Open mpmdean opened 6 years ago
It seems to work for the data taken last summer. It also reduces a lot of the noise across the ccd.
Looks like averaging the overscan region horizontally (which is labeled as underscan because the sides of the chip switch between integration mode and single photon counting mode) and subtracting from the data array works best.
However, some changes to the configuration files of the CCD have decreased the "flashing" and the inherent noise also seems less so the overscan correction doesn't appear as useful now. But we are