Temperature Dependence of the Ginga Clock Rate

John E. Deeter¹ and Hajime Inoue²

[1]Department of Astronomy, University of Washington
[2]Institute of Space and Astronautical Science

Abstract:

The period of the clock aboard Ginga varies by one part in 10⁶, mainly due to a strange temperature dependence with a maximum near 18°C. This temperature dependence makes it difficult to assign times of observation to the data with millisecond accuracy, particularly during remote orbits during which it is not possible to compare the onboard clock with a reference clock at the Kagoshima ground station. In this paper we investigate the relationship between satellite temperature and clock period, and show how this relationship can be used to estimate the clock period from the temperature data even when direct clock comparisons are not available. The resulting corrected times of observations are accurate to 2-3 milliseconds, even during remote orbits. We show that with some simple precautions this method can be used to achieve 100 µs accuracy in time assignment for Astro D, without actively controlling the clock temperature with an oven.

• Introduction
• Analysis and Results
  • Determination of Ginga Clock Rate
  • Determination of Clock Temperature
  • Clock Period-Temperature Correlation
  • Recovery of Clock Phase from Temperature Data
• Discussion
• Recommendations for Astro D
• References

• A 34kB gzipped PostScript reprint
• A 762kB gzipped tar copy of the figures