|COS Instrument Handbook for Cycle 25|
Prolonged exposure to light causes the COS FUV detectors to become less efficient at photon-to-electron conversion, a phenomenon called “gain sag”. The more a particular region of the detector has been used, the smaller the “pulse height” of the charge cloud generated by an individual photon becomes. As long as all pulse heights are above the minimum threshold needed to distinguish real photons from background events there is no loss in sensitivity. However, as the average pulse height at a location on the detector approaches and drops below this threshold, real photon pulses are increasingly misidentified as background and the effective throughput decreases. Since the amount of gain sag increases with the total amount of previous illumination, these effects appear first on regions of the detector that are illuminated by the bright Lyman-a airglow line, but eventually the entire spectrum becomes affected.STScI is undertaking a number of actions to mitigate the effects of gain sag and extend the lifetime of the COS FUV XDL detector. Primarily, the position of the science spectrum on the COS FUV detectors is periodically moved to an un-sagged region, to temporarily eliminate the “Lyman-a” gain-sag holes and other gain-sag artifacts. On July 23, 2012 the spectrum was moved from its original lifetime position (LP1) to its second lifetime position (LP2). LP2 is offset by +3.5" from LP1 in the cross-dispersion direction. On February 9, 2014, the spectrum was moved to the third lifetime position (LP3), which is offset by -2.5" from LP1, for all modes except the G130M 1055 and 1096 central wavelengths. These settings have wide cross-dispersion profiles that would be severely impacted by the proximity of LP3 to LP1, and so they continue to be executed at LP2. The G130M 1222 central wavelength will execute at LP3, but has been operated at a higher voltage setting to minimize the impact of gain-sagged regions.Preparations are underway for the move to the fourth lifetime position (LP4). The move is currently planned for July 2017, during Cycle 24. LP4 is located at -5.0" from LP1. The spectral resolution at LP4 is expected to be approximately 15% lower than its current value at LP3, although this will be confirmed with calibration observations taken after the move occurs.Starting with the move to LP3, and continuing at LP4, a new spectral extraction algorithm was implemented. This TWOZONE algorithm uses the shape of a point source profile to define the region over which counts are included in the extracted spectrum and to decide when bad pixels in the profile wings compromise the accuracy of the spectral extraction. Sources that have substantial spatial extent may have significant overlap with the gain-sagged regions and may require specialized extractions that are currently not performed with calcos. For these reasons, observations of extended sources will not be optimally calibrated. Users should set the optional APT parameter EXTENDED=YES to flag such sources (see Section 5.9), even if the calcos pipeline calibration will not treat extended sources differently from point sources.Throughput and most other calibrations at LP3, are very similar to those at the original position, and are expected to remain so at LP4. See the COS website and the COS Instrument Science Reports (ISRs) for additional information about the calibration of the different lifetime positions.Gain sag is an inevitable result of using the detector. Gain sag holes will eventually appear at LP4 as well, with the timing of their appearance depending on the locally accumulated signal. The use of all four FP-POS positions, which is now required for most COS FUV observations (see Section 4.1.7), will distribute the high geocoronal Lyman-a flux more uniformly over the detector, and thus will significantly delay the re-appearance of these holes.In order to continue mitigating the effects of gain sag in the FUVB detector caused by airglow emission, when observing with the G130M grating STScI reserves the right to switch the central wavelength setting of any program to another G130M central wavelength setting (cenwave). This is to ensure that the G130M usage in Cycle 25 is such that a single cenwave is not unreasonably causing the detector to sag. Users that require a specific G130M cenwave must justify it in the Phase I proposal.