The list of successful proposals generated by the review and merger process will be fed into the SPIKE scheduler to produce two mission time lines: a ``long-term'' time line, covering one year of observations, and a ``short-term'' time line, covering a week of observation and updated every week. SPIKE automatically takes into account the various constraints on ASCA observations and chooses the most efficient overall time line. These constraints are described below for those observers who are interested in how ASCA samples targets. Short-term and long-term timelines can be found on the ASCA GOF homepage, the URL of which is:
Please look under Timelines .
The constraints on what and when ASCA can observe fall into three broad categories:
The mirror axis is parallel to the plane of the solar paddles (see the drawing of the spacecraft in Figure 4.3) which means that there is a hard limit on the Sun angle (Sun-Earth-target angle) for any ASCA observations of -, for operational reasons. For observations that are not time-constrained, a range of - is used for scheduling purposes, to increase the margin for safety, and to ensure the quality of the SIS data. Larger Sun angles tend to warm up the CCDs and smaller Sun angles increase the scattered optical light, both of which lead to degradation in data quality. Observations outside this range of - should be requested only when absolutely necessary for time-critical observations. Depending on other circumstances, observations outside the range - may be refused by the ASCA operations team at ISAS.
To work out when ASCA can observe a source, proposers should use the program viewing which is described in the appendix to this document.
ASCA orbits at an altitude that varies between 500 and 600 km, with an inclination of 31.5 degrees. The orbital period is 95 minutes. This orbit imposes the following sampling constraints.
Satellites launched into low-Earth equatorial orbits inevitably pass through the South Atlantic Anomaly (SAA). Because of orbit precession, SAA passages occur on about nine of the fifteen orbits per day. During a passage, the high particle flux renders the instruments unusable (the high voltage of the GIS, in fact, is turned off), but even after emerging from the SAA, the level of induced radiation, seen as background, may be very high. The practical effect is to blackout 15-30 minutes per orbit for about nine orbits per day.
Sources not at the orbital poles will suffer Earth occultation: data taken near the limb of the Earth (elevation in the range 0-) suffer absorption in the Earth's atmosphere, and hence are discarded together with negative elevation data (i.e., occultation by the solid Earth). We will refer to both as Earth occultation.
Typically, Earth occultations last about 30 minutes and occur every satellite orbital period (95 minutes). Sources near declination enter the continuous viewing zone (CVZ) of ASCA, every 56.5 days, for as long as 10 days at a time. However, the actual coverage of such sources will not be continuous due to SAA and other constraints described below. In particular, targets in the CVZ are never far away from the Earth's horizon, and SIS data may suffer badly due to the proximity to the bright Earth limb.
In the six orbits per day which do not pass through the SAA, there are still regions of high particle background where the geomagnetic cut-off rigidity (COR) is low. Both SIS and GIS will continue to collect data regardless of COR, although in some cases the increased background may lead to telemetry saturation. Typically, data taken in lowest COR regions are discarded in analysis; the threshold should be chosen according to the source intensity and the type of science. A limit of 4 GeV/c is suitable for bright sources and hardly excludes any data; a limit of 6 GeV/c is more conservative, and typically leads to a loss of effective exposure (of non-SAA, non Earth occult time) of about 10 %.
The quality of SIS data taken near the sunlit limb of the Earth suffers in several ways, due to the sensitivity of the CCDs to scattered (optical) sunlight, as well as to scattered Solar X-rays (including geocoronal oxygen emission line). The symptoms are: (1) Increased rate of flickering pixels (see section 7.6.5), potentially leading to telemetry saturation. Regions of the worst affected chips are masked off; (2) Offsets in channel vs. energy scale, caused by the flood of optical photons that raises the zero-level; and (3) Increased uncertainties in the soft X-ray background, particularly at the O line region.
Local background subtraction may take care of point 3 above for point sources, but it is serious for extended sources. Moreover, sensitivity to optical light is dependent on CCD clocking mode (see section 7.5.1): typical threshold for the Bright Earth angle is - in 4-CCD mode, typically resulting in 10% reduction in effective exposure time; 10-15 in 1-CCD mode; - in 2-CCD mode. However, sources in or near the CVZ (see 4.4.2 above) are viewed close to the horizon and may suffer much greater losses of effective SIS exposure time. This can be avoided by deliberately scheduling the observations away from CVZ, if the schedulers are warned of this possibility.
Figure 4.4 shows a typical sampling window that observers can expect (a PV observation of the early type star Pup). The observing efficiency of ASCA is pretty much as expected: a day's observation will yield 35-65 ks of good data, the amount depending most on the extent of Earth occultation.
Direct contact between the satellite and the ground stations will be possible for (typically) nine to twelve of the fifteen orbits per day: five from Japan and the remainder from the Deep Space Network (DSN; nominally, nine passes per day, distributed unequally over orbits). During a contact, data are telemetered down to the ground station and commands transmitted up--however, only the Japanese contacts will be used for this latter purpose. These five orbits are known as contact orbits, while the others are known as remote orbits. One day of the week is ``holiday'' and has no contact orbits. ASCA cannot record or transmit data taken during DSN downlinks, therefore each DSN downlink results in an interruption of data for about 8.5 min.
Because of hot and flickering SIS pixels, low bit rate cannot be used in practice. Fortunately, it has proved possible to observe for about two thirds of the time in high bit rate (32,768 bps), the rest in medium bit rate (4,096 bps). With nine data dumps a day, most of the time the satellite will be high bit rate.
