MIPS is optimized to provide calibrated images of sources that are small enough that they can be chopped on and off of the arrays, particularly at 70 and 160 µm. The Total Power (TP) mode AOT provides a way to accurately measure extended emission as well, by chopping between an internal dark position and the sky for each array. Experienced MIPS observers will note that the strategy for the 70 µm array changed from pre-launch expectations; now the 70 µm TP observations use side A of the array (the good side) and do indeed chop between the sky location to be measured (using the default 9.98´´ pixel scale), and an internal dark position.
The basic observation sequence begins with a frame at the dark position followed by a stimulator flash frame. Six more frames are taken at the dark position to allow the latent image from the stimulator flash to fade, then the scan mirror is used to change to the sky position, and a single image is taken of the source region. The scan mirror then moves back to the dark position, and 6 more frames are taken, allowing time for the latent image of the sky to fade. A final stimulator flash completes the cycle. This pattern (less the initial dark and stimulator flash frames) is repeated 5 times, providing 5 sky - dark image pairs per AOT cycle. This time-consuming process should provide excellent data on the germanium arrays. At 24 µm, image latents are better behaved. However, a 24 µm stimulator flash is included in this AOT (the only AOT that uses the 24 µm stimulator) in order to provide a reference signal for applying the droop correction to the data.
This mode uses the same imaging arrays at 24, 70 and 160 micron, and therefore is affected by the same problems (see Chapter Chapter 7). Unlike standard photometry, this mode does not uses the scan mirror to dither between positions in the sky, and this adds additional problems with latencies (24 µm), and strong flux non-linearities (70 and 160 micron). Even for observers with experience with Ge:Ga detectors, this mode can be particularly difficult to deal with.
3.1.13 Photometry Raster Map
Scan maps are very efficient, and obtain data at all 3 wavelengths at once. However, for objects that are somewhat larger than 2´ in diameter, but smaller than the 30´ minimum scan map size, the photometry raster map option could be used to construct a map in one or more of the MIPS bands.
The raster map uses the small field dither pattern identical to a single point Photometry/Super Resolution observation (same scan mirror and spacecraft offsets as described earlier). After completing observations in this fashion in a single field-of-view, the spacecraft is offset to an adjacent observer-specified field, and the normal photometry dither pattern is repeated. The observer specified the number of nearly-full-frame offsets in the along scan and cross-scan directions. The raster option returned high-quality 24 µm results because of the high stability of that array. Observers needed to specify offsets that gave some overlap between fields if they were interested in extended emission within the field of their raster map. For full coverage raster maps at 24 μm the observations should have used ½ array offsets, at most.
Another solution to the general problem of covering a small region of sky was to use cluster mode targets instead of raster maps.
Table 3.4 summarizes the expected frames per observation cycle that were obtained in the MIPS photometry and super-resolution observing modes, and the integration time per pixel for those modes.
Table 3.4: MIPS Photometry and Super-Resolution Summary.
Frames /Obs. Cycle1
Approximate Integration Time per Pixel per Cycle2
Compact Source Photometry
42, 140, 420 sec4
30, 100 sec
6, 20 sec
Large Source Photometry
10 / 10
30, 100, 300 sec
6 / 6
18, 60 sec
10 / 106
3, 10 sec
Compact Source Super Resolution
42, 140, 420 sec
8 / 8
24, 80 sec
18, 60 sec
Large Source Super Resolution
10 / 10
30, 100, 300 sec
32 / 328
24, 80 sec
1 Two values indicate number of frames on-source/off-source.
2 For 3 and 10 second exposure times (and 30 seconds at 24 µm) respectively. Times are per pixel on a given sky position in ''MIPS seconds.'' Actual exposure times are 1.05 times longer (see also Table 8.10).
3 At 24 µm, 2 additional frames are taken per AOR, so total integration time will be longer than shown here by 2 times the exposure time. See also next note.
4 For the first cycle in an observation at 24 µm, exposure time is 1 second shorter than shown in this table. See also previous note.
5 The 10 160 µm frames combine to provide a 2´x >5´ filled field of view containing two images of the source.
6 The 10 160 µm frames combine to provide a 4´x5´ filled field of view containing a single image of the source
7 The 3x10 160 µm frames combine to provide a 2´x >5´ filled field of view containing six images of the source sampled at sub-pixel shifts.
8 This indicates total number of frames, not total number of frames on-source (which is 8, just like the compact source)