This is the trending report for the grating stability in FLAMES/UVES, derived from the MOS format check calibrations taken with wavelength setting 520, 1x1 binning and Plate 1. This setup measures the position of CD#3. It is partly redundant with the 580 P1 plot (which monitors the same cross-disperser) but is displayed for completeness.

Since 2010 these data are taken daily as part of the Daily Health Checks.

Calibrating SCIENCE observations made during the night with calibrations taken during the day relies on the stability and reproducibility of the positioning of the instrument components. To insure an accurate wavelength calibration, one of the critical criteria is the accuracy of the positioning of the gratings. If the highest possible accuracy is required for the wavelength calibration, an attached wavelength calibration should be taken with the science. Alternatively, since the exact grating position can not be measured in the science image itself, we rely on the daily health check format check data to verify the stability and reproducibility from day to day, assuming that this is a reaasonable indication of the reproducibility between nighttime science and daytime calibrations.

With this in mind the important aspects in these plots are the day to day scatter. The absolute level, within reason (i.e. plus or minus 10 pixels) is not important. From time to time there are sudden jumps in the mean levels, often due to earthquakes, but often the cause of the jump is far from obvious. Such jumps are not in themselves a cause for immediate concern. Only if they are happening frequently, and liable to be happening during the night. Or if they take the instrument so far from the Physical Model that the pipeline will have serious difficulties to converge to a solution, as a rule of thumb, intervention to realign should be considered if the residuals exceed plus or minus 10 pixels.

The individual plots describe, separately for the UPPER CCD (upper row) and the LOWER CCD:

• the mean residuals in X and Y of the fitted position of the emission lines with respect to the positions expected from the Physical Model;
• the number of emmission lines selected in the final solution; trends in this parameter can be due to lamp aging effects.

Find more information about FLAMES/UVES Grating Position monitoring here.