All RGS CCDs are operated in the so-called ``frame store'' mode, in which half of each CCD is exposed to the sky. The contents of this half are transferred to the second shielded half, which works as a storage area before readout, while the first half is acquiring the next frame. Thereby, the two-dimensional dispersed photon distribution is stored.
The standard science mode of operation of the RGS instrument is called ``Spectroscopy''. It consists of a two-dimensional readout of one or more CCDs over the full energy range. Each RGS1 CCD is read via two nodes, resulting in an accumulation time of 4.8 sec when reading the eight CCDs. Starting in August 2007, RGS2 CCDs are read via a single node. RGS2 frame times are therefore twice as long as RGS1 frame times (accumulation time is 9.6 sec when reading the eight CCDs).
| Mode | Description |
| Spectroscopy | 2-D readout of up to all CCDs over the full energy range with a long duty cycle (4.8 sec accumulation time for RGS1, 9.6 sec for RGS2, for eight CCDs). Each CCD readout takes 0.6 sec in RGS1 and 1.2 sec in RGS2. Diagnostic images are taken in parallel. |
| Spectroscopy HCR | Similar to Spectroscopy, but no diagnostic images are taken.
For high count rates ( 70 cts s ). |
The on-chip binning (OCB) factor in this mode is 3x3 pixels, after which other operations are performed on board to reduce the data rates to within the RGS telemetry bandwidth. After applying a low signal threshold, hot columns and hot pixels defined by a look-up table are rejected. The remaining pixels are treated by a programmable processor (DPP). Pixels exceeding an upper signal threshold and pixels with complex structures due to cosmic rays are rejected, and only events which fit in a pattern are transferred to the ground including information on their shape.
Each individual chip or any combination of chips can be read out. The energy range covered depends on which CCDs are read out and on the positioning of the source within the field of view. A rough estimate of which energies are sampled if a given CCD is read out is provided in Table 9.
In parallel to the recording of science data, diagnostic images are transferred to the ground. One complete CCD of each RGS is read in this way every 1500s. The on-board data processing is bypassed and the entire CCD image or ``Q-dump'' is transferred to ground. These data are used by the instrument and calibration teams for dark current and system noise level verification with the aim to identify possible instrument degradation with time and/or changes affecting the health of the RGS instruments. These data are now routinely used to improve the calibration of the CCD offsets, e.g. in the case of visual light background.
This mode has been used for almost all observations and is the
recommended configuration. However, for bright sources with an
expected RGS count rate larger than 70 counts s, the
Spectroscopy High Count Rate (HCR) mode, that is the same as the
Spectroscopy mode but without diagnostic frames being taken in
parallel to reduce the telemetry rate, should be used.
Most X-ray sources observed with the RGS do not suffer significantly
from pile-up. However, when source count rates approach or exceed 0.04
photons/cm
/s in RGS1 (or half that value because of the longer
readout time in RGS2) the potential effects of pile-up should be
considered. Since pile-up most often results in the combination of two
first-order events to become a single apparent second order event, a
useful technique to assess the importance pile-up in RGS data is the
comparison of first and second order fluxed spectra. In the absence of
pile-up, the effective area calibration ensures that first and second
order fluxes at a given wavelength agree within a few percent. Hence,
examining the ratio between first and second order can help determine
whether pile-up has played a role.
The effects of pile-up can be mitigated by reducing the accumulation time. Up to now, two relevant methods exist to achieve this, separately or in combination, by either reading fewer CCDs - and inevitably reducing the wavelength coverage - or alternatively reading the most brightly illuminated CCDs more often.
Some basic characteristics of the RGS science modes are listed in Table 11.