ACS Status

Linda Smith, lsmith@stsci.edu, David Golimowski, golim@stsci.edu, & Norman Grogin, nagrogin@stsci.edu

The Wide Field Channel (WFC) and the Solar Blind Channel (SBC) of the Advanced Camera for Surveys (ACS) are continuing to perform well after the repairs made during Servicing Mission 4 (SM4). All aspects of the WFC performance were checked during the servicing mission observatory verification (SMOV) period. The replacement electronics have given the WFC detectors new characteristics as described below.

The replacement CCD electronics box (CEB-R) is equipped with two correlated-double sampling (CDS) modes (clamp and sample, dual-slope integration), which can be used to measure the charge accumulated by each pixel during read out. The clamp-and-sample mode was used successfully in the old CEB, but the dual-slope integrator (DSI) offers lower read noise at the expense of more bias structure. Post-SM4 image analysis confirms that the DSI yields significantly lower read noise (3.9–4.7 compared to 5.5 electrons), and that bias frames exhibit a 5–10 digital-number (DN) gradient spanning the rows of each image quadrant. The bias gradient is stable and can be precisely removed during normal image reduction and processing. Consequently, the DSI has been selected as the default CDS mode for post-SM4 observations.

Another characteristic of the new WFC electronics is the presence of faint horizontal stripes that extend along the rows of the CCD and across the quadrant boundaries. The stripes are constant along each row of pixels, but they are not stable from frame to frame. The stripes are caused by low-frequency (1 mHz to 1 Hz), 1/f-noise on the reference voltage, generated by the application-specific integrated circuit (ASIC) used to offset the pixel signal after CDS is performed. The stripes are present in all WFC calibration and science images, regardless of CDS mode. The contribution of the stripes to the global read noise statistics is small (the peak-to-peak deviation is approximately 2 DN), but the correlated nature of the noise may affect photometric precision for very faint sources. The ACS team has developed algorithms for removing the stripes from calibration and general science images, and we are in the process of releasing these algorithms to the community as standalone software packages that operate on the images independently of the CALACS software package.

The level of ACS/WFC amplifier crosstalk after SM4 is the subject of a recent instrument science report (ISR 2010-02; view here) by A. Suchkov, N. Grogin, M. Sirianni, E. Cheng, A. Waczynski, and M. Loose. When each quadrant of the two WFC CCD detectors is read out, electronic crosstalk between the amplifiers induces faint, typically negative, mirror-symmetric ghost images on the other three quadrants. The effect is strongest for high-signal pixels. Analysis of pre-SM4 crosstalk showed that its impact on ACS/WFC science is not significant and can be ignored in most science applications. Analysis of the crosstalk after SM4 shows that the crosstalk due to low-signal pixels is much weaker than before SM4 and does not produce ghosts similar to those seen in pre-SM4 images. For high-signal pixels, we find substantial differences between the gain = 1 e/DN and gain = 2 e/DN cases. For the default gain setting of 2, the crosstalk is similar to what it was before SM4: up to 5–8 e per pixel on the same CCD. For gain = 1, the crosstalk is ~100 e per pixel for saturated pixels on the same CCD, which is more than an order of magnitude above the pre-SM4 level. The crosstalk from saturated pixels is ~20–30 e per pixel on the other CCD, which is also much higher than it was before SM4.