Inside, the dome was much calmer, and we got to work aligning the calibration unit.
During the calibration process, we put known shapes (called "modes") on the mirror and measure the resulting image distortions with the wavefront sensor. The more modes you can sense and correct, the better your final image will be. Thankfully the Arcetri team had laid the framework for us, providing a validated set of 500 modes for us to use for calibration.
Over the course of the next several days, we took closed-dome (off-sky) calibrations with up to 500 modes - the highest number of modes we've ever been able to calibrate on either side!
While we don't yet know if the 500 mode calibration can be stable on a real star (the seeing was too poor to use it last night), we our new 400 mode calibration performed well on-sky. For the first time, both the right and left sides have 400 modes of correction available, which should make for some excellent high-resolution images this fall.
quote of the day(s): "That's a great modal plot!" - Alfio, approving of the performance of our 500 mode calibration. We breathed a sigh of relief after that!
[He was referring to the plot below, created closed-dome using simulated turbulence. The x axis is the mode number (low spatial frequencies on the left, high on the right) and the y axis is the wavefront error in each mode. The red line is the raw wavefront, and the white line is after AO correction]