TITLE:   GCN CIRCULAR
NUMBER:  6797
SUBJECT: Swift Status and XRT Calibration Uncertainties
DATE:    07/09/18 10:57:59 GMT
FROM:    David Burrows at PSU/Swift  <dnburrows@gmail.com>

D. N. Burrows (PSU), J. P. Osborne (U. Leicester), G. Chincarini 
(OAB), and N. Gehrels (NASA/GSFC) report on behalf of the Swift XRT team:

The Swift observatory continues to progress toward full operations. 
Currently BAT and XRT are turned on and operating.  BAT is detecting 
GRBs and distributing Notices in real time (Barthelmy et al. GCN 
6781).  UVOT remains off and autonomous GRB slewing remains disabled 
until completion of the spacecraft gyro calibrations, which are still 
on-going.  XRT is performing some delayed observations of GRBs and 
posting results by manually generated Circulars (such as GCN Circular 
6796).  XRT Notices are not yet enabled for automatic distribution.

We would like to issue the following general caveats for performance 
of the XRT following the observatory safe-hold recovery operations:

1) Position determination: The safe-hold event resulted in all 3 
Swift instruments being turned off and the observatory cooling down 
to survival temperatures.  The observatory is now operating at its 
normal operating temperature once again, but there is a possibility 
that the thermal cycle caused by the safe-hold may have resulted in a 
shift to the XRT boresight.  There is also a known boresight shift 
introduced into the ACS software during the recovery operations as 
part of the change to a redundant gryo.  Initial indications suggest 
that the XRT boresight shift is small, but until XRT boresight 
recalibrations are complete, we cannot guarantee the same level of 
accuracy in XRT positions obtained prior to the 
safehold.  Recalibration of the XRT boresight is our top priority and 
we will provide updates on the results as they become available.  In 
the meantime, we are increasing XRT position uncertainties (to 6-8 
arcseconds) to account for the uncertainty introduced by this event.

2) Spectroscopy: the XRT team has been working for many months 
towards a change in the CCD substrate voltage in order to reduce dark 
current and minimize the effects of the elevated operating 
temperatures of the XRT CCD.  This voltage change has now been 
implemented and we are recalibrating the instrument.  Until the 
recalibration is completed, there may be small gain changes (of order 
5%) and QE changes (of order 10% for E>6 keV) that are not correctly 
accounted for by the current CALDB files.  We will update the CALDB 
files as soon as possible.  Meanwhile, we urge caution in the 
interpretation of spectral data, and we urge all users to process 
data through xrtpipeline themselves using the latest available CALDB 
files until the changes have worked their way into the standard 
production pipelines at the Swift Data Center.  In particular, we 
note that small gain errors can introduce spurious spectral features 
associated with the Oxygen K edge, the Si K edge, and the Gold M edge 
in the instrument response.  These can be handled during spectral 
fits by allowing a gain adjustment in XSPEC to minimize these features.