Week 8 of Early Operations
A first theme was ongoing investigations into the apparent slow temporal variations of the relative optimal focus offset between different filters. The team performed in-dome tests with the laser tracker metrology system to further characterize the hexapods that control the relative positions of the optical elements. The team ran multiple on-sky tests executing a series of filter changes while making detailed wavefront measurements by pistoning the full focal plane to intra- and extra-focal positions. These measurements have been repeated on multiple nights, in some cases multiple times within a single night, to characterize the system response in a variety of environmental conditions. Analyses of those data are ongoing.
A second area of study was optimal control for hexapod motion during the 30-second exposures for a typical LSST visit, and the impact of various strategies on both the image quality and pointing drift. Since the start of on-sky engineering, the default approach has been to apply small optical corrections with the hexapods while exposuring to continuously maintain optimal focus and alignment. However, those small corrections can sometimes induce a small pointing drift during the exposures. Following several small-scale tests running with and without hexapod compensation while exposing, the team began running full-night tests with pre-LSST observations driven by the Feature Based Scheduler (FBS) with hexapod compensation turned off. Initial results were encouraging, with improvements in the delivered median PSF ellipticity, at the level 0.04 during the survey observations.
The team also began a multi-night campaign of observations to develop a refined active optics system (AOS) look-up table (LUT) for the elevation and rotator angle that extends the current LUT that includes hexapod degrees of freedom to additionally include mirror bending modes. These observations involve running the AOS closed-loop to convergence at a grid of elevation and rotator angle positions. Those data will be supplemented by the FBS pre-LSST observations that naturally span a range of elevation and rotator angle position.
A new all-time record for Rubin for delivered image quality was achieved on 17 December, with a median PSF FWHM across the field of view of 0.58”.
The team started a series of in-dome calibration tests as part of a project to update the sequencer (CCD readout controller) firmware for the LSST Camera to use the natural frequency of the electronics (readout electronics board system clock). This update is expected to reduce structure in the calibration biases. Measurements this week included a short set of biases and flats in preparation for a next stage of tests, including more extensive calibration measurements planned for early 2026.
The team also introduced an abbreviated daily in-dome calibration script so that the measurements can be completed during dark time and observers can get their needed rest after routinely running on-sky programs until -12 deg morning twilight.
Looking ahead, the team is preparing for close out the calendar year 2025 by completing the multi-night campaign for the refined elevation and rotator angle LUT, and then running several full nights of pre-LSST observations with a few different FBS configurations to provide a more complete view of the current distribution of performance, evaluate the suite of the feature deployments and configuration updates made during the past two months, and inform any adjustments for the survey strategy at the start of LSST taking into account the as-built system performance.