the thru
ster alone only solved a portion of the system operations task. The highlig
ht of the test program proved to be the vacuum firing of the ion engine dur
ing the spacecraft's solar thermal vacuum test. A comparison of the preflig
ht data with postflight data shows that high confidence was warranted for e
xecuting a successful flight to the asteroid and beyond. (C) 2000 American
Institute of Physics. [S0034-6748(00)51402-6].