[vsnet-grb-info 5253] Swift autonomous slewing re-enabled

[GCN Circulars]

TITLE:   GCN CIRCULAR
NUMBER:  6946
SUBJECT: Swift autonomous slewing re-enabled
DATE:    07/10/19 21:31:28 GMT
FROM:    Neil Gehrels at GSFC  <gehrels at milkyway.gsfc.nasa.gov>

Neil Gehrels (NASA-GSFC) reports on behalf of the Swift team:

This is an update on Swift status following earlier Circulars
(Gehrels 6760; Barthelmy et al. 6781; Burrows et al. 6791;
Gehrels 6825).  As of today, Swift is re-enabled for full-time
autonomous GRB slewing.  The gyro problems of the past 2 months
are resolved and the observatory is back to slewing accurately
within its 3 arcmin requirement.  BAT and XRT continue to be
fully operational and performing science, as they have been for
several weeks.  UVOT is in engineering mode, and will be activated
for science observations in the middle of next week.  The TOO web
site is back on line.

More details on the problem and solution are as follows:  On
August 10 gyro #3 (known as Two Axis Rate Assembly - TARA 3)
exhibited jumps in its bias offset level (voltage output when gyro
is not rotating) that caused the spacecraft to go into safe hold.
Redundant TARA 2 was switched in for recovery.  It was then found
that TARA 2 is slightly misaligned from TARA 1, with the consequence
that slew accuracy was degraded.  Due to memory limitations of the
onboard spacecraft computer, assumptions were made in the matrix
manipulation mathematics that did not permit a simple correction
for the misalignment.  However, a clever technique was found by
Craig Markwardt to use available parameters to combine signals
from all 3 gyros for accurate slewing.  This mode of operation,
which is now the baseline, is called TARA 123 and uses TARA 3 to
correct the TARA 1 and 2 alignments.  The TARA 3 term comes in at
low weight and, as such, is not sensitive to its small bias shifts.
A more elaborate software patch has been developed that could be used
in the event of a failure of TARA 3.