Telescope PerformanceScientific Instrument OverviewOrbital ConstraintsGuide Stars and Target AcquisitionSystem OverviewAs shown in Fig. 1, HST's Scientific Instruments (SIs) are mounted in bays behind the primary mirror. The Wide Field Planetary Camera 2 occupies one of the radial bays, with an attached 45 degree pickoff mirror that allows it to receive the on-axis beam. Three SIs (Faint Object Camera, Near Infrared Camera and Multi-Object Spectrometer, and Space Telescope Imaging Spectrograph) are mounted in the axial bays and receive images several arcminutes off-axis.
Figure 1: The Hubble Space Telescope
Major components are labelled, and definitions of V1,V2,V3 spacecraft axes are indicated.
During the servicing mission in December 1993, the astronauts installed the Corrective Optics Space Telescope Axial Replacement (COSTAR) in the fourth axial bay (in place of the High Speed Photometer). COSTAR deployed corrective reflecting optics in the optical paths in front of the Faint Object Camera, thus removing the effects of the primary mirror's spherical aberration. In addition the Wide Field and Planetary Camera (WF/PC) was replaced by the WFPC2, which contains internal optics to correct the spherical aberration.
The Fine Guidance Sensors (FGSs) occupy the other three radial bays and receive light 10-14 arcminutes off-axis. Since at most two FGSs are required to guide the telescope, it is possible to conduct astrometric observations with a third FGS. Their performance is unaffected by the installation of COSTAR.
For an overview of the SIs, see Scientific Instrument Overview. Detailed information about each SI is contained in separate Instrument Handbooks.
HST receives electrical power from two solar arrays (see Fig. 1), which are turned (and the spacecraft rolled about its optical axis) so that the panels face the incident sunlight. During the 1993 servicing mission the astronauts installed new solar arrays, which have significantly reduced the thermally induced vibrations that the old arrays had been producing. Nickel-hydrogen batteries provide power during orbital night. The two high-gain antennas shown in Fig. 1 provide communications with the ground (via the Tracking and Data Relay Satellite System). Power, control, and communications functions are carried out by the Support Systems Module (SSM), which encircles the primary mirror.
In addition to STIS and NICMOS, the second servicing mission will replace several additional pieces of equipment. One of the FGSs (most likely FGS 2) will be replaced. The new FGS will have an adjustable fold flat to recover some of the performance capability lost by spherical aberration. A spare magnetic tape recorder will replace the failed ESTR-2. A solid state recorder (SSR) will replace ESTR-1, and will provide for a factor of 10 greater on-board data storage volume. This extra storage will be necessary to support parallel operations of the WFPC2, STIS, and NICMOS. It will also provide increased flexibility in scheduling HST observations, reducing the tight coupling with the TDRSS system.
To discuss HST pointing, it is useful to define a coordinate system that is fixed to the telescope. This system consists of three orthogonal axes: V1, V2, and V3. V1 lies along the optical axis, V2 is parallel to the solar-array rotation axis, and V3 is perpendicular to the solar-array axis (see Fig. 1). Power and thermal constraints are satisfied when the telescope is oriented such that the Sun is in the half-plane defined by the +V1 axis and the positive V3 axis. The orientation that optimizes the solar-array positioning with respect to the Sun is called the "nominal orientation."
It should be noted that the nominal orientation angle required for a particular observation depends on the location of the target and the date of the observation. Observations of the same target made at different times will, in general, be made at different orientations.
Some departures from nominal orientation are permitted during HST observing (e.g., if a specific orientation is required at a specific date, or if the same orientation is required for observations made at different times). Roll is defined as the angle about the V1 axis between a given orientation and nominal orientation. Off-nominal rolls are restricted to approximately 5 degrees when the sun angle is between 50 degrees and 90 degrees, < 30 degrees when the sun angle is between 90 degrees and 178 degrees and is unlimited at anti-sun pointings of 178 degrees to 180 degrees.
HST utilizes electrically driven reaction wheels to perform all maneuvering required for guide-star acquisition and pointing control. A separate set of rate gyroscopes is used to provide attitude information to the pointing control system. The servicing mission restored or replaced three gyros that had failed since the original launch, so that the spacecraft currently has a total of six operational gyros. Any three of these are the minimum required for telescope pointing control.
