Previous Article Next Article Table of Contents


Imaging CCDs Optimized to Minimize Effects of Saturated Stars (1Dec95) (from KPNO, NOAO Newsletter No. 44, December 1995) As our imaging CCDs have gotten larger, more bright stars appear in a given CCD frame. Stars that produce signals well beyond saturation in one or more pixels have long been a problem for CCD observers, particularly in previous generations of CCDs that had shallower well depths and therefore bled more easily. The Tektronix/SITe CCDs that we currently use for imaging are much more robust, but we do receive occasional comments and questions from observers on the effects of saturated stars on their data. The most common issue is a "trail" or "streak" to the right of a saturated star (along rows). The amplitude of this trail is usually small (typically a few ADU relative to the sky level), but it can be worrisome. This phenomenon is caused by the saturation of a component in the CCD electronics due to the large amount of charge in the saturated star and the subsequent recovery time. This effect has been referred to as "overshooting" or "streaking" or "shadowing." We have undertaken a project to minimize the effects of bright stars on CCD imaging data. During the summer shutdown, the video circuitry in the CCD electronics was adjusted in order to minimize the effects of overshooting for each of our imaging CCDs. This optimization had to be done for a particular gain value. We chose this gain value such that the digitization limit (32,767 ADU) encompassed as much of the linearity range of the CCD as possible, while avoiding gross undersampling (in e^- /ADU) of the readnoise. No other properties of the CCD besides the overshooting phenomenon (such as read noise, linearity, etc.) were changed by this optimization. The optimized gains for each imaging CCD are given in the table below (and in the Direct Imaging Manual, available on the KPNO home page on the World Wide Web). The ICE software has been updated to reflect these newly optimized gains. The default gains in ICE for these CCDs are the optimized gains. After an obsinit, even if one does not edit the detpars parameter set, one will by default be using the optimized gain. We recommend that all users concerned with the effects of saturated stars on their imaging data use the optimal gains. Also, in terms of grossly saturated stars (> > 200,000 e^- per pixel), we remind the reader that many of the bleeding/trailing phenomena associated with these are unavoidable, so the observer should consider the benefits and costs of moving these stars off the frame or shortening the exposure times if they are causing problems with program objects. Finally, we suggest the use of row and column plots in the analysis of such phenomena seen on the image display, as this provides a more quantitative assessment of the effect. While the effect may look like trouble from the image display, measuring the amplitude and comparing this to other sources of noise is critical for evaluating the effect that any residual overshooting may have on the data. CCD Optimized Gain Optimized ICE Gain Index S2KA 3.8 e^-/ADU 2 T2KA 5.4 e^-/ADU 3 T2KB 3.2 e^-/ADU 4 T1KA 4.5 e^-/ADU 3 Taft Armandroff, Rich Reed, Phil Massey
Previous Article Next Article Table of Contents