3.1.10 Super Resolution - 70 µm Large Source (Large Field), Fine Pixel Scale
The 70 µm large field fine scale mode is nearly equivalent to doing a 2x2 raster map of the 70 µm small field fine scale mode which is centered on the target. It differs from the simple raster in that the scan mirror chops in opposite directions for the top and bottom rows of the 2x2 map. Thus, the result is a 2x4 fine scale map roughly centered on the target. The actual map center is offset slightly so that the central target position is not on the edge of the array in at least one of the raster positions. See Figure 3.15. Note that the viable part of the array is a rectangle (1.25´x2.5´), but since the orientation of the field of view on the sky is a function of time and generally observers could not specify the time of their observation, the largest object that should have been observed using this mode is <2´.
Figure 3.15: Strategy for obtaining super resolution observations over a field of 2.5´x2.5´. At each pair of fields, a set of 16 exposures is obtained similar to the pattern illustrated in Figure 3.13. Note that for simplicity, the array is portrayed as a square; in-orbit realities mean that only half of the array is functional, and the pattern as portrayed here is centered on side A.
In order to cover roughly a 2.5´x2.5´ field around the target, a cluster target should have been constructed with offsets of (0, +60´´) and (0, -60´´) (offsets only). This saved the slew tax from mapping with 2 AORs.
3.1.11 Super Resolution - 160 µm Compact Source (Small Field)
The 160 µm pixels are nominally small enough to provide good sampling of the PSF for high-resolution image reconstruction. However, the impact of cosmic rays and the complicated behavior of the Ge detectors mean that integration time and PSF sub-sampling beyond what is provided by the normal 160 µm photometry mode were needed to obtain good super resolution results. Unlike the 70 µm array, there was no separate pixel scale or tailored AOT available for 160 µm super-resolution observations. The recommended method for specifying such an AOR using the available 160 µm pixel scale was an observation consists of 24 individual frames that provide three 2´x5.3´ images, each containing 2 images of the source, with ½ pixel sampling in both the scan and cross-scan direction. The third image has the same cross-scan direction sampling of the PSF as the first image, but is placed on different pixels of the array to provide extra protection against unusual behavior of the Ge:Ga pixels and intra-pixel responsivity variations.