The slew rate is limited to approximately 6 degrees per minute of time. Thus, about one hour is needed to go full circle in pitch, yaw, or roll. Upon arrival at a new target, up to 9 additional minutes must be allowed for the FGSs to acquire a new pair of guide stars. As a result, large maneuvers are costly in time and are generally scheduled for periods of Earth occultation or crossing of the South Atlantic Anomaly.
The telescope does not generally observe targets within 50 degrees of the Sun, 15.5 degrees of any illuminated portion of the Earth, 7.6 degrees of the dark limb of the Earth, nor 9 degrees of the Moon. Following the first servicing mission, the telescope is again allowed to point directly away from the Sun.
There are exceptions to these rules for HST pointing in certain cases. For instance, the bright Earth is a useful flat-field calibration source. However, there are onboard safety features that cannot be overridden. The most important of these is that the aperture door shown in Fig. 1 will close automatically whenever HST is pointed within 20 degrees of the Sun, in order to prevent direct sunlight from reaching the optics and focal plane.
Objects in the inner solar system, such as Venus or comets near perihelion, are unfortunately difficult or impossible to observe with HST, because of the 50 degree solar limit. When the scientific justification is compelling, observations of Venus and time-critical observations of other solar-system objects lying between 45 degrees and 50 degrees of the Sun may be carried out (this capability was successfully demonstrated in Cycle 4).
Commands to HST originate at the STOCC and are sent via land-line or communications satellites to the TDRSS ground station at White Sands. From there the commands are sent via the appropriate TDRSS to HST. Scientific data are sent from HST to the STOCC via the reverse path, and then from the STOCC to the STScI via dedicated high-speed links.
The TDRSS network supports many spacecraft in addition to HST. Therefore, use of the network, either to send commands or return data, must be scheduled. Because of limited TDRSS availability, command sequences for HST observations are normally uplinked periodically and stored in the onboard computers. HST then executes the observations automatically.
It is possible for observers at STScI to interact in real-time with HST for specific purposes, such as certain target acquisitions. In practice, real-time interactions are difficult to schedule. Historically, during normal operations, fewer than 50 real-time interactions have been required per year.
HST currently uses two onboard tape recorders to store scientific data. After the second servicing mission a much larger capacity Solid State Recorder will be available. Except when real-time access is required, most HST observations are stored to a recorder and read back to the ground several hours later. There are, however, limits to the amount of data that can be handled by the ground system supporting HST. Some scientific programs requiring very high data-acquisition rates cannot be accommodated, because the SIs would generate more data than either the links or ground system could handle.
Telescope PerformanceTable 2: HST Optical Characteristics and Performance
[1] Expected performance.
Software has been developed at STScI for detailed simulations of HST images, which agree well with actually observed (pre- and post-repair) images. This Telescope Image Modelling (TIM) software, with its point-spread-function (PSF) library, is available for downloading from STEIS. Another set of software specifically designed to simulate HST camera images, called TINYTIM, is also available.
A figure showing actual measurements of the PSF and encircled energy achieved with the Faint Object Camera can be found in the FOC Instrument Handbook.
Each of the three FGSs covers a 90 degree sector in the outer portion of the HST field of view (FOV), as shown in Fig. 2. Optics within the FGS, using precision motor-encoder combinations, select a 5" x 5" region of sky into an x, y interferometer system. Once an FGS is locked onto a star, the motor-encoders are driven to track the interference fringe of the guide star. The encoder positions are used by the PCS software to update the current telescope attitude and correct the pointing.
The FGSs have two guiding modes: Fine Lock and Coarse Track. Fine Lock was designed to keep telescope jitter below 0.007" rms. Previously, there were periods of 2 to 5 minutes each orbit when there was increased jitter (0.020-0.050") due to thermal effects in the Solar Arrays. The combination of the new Solar Arrays and tuning of the pointing control system has successfully eliminated these effects. The telescope jitter is now routinely below the 0.007" rms level. A drift of up to 0.05" may occur over a timescale of 12 hours and is attributed to thermal effects as the spacecraft and FGSs are heated or cooled. Observers planning extended observations in 3 0.1" STIS slits should execute a target peak up maneuver every 4 orbits.
Coarse Track is now believed to cause degradation in mechanical bearings in the FGSs, and accordingly is no longer available as a guiding mode.
