Roberto reported that, in some cases, the values for the A, B1 and B2 coefficients for f/8 drift compared to the usual values seen in the TCS. For example, A shows almost 600 units and  B2 goes to 720, which is higher than usual. Roberto forced a f/8 correction and the values go much closer to the "default" values (A around -230 and B2=323). I took test images before and after the correction and the FWHM improved from 4.0–4.2 pixels down to 3.5–3.7 pixels, or from 5.5 to 4.5 pixels in another case. It looks like the values change to the higher values when filters are changed (JX -> HX, for example). Is it possible to check if this is the expected behavior of the system?

The 2nd-half night has a problem in star elongation, which may be due to the strong wind, suggested by Sean. We decided to use guiding for the 180s exposures. However, the guidling failed at the begining, but it comes back after several exposure.