Max-Planck-Institut für extraterrestrische Physik
ROSAT Zeit Korrelation, SCC - UTC
|English version||MPE HEG Röntgen-Astronomie Wave ROSAT Kalibration Space Craft Clock|
Space Craft Clock
This document describes the correlation between UTC and the ROSAT onboard spacecraft clock (SCC) for the whole mission since launch. Data are given as pairs (UTC / SCC) together with the result of a linear regression and plots of the residuals (which are typically below 10msec).
How it works
The standard correlation between S/C clock and UTC does not permit the timing of photon-events with an accuracy better than a few hundred milliseconds. Therefore a trick has been implemented in the onboard software:
The central data processing unit ('ZDE') of the focal instrumentation runs in a software cycle which is started every second by a pulse provided by the S/C clock. This cycle can be interrupted by, e.g., the reception of telecommands, health signals from the instrumentation, or by the telecommand pulse 'RESET' which causes the computer to make a warmstart. Upon reception of this interrupt, the ZDE enters an idle loop (after completion of a couple of necessary actions which require exactly 532.8 msec). This idle loop is stopped at the next second pulse coming from the S/C clock. The time between the two interrupts is measured by a software counter, one 'count' corresponds to 0.0113774 msec. Subsequently, the counter value is put into a telemetry frame which contains as header information the actual S/C clock. Whenever the RESET command is sent, the time when the command 'leaves' the antenna on ground is recorded. These informations (CMD-time on ground, ZDE-counter, and S/C-time of telemetry frame) are combined in order to perform the clock correlation. As a further information, the distance between ground station and ROSAT is needed. This is part of a post-facto attitude analysis at GSOC and is delivered offline a few days later. This time-correlation by command is made typically once per week
A comparison of all fits shows that the drift of the clock was continously changing since the begin of the mission.
The S/C clock cannot be set to a specific value but is reset to zero once per year (by purpose, additionally by accident) in order to avoid clock-overflows. Since these resets have been always in January or February, a period between two consecutive clock resets correspond approximately to calendar years but sometimes these 'years' contain more than 365 or 366 days.
For most applications, it is sufficient to perform a linear regression to the correlation data within a period. Special attention has to be given to leap seconds: one of the clocks (UTC) 'jumps' occasionly; therefore one has to subtract a second from SCC in order to get a fit to the data.
Data, plots, and fits
The photon event time (not S/C time!) is delivered to the users in event tables. These photon event times can be transformed to the corresponding UTC by the following formula:
'eventtime' are as given in event tables, 'drift' and 'reset' can be found in the tables below, 'offsets' are added by SASS in order to create homogeneously increasing event-times regardless of any S/C clock resets.
Plots (jpeg) of the residuals can also be found separately for the years 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1998end, or also as PostScipt files 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1998end.
An ASCII-table containing all data can be found in timcor.txt.
Peter Predehl (predehl @ mpe . mpg . de)