Guide-star acquisition times are typically 9 minutes. Reacquisitions following interruptions due to Earth occultations take about 6 minutes. It is also possible to take observations (primarily WFPC2 "snapshot" exposures) without guide stars, using only gyro pointing control. The absolute pointing accuracy using gyros is about 14" (one sigma), and the pointing drifts at a rate of 1.4 +/- 0.7 mas s**-1.
The main factors that limit the observing efficiency are (1) the low spacecraft orbit, with attendant frequent Earth occultation of most targets; (2) interruptions by passages through the South Atlantic Anomaly; (3) the relatively slow slew rate; (4) telemetry constraints; and (5) the performance of the scheduling algorithm.
Scientific Instrument Overview
Tables 3(a)-(e) provide a summary of the capabilities of the SIs. For some applications, more than one instrument can accomplish a given task, but not necessarily with equal quality or speed.
The following subsections contain brief descriptions of the five SIs. After examining Tables 3(a)-(e), prospective proposers should read these descriptions in order to determine which SIs are likely to be most suitable for their programs. Revised and updated Instrument Handbooks, which discuss the SIs in detail, have been distributed to institutional libraries, and are available from STScI. The Instrument Handbooks must be consulted before actual preparation of observing proposals. In addition, exposure simulators for the SIs are available to assist in the estimation of exposure times, as described in the Synphot User's Guide; users may contact the STScI Help Desk for more information.
Data from the following four SIs, which were removed from HST during the December 1993 servicing mission (WF/PC, HSP), or are planned for removal in Feburary 1997, are now available only in archival form:
Table 3: HST Instrument Capabilities
Table 3: HST Instrument Capabilities (continued)
Notes to Tables 3(a)-(e):
The WFPC2, which is HST's only on-axis instrument, is designed to provide digital imaging over a wide field of view (FOV). It has three "wide-field" charge-coupled devices (CCDs), and one high-resolution (or "planetary") CCD. Each CCD covers 800 x 800 pixels and is sensitive from 1200 to 11,000 Å. All four CCDs are exposed simultaneously, with the target of interest being placed as desired within the FOV.
The three Wide Field Camera (WFC) CCDs are arranged in an "L"-shaped FOV whose long side projects to 2.5', with a projected pixel size of 0.10". The Planetary Camera (PC) CCD has a FOV of 35" x 35", and a projected pixel size of 0.0455". A variety of filters may be inserted into the optical path. Polarimetry may be performed by placing a polarizer into the beam. A ramp filter exists that effectively allows one to image a small 3 10"object in an arbitrary 1-3% bandpass at any wavelength between 3700 and 9800 Å, by appropriately positioning the target within the FOV.
The WFC configuration provides the largest FOV available on HST, but undersamples the cores of stellar images; the PC configuration samples the images better, over its smaller FOV.
The f/96 camera (plus COSTAR) has a FOV of 7" x 7" and a pixel size of 0.014" x 0.014" in its standard 512 x 512 format; a field of 14" x 14" can be used with the 512 x 1024 pixel format with a (rectangular) pixel size of 0.028" x 0.014". This camera provides two occulting fingers (0.3" and 0.5" wide) at the entrance aperture. The f/96 camera also has three polarization analyzers for polarimetric imaging.
Usage of the f/48 camera is currently restricted to LONG SLIT spectroscopy only; see the FOC Instrument Handbook for further details.
The FOC detector is a three-stage image intensifier, optically coupled to a television tube. Software centroids the individual photons. A variety of options is available for the size and shape of the area that is scanned and the spatial resolution. The dynamic range and limiting magnitude of the FOC depend on the readout format and the desired signal-to-noise ratio, but the limiting magnitude in a broad bandpass is about mU = 27.
In comparing the FOC and WFPC2, proposers should note the following: (1) the FOC provides the higher angular resolution at all wavelengths, and the WFPC2 provides the larger field of view; (2) the FOC is faster below about 4500 (Å), while the WFPC2 is faster above 4500 (Å).
Each camera carries 19 independent optical elements providing a wide range of filter options. Cameras 1 and 2 have polarimetric filters, Camera 2 has a 0.3 arcsec radius coronographic spot and optimized cold mask, Camera 3 has three separate grisms providing slitless spectroscopy over the full NICMOS wavelength range. A variety of standard dithering and chopping (for background and sky mapping) sequences are available.