All four instruments must operate at the same bit rate.
The data recorder on the satellite can store (32,768 s) of data in medium bit rate, and (4,096 s) in high bit rate. The choice of bit rate depends largely on the interval between contacts.
How the bit rate constrains an observation is related to feasibility and is discussed in Chapters 8 & 9.
The ASCA spacecraft can rotate about all three of its axes at a maximum rate of 0.15 degrees per second. In the PV phase, the Attitude Control System has been performing as follows:
Pointing stability: Z-axis, 3-sigma 5 arcsec Around Z-axis, 3-sigma 20 arcsec Attitude determination: By on-board CPU , after maneuver < 18 arcmin By on-board CPU , in pointing 11 arcsec On the ground, after manoeuvre < 18 arcmin On the ground, in pointing 4 arcsec Post-manoeuvre settling time: 15-20 minutes Movement, per orbit, of image due to thermal expansion: 20 arcsec
The factors behind the choice of mode for the GIS remain as before. For the SIS, however, the presence of hot and flickering pixels, which wastefully occupy telemetry, has raised the count rates at which the telemetry saturates. Please refer to Chapter 9 for information about choosing SIS modes.
The placement of the target in the field of view depends on mode and on the type of source observed. This is illustrated by Figure 4.5 which shows the currently adopted default pointing positions with respect to the layout of the SIS chips. If a single point source is to be observed, the mission operations staff are using three standard positions, depending on how many CCDs are active. These take into account inter-chip gaps, the support grid on GIS windows, and the misalignment of the four detectors.
For extended or multiple sources, the operations team will do their best not to hide something interesting between the chips or under the GIS window grid. The figure also shows that about one third of chips 3 & 2 of SIS0 suffer from an optical/UV light leakage (see Chapter 7). Note that, due to the square shape of the CCD chips, the sky coverage achieved is a function of the spacecraft roll angle. On a given date, there is a freedom in roll of typically 15 degrees, and observations of extended sources are usually done in ``best available roll angle'' bases. That is to say, after the observation is scheduled, PI and the GOF request a specific chip combination and a small adjustment in roll to best achieve the science. This usually involves some (but an acceptable level of) compromise in sky coverage and/or chip selection. If a specific orientation is essential for the science, and achievable (not all roll angles are possible for a given point on the sky), such an observation should be requested as time critical.
The operational load of commanding the satellite, together with the performance of the attitude control system, impose restrictions on the minimum amount of observing time. These are described below.
The ASCA project uses the GIS exposure time, with a liberal COR threshold of 4 GeV/c, as the basis for observation scheduling and as the completion criterion. In scheduling, margins of 5-10% of the approved time and 3-5 ksec (1 ASCA orbit, to account for acquisition of guide stars and subsequent fine-tuning of attitude) are added. This practice results in a clear majority of observations achieving the approved exposure. Note, however, that the effective GIS exposure will be smaller than this by 10% if a more conservative COR cut is applied, and that the effective SIS exposure time is typical smaller yet by another 10%. The observations are deemed completed if 95% of approved time is achieved for category A targets, 70% for B targets; no such criterion exists for C targets after AO-3. Observers can request supplementary observations if these completion criteria are not met.
Previously, proposers had to request not less than a consecutive 20 ks for a single observation (typically half a day of satellite time). Exceptions were made, however, for sets of multiple, close pointings at large extended objects (see below). Since AO-5 this policy has been relaxed somewhat in the sense that monitoring proposals of extended and point sources may be accepted in which each snapshot is not less than 10 Ks (but the minimum of 20 Ks or two 10 Ks snapshots still applies). Acceptance of such proposals will, however, be subject to the discretion of mission operations. Note that within the above restrictions, the requested time does not have to be a multiple of 10 Ks.
In general, because the attitude has to be verified on the ground after each contact, pointing the satellite at many different positions during one day risks the accumulation of large--possibly catastrophic--drifts in attitude. However, if the positions are close, i.e., the maneuvers are small, the risk is much lower. It is possible, therefore, to request a maximum of nine close pointings per day (one for each contact), each yielding 4-5 ks of data. Strictly, ``close'' is interpreted to mean a step-size of one GIS field of view (50 arcmin). Larger steps are possible, but may be vetoed by the mission operations staff. For advice on this issue, please contact the ASCA GOF by email at email@example.com. Note that this facility is for extended objects, not for groups of point sources.
The lowering of overall efficiency due to the inclusion of time-critical observations has turned out to be less than anticipated. The ASCA mission observations staff are willing to perform up to one time-critical observation per week.
Figure 4.3: Drawing of the ASCA spacecraft showing the four XRT, the Sun Shield and the Solar Paddles.
Figure 4.4: Figure 4.4: A typical sampling window of ASCA, this example for the PV phase observation of Pup. Top four panels contain plots of four important factors discussed in this section, with a typical threshold (two alternatives in the case of COR). The bottom panel is the actual light curve obtained with GIS-2, which was filtered with the constraints of non-SAA, elevation , and COR > 6 GeV/cc.
Figure 4.5: The layout of the four chips that make up each SIS camera, shown with the default pointing positions for the various clocking modes (1-CCD, 2-CCD, etc.). Also shown are the areas afflicted by UV/optical light leak. This diagram is shown in detector coordinates used for the X-ray images.