12.5 Mode = ACQ/SEARCH
Config = COS/NUV

12.5.1 Aperture or FOV

12.5.2 Spectral Element

12.5.3 Wavelength

12.5.4 Optional Parameters

12.5.5 Number of Iterations

12.5.6 Time Per Exposure

12.5.7 Special Requirements

This mode invokes a procedure in the COS onboard flight software that searches for the target and locates it in the selected COS aperture for subsequent science exposures. A series of exposures are taken at different HST pointings in a spiral pattern from the initial pointing, and then the HST is repositioned to maximize the light intensity through the COS aperture. This mode is required for targets whose coordinates are not known to at least 0.4". Most COS observation sequences that use ACQ/SEARCH mode will also need to use other acquisition exposures (ACQ/IMAGE, ACQ/PEAKD, and/or ACQ/PEAKXD) to achieve proper centering of the target in the science aperture.

Details on the target acquisition sequence and limiting magnitudes for the PSA and BOA can be found in the COS Instrument Handbook.

This method in the NUV uses undispersed or dispersed light from the object to be observed. In dispersed light, the total counts from all three stripes are used to maximize the photon counting statistics. There are no appreciably bright airglow lines in the NUV portion of the spectrum, but subarrays are still used to exclude portions of the detector which are not illuminated by the input spectrum or image; this improves accuracy by reducing the contribution from background counts. The subarray dimensions depend on the selected spectral element; see the COS Instrument Handbook for details.

12.5.1 Aperture or FOV

The following apertures are allowed:

•	PSA Primary Science Aperture

•	BOA Bright Object Aperture

Use of the BOA is recommended on bright targets to attenuate the flux and allow the acquisition to proceed without triggering bright object violations.

12.5.2 Spectral Element

Enter a spectral element from Table 12.3 for the COS/NUV configuration. If a grating is selected, this will generally be the same as the one used for the subsequent TIME-TAG or ACCUM science observation. If the MIRRORA or MIRRORB spectral element is selected, the search for the target will be done in undispersed light. Use of the MIRRORB spectral element (possibly in conjunction with the BOA aperture) will attenuate the flux and allow undispersed-light acquisitions to be done on targets that would otherwise trigger bright object checks.

12.5.3 Wavelength

Enter the value of the central wavelength in Angstroms. Table 12.3 gives the allowed values of the central wavelength for each grating.

Note: No wavelength should be specified if MIRRORA or MIRRORB is the spectral element.

12.5.4 Optional Parameters

SCAN-SIZE
= 2, 3, 4, 5 (points)

This is a required parameter that specifies the size of the square spiral search region in terms of the number of dwell points on a side (e.g. 2x2, 3x3, 4x4, or 5x5). A larger search pattern samples a larger area, but the time required scales as the square of SCAN-SIZE.

Note: 2x2 dispersed light searches with the NUV channel should be avoided in most cases, as simulations show significantly degraded accuracy; a 2x2 search can even worsen the alignment of a target that is initially well centered. For more details on target acquisition strategies, please consult the COS Instrument Handbook.

STEP-SIZE
= 1.767 (default); 0.2 - 2.0 (arcsec)

Specifies the size in arcseconds of each step in the spiral search. Both the PSA and BOA are 2.5 arcsec in diameter. The recommended offset of 1.767 arcsec is the offset at which diagonal dwell points just overlap, and the maximum spacing that provides continuous coverage. Simulations show that offsets larger than 1.767 arcsec will introduce errors due to unsampled areas within the search pattern.

CENTER
= DEF (default), FLUX-WT, FLUX-WT-FLR, BRIGHTEST

Specifies the method used for locating the target within the search pattern. Two methods are available: (1) a flux-weighted centroiding algorithm (FLUX-WT or FLUX-WT-FLR), or (2) a return to the brightest dwell point (BRIGHTEST).

If FLUX-WT-FLR is used, the minimum number of counts measured from any of the dwell points (the "floor") will be subtracted from the number at other dwell points before computing the centroid. The idea behind this threshold is to reduce the contribution of background counts. Note that this has the effect of eliminating the dwell point with the minimum counts from the centroid calculation. This drastically reduces the centering accuracy for SCAN-SIZE=2; therefore FLUX-WT-FLR is not allowed if SCAN-SIZE=2 (FLUX-WT is recommended).

Because of the higher detector background of the NUV detector, use of the floored threshold is considered necessary to achieve acceptable accuracy with the flux-weighted centroiding algorithm. Therefore, the default is to use FLUX-WT-FLR with a SCAN-SIZE of 3 or larger. See the COS Instrument Handbook for recommendations on using these different methods.

12.5.5 Number of Iterations

The Number_Of_Iterations must be 1 in this Mode.

12.5.6 Time Per Exposure

Enter the total time of data collection at each dwell point as Time_Per_Exposure. Time_Per_Exposure must be an integral multiple of 0.1 seconds. If it is not, its value will be rounded down to the next lower integral multiple of 0.1 sec, or set to 0.1 seconds if a smaller value is specified.

The procedures to determine the exposure time for the ACQ/SEARCH Mode exposure are given in the COS Instrument Handbook. The exposure time will be repeated at every dwell point in the search.

12.5.7 Special Requirements

The special requirement POSition TARGet <X-value>,<Y-value> is not permitted on ACQ/SEARCH exposures.