The three cameras are designed to be operated independently and simultaneously. Accordingly, proposers are strongly encouraged to request use of all three cameras in their Phase I submission. In cases where the user has no interest in obtaining multi-camera observations then the user will be expected to specify survey observations through standard filters at the Phase II submission. Details of our proposed survey programs are in the NICMOS Instrument Handbook.
The three 1024 x 1024 pixel detectors supporting spectroscopy, target acquisitions, and limited imaging applications are:
For positional measurements, the useful magnitude range is mv = 4 to 17 mag, and the precision is about +2 mas. Generally, a target star and 5 to 10 reference stars within the FOV annulus would be observed at several epochs to yield relative proper motions and parallaxes. Note that the effective FOV for parallax observations is reduced to about 4" x 5", since for observations six months apart the telescope's roll angle will differ by 180 degrees (see HST Maneuvering and Pointing).
The "TRANS" mode for double-star and angular-diameter measurements is available, and considerable data-reduction and analysis software has been developed at STScI. This mode is nominally capable of measuring (1) separations down to 5 mas and magnitude differences of up to 4 mag, for double stars whose primaries are as faint as about mv = 14 mag, and (2) angular diameters above 20 mas. Similar measurements on non-stellar objects are also possible.
A new FGS is scheduled to be installed during SM97. Its utility as an astrometric device will be evaluated during the post-servicing mission observatory verification period. However, proposals should be based on FGS-3, currently being used for astrometry, which will remain on board the spacecraft.
In order to avoid confusion with the spacecraft's V2 and V3 axes, we define two new axes in Fig. 2, U2 and U3, which are fixed in the focal plane as projected onto the sky. At nominal roll (see HST Maneuvering and Pointing), the U3 axis points toward the anti-Sun.
Table 4 lists the relative effective locations of the SI apertures; the U2,U3 coordinate system of Fig. 2 is used, and the linear dimensions have been converted to seconds of arc using a plate scale of 3.58 arcsec mm**-1. The locations of the WFPC2, FOC and FGS apertures are accurate to about +/-1", the predicted locations of the STIS and NICMOS apertures could be in error by as much as 10".
Table 4: Nominal Effective Relative Aperture Locations
[1] Predicted locations for STIS and NICMOS.
[2] FGS-2 to be replaced during SM97, location will remain unchanged.
Figure 2: Effective aperture locations for the instruments available after SM97 (see also Table 4).
In order to avoid exceeding the MAMA bright object limits, observations must never:
Orbital ConstraintsTable 5: HST Nominal Orbital Parameters (Epoch 1996.3)
The length of target occultation decreases with angle from the spacecraft orbital plane. Targets lying within 24 degrees of the orbital poles are not geometrically occulted at all during the HST orbit. However, the size of the resulting "Continuous Viewing Zones" (CVZs) is substantially reduced by the Earth-limb avoidance angles. Note also that scattered Earth light may be significant when HST observes near the bright Earth limb.
Since the orbital poles lie 285.5 from the celestial poles, any target located in the two declination zones near +/- 615.5 will be in the CVZ at some time during the 56-day HST precessional cycle. The maximum uninterrupted length of an observation may then be up to 7 days, although passages through the SAA (see below) will force gaps in coverage after a maximum of 8 orbits.
A detailed examination of all the observing constraints has shown that there are two distinct CVZ regions. The "heart" of the CVZ are those positions where substantial scheduling opportunities exist, and these observations will be treated as any other observation. The "wings" of the CVZ are those positions where limited scheduling opportunities exist, and so it is possible for those observations to come into conflict with other observations or contingencies. If a proposer elects to request CVZ time in the "wings", and the observations are not obtained in the limited CVZ opportunities, then those observations may NOT BE EXECUTED.
This positional uncertainty affects observers of time-critical phenomena, since the target could be behind the Earth at the time of the event. In the worst case, it will not be known if a given event will be observable until a few days before the event.
Guide Stars and Target Acquisition
The Guide Star Catalog (GSC), which resulted from the digitization and analysis of the plate collection, contains information, including coordinates and magnitudes, on about 18 million objects to 14.5 mag.
The observation summary that is part of each Phase I observing proposal must include celestial coordinates for all fixed targets, but these coordinates need only be of sufficient accuracy for the scientific and technical reviews.
More accurate target coordinates are required in Phase II. Stellar proper motions (and parallaxes) will also be requested during Phase II, since even relatively small stellar motions can be surprisingly significant. Phase II proposers should be prepared to supply such information themselves.
