Calibrations -- details on "to do" list

Improve calibrations, both pedestals and "mip" peaks.
Pedestals from calibration runs?
Pedestals from later runs?
better mip peaks

The pedestal calibrations attempt to find the mean and width (sigma) of the pedestal. We cut ~3 sigma above the pedestals to separate signal from noise. The problem is that we can't fit the pedestals in the runs we use for the analysis -- we have zero suppression on for all data-taking runs. The only pedestals we can look at in real data taking runs are the one of every 16 channels which are not zero suppressed. In order to set up zero suppression, we do periodic standalone runs (with no beam) without zero suppression. These set the zero suppression cuts used online. These calibration runs may give us better pedestals than we get in any other way -- I'm not exactly sure how Sangsu has been doing this so far. Unfortunately, our scripts which did the pedestal runs somehow because broken. They typically did not run for as many events as they should have. The best pedestal runs are from very late in run 3. We do not have clear information on whether or not the pedestals varied much with time -- be believe it was better than in previous runs however. This is something that needs study.

The mip peaks can obviously only come from runs with the beam on. These are most important in the analysis of data from strips.

In general, all the gains (i.e. channels per mip) in a given AMU/ADC chip (32 adjacent channels) are very similar. We may choose to get the final parameters of these mip peaks from the non-zero suppressed channels in the runs (every 16th channel) and assume they can be applied for all channels in the chip.

This web page contains a collection of links related to calibrations.