MULTIACCUM is the only observing mode for the IR channel. An exposure in
MULTIACCUM mode begins with an array reset followed by an initial readout. Next, one or more nondestructive readouts are obtained at user-selectable times. All of the readouts, including the initial readout, are recorded onboard and returned to the ground for analysis. The difference between each successive pair of reads is the image data accumulated between reads.
There are two major advantages of this approach. First, the multiple readouts provide a way to record what is happening in a pixel before it saturates, increasing dynamic range. Second, the multiple readouts can be compared to remove cosmic ray effects. See the
WFC3 Instrument Handbook for more information.
Placement of the target on the detector is controlled by the specified Aperture, the
POSition TARGet <X-value>,<Y-value> special requirement (if used), the telescope orientation (via the
ORIENTation <angle1> TO <angle2> special requirement or by default), and in some instances according to the Spectral Element. The apertures for the IR channel and their valid combinations with spectral elements are defined in
Table 14.3. The current values of the aperture coordinates of the Aperture+Spectral Element combinations in
Table 14.3 may be found on the
HST Apertures Web Page.
The
IR aperture is designed for placing targets at the “optimum center” of the detector. The default location within this aperture will be routinely adjusted by STScI to reflect any changes in detector performance. This aperture is appropriate for targets that are small compared to the scale size of defects in the chips.
The IR-FIX aperture defines the geometric center of the detector and will remain FIXED in aperture coordinates. This location will
not be adjusted for changes in detector characteristics, and should be used to specify the location of the target relative to the detector. This geometric center aperture is appropriate for pointings designed to position an extended scene within the WFC3 FOV.
Three apertures are provided that use the same pointing of the telescope as used for three associated UVIS apertures. Using an associated pair of apertures for UVIS and IR exposures will avoid a small angle maneuver between the exposures. The IR apertures are
IR-UVIS,
IR-UVIS-CENTER, and
IR-UVIS-FIX, which are associated with, respectively, the apertures
UVIS,
UVIS-CENTER, and
UVIS-FIX.
IR subarrays are specified by selecting the appropriate aperture. The
IRSUBnn or
IRSUBnn-FIX apertures will result in the use of the subarray readout mode of the IR detector with the size of the subarray being that indicated by the aperture name (
nn =
512,
256,
128, or
64). The use of the subarray readout mode will result in different sample times than for full detector readouts listed in
Table 14.4. Not all combinations of subarray size and sample sequence are supported. See the discussion under the
SAMP-SEQ optional parameter (
Section 14.3.4) for more details. The subarray readouts will have a border of five reference pixels added around the edge of the subarray used for imaging, making the total data sizes 1024x1024 (full-frame), 74x74, 138x138, 266x266, and 522x522 pixels.
The IRSUBnn apertures will place the target at the "optimum center" of the corresponding subarray; note that these positions may be different for the different subarrays. The default position of each of these apertures will be updated by STScI to reflect changes in instrument performance. These apertures are appropriate for targets that are small compared to the scale size of defects on the detector.
The IRSUBnn-FIX apertures define the geometric center of the subarray and will remain fixed in aperture coordinates. These locations will
not be adjusted for changes in detector performance.
Five apertures are specialized for use with the two IR grisms (G102 and
G141) and to obtain band pass filter images for use as wavelength zero-point references. According to the FOV, the apertures are named
GRISMmm, where
mm = 1024 (full frame), 512, 256, 128, 64. The subarrays are the same as the
IRSUBnn apertures. The fiducial pixel for each Aperture+Spectral Element combination is optimized to best position the first-order spectrum in the FOV. For
GRISM1024,
GRISM512, and
GRISM256, the same fiducial pixel is used for
G102 and
G141 and for reference band pass filter exposures. For
GRISM128 and
GRISM64 different fiducial pixels are used for
G102 and
G141 that best center each first-order spectrum in the FOV. The fiducial pixel for a bandpass filter exposure with those two apertures is midway between the two grism fiducial pixels.
SAMPSEQ=RAPID, SPARS10, SPARS25, SPARS50, SPARS100, SPARS200, STEP25, STEP50, STEP100, STEP200, STEP400
A required parameter specifying the name of a predefined sequence of times from the start of the exposure at which the nondestructive readouts (samples) are performed. The number of readouts (up to 15, plus one for the initial readout) taken for each exposure is controlled by the
NSAMP parameter (see below).
Table 14.4 gives the sample times (defined as the time from the start of the initial readout to the start of a given readout) for each sequence and image size. Different types of sequences are provided. The
RAPID sequence provides linear sampling as fast as possible (limited by the readout time for the selected image size) and is intended for bright targets that could saturate in the other sample sequences. All sample sequences with the full detector apertures are supported. But note that only a limited number of combinations of subarray size and sample sequence are supported.
Sequences STEP25, STEP50, STEP100, STEP200, and STEP400 begin with four rapid samples (five readouts), switch to logarithmic spacing up to the given number of seconds (25-400), and then continue with linear spacing for the remainder of the sequence with adjacent steps separated by 25-400 seconds depending on the selected sequence. These sequences are intended to compensate for any nonlinearities near the start of the exposure and to provide increased dynamic range for images that contain both faint and bright targets.
Sequences SPARS10, SPARS25, SPARS50, SPARS100, and SPARS200 begin with one rapid sample (two readouts) then provide linear spacing to allow observers to "read up the ramp" at evenly spaced intervals. The variety of sampling intervals allows this basic strategy to be applied over a wide range in target flux.
The different sequences are designed to efficiently fill the target visibility period with one, two, or several exposures. See the
WFC3 Instrument Handbook for recommendations on which sequences to use in different situations.
A required parameter specifying the number of samples in a predefined sequence that should actually be taken, not counting the initial readout.
Table 14.4 defines 15 sample times for each sequence. If an
NSAMP value smaller than 15 is used, samples will be taken at only the first
NSAMP times from this table.
The total number of readouts will be NSAMP plus one (for the initial readout, giving a maximum of 16 readouts for a single execution of a
MULTIACCUM exposure). Each readout will be recorded and will appear in the final data set.
Enter the number of times this exposure should be iterated, and the duration in seconds of each iteration. This option should be used in observational situations when two or more identical exposures should be taken of the same field. If the
Number_Of_Iterations is n, the exposure will be iterated n times.
If the exposure is a Spatial Scan (see
Section 7.3.3) and
Number of Iterations > 1, a small slew will be inserted between the exposures so the scans will repeat the same path on the detector each time. This will sacrifice visibility time. Consider alternating the Scan_Direction instead.
Time_Per_Exposure must be
DEF in this Mode. The exposure time is unnecessary, because it is specified by
SAMPSEQ and
NSAMP.
Table 14.4 shows the sequence of 15 sample times corresponding to the different
SAMP-SEQ values, in seconds from the start of the initial readout to the start of the readout for the given sample. These values are given to the nearest millisecond