Target Acquisition without the Ground System
No acquisition means that pointing control will be entirely on gyros, and FGSs will not be used. The telescope is slewed on gyro control from the last target. The pointing accuracy depends on the size of the slew from the previous target. The best-case uncertainty is 0.010" for each 30" displacement in the slew to the target.
Blind acquisition means that guide stars are acquired and the FGSs are used for pointing control. The pointing is accurate to the guide star position uncertainty, which is about 1".
Onboard acquisition means that software onboard the spacecraft specific to the science instrument in use will be used to center the fiducial point onto the target. On-board target acquisitions will be needed for all STIS spectroscopic observations, and for NICMOS coronographic observations. The WFPC2 and FOC do not have onboard acquisition capabilities. For specific information on methods and expected pointing accuracies, see the Instrument Handbooks for the instrument to be used.
Early acquisition means using an image taken on an earlier visit to provide improved target coordinates for use with subsequent visits. Usually the WFPC is used to make the acquisition image several months in advance of the science observations. The observer will analyze the early acquisition image and provide the improved coordinates to the scheduling system.
Target Acquisition with the Ground System (OPUS)
Target acquisitions that cannot be accomplished reliably or efficiently via one of the above methods may still be possible by transmitting relevant data to the STScI, analyzing it to determine the needed pointing corrections, and then providing those corrections to the telescope. This description covers two kinds of activities, the "interactive acquisition" and the "reuse target offset", both of which are described briefly here. The observer who uses these capabilities will be supported by staff of the OPUS (the Observation Support/Post-Observation Data Processing (OSS/PODPS) Unified System) of the Data Systems Division.
Interactive acquisition, or real-time target acquisition, uses the ground system software to calculate the small angle maneuver to move the aperture onto the target. This method is available for all science instruments except the astrometry FGS. High data rate TDRSS links are required at the time the data is read out of the instrument to transmit the data to the ground, and at a subsequent time to re-point the telescope before the science observations, which adds a difficult constraint to the scheduling. The GO, or a designated representative, must be present at the STScI at the time of the acquisition. The acquisition data, usually an image, is analyzed by OPUS personnel to compute the image coordinates and centering slew for the target identified by the GO.
Reuse target offset means using an offset slew derived from an onboard acquisition done on visit 1 to reduce the amount of onboard acquisition time required for subsequent visits to the same target. The data from the initial visit are analyzed by OPUS personnel to provide the offset slew to be repeated for subsequent visits. All subsequent visits to the target must use the same guide stars as the initial visit, which limits the time span of all visits to a few weeks. There are additional instrument-specific requirements. The GO is advised to contact the STScI Help Desk if this capability is required.
Two specific aspects of solar-system observations are discussed below: the initial acquisition of a moving target, and the subsequent tracking of the target during the scientific observations. Only an overview of the current moving-target capabilities is given here. Phase I proposers are encouraged to consult the STScI Help Desk for more detailed information.
Tracking Capabilities
HST is capable of tracking moving targets with the same precision as for fixed targets (see Guiding Performance). This is accomplished by maintaining FGS Fine Lock on guide stars, and driving the FGS star sensors in the appropriate path, thus moving HST so as to track the target. Tracking under FGS control is technically possible for apparent target motions up to 5 arcsec s**-1. In practice, however, this technique becomes infeasible for targets moving more than a few tenths of an arcsec s**-1. It is currently possible to begin observations under FGS control and then switch over to gyros when the guide stars have moved out of the FGS field of view. If sufficient guide stars are available, it is possible to "hand off" from one pair to another, but this will typically incur an additional pointing error of about 0.3".
Targets moving too fast for FGS control, but slower than 7.8 arcsec s**-1, can be observed under gyro control, with a loss in precision that depends on the length of the observation.
The track for a moving target is derived from its orbital elements. Orbital elements for all of the planets and most of their satellites are available at STScI. Moreover, STScI has access to the ASTCOM database, maintained by the Jet Propulsion Laboratory (JPL), which includes orbital elements for all of the numbered asteroids and many periodic comets. For other objects, the GO must provide orbital elements for the target in Phase II.
It is also possible to obtain data while HST scans across a small region of the sky. In all cases the region scanned must be a parallelogram (or a single scan line). Two types of "spatial scans" (i.e., raster scans) may be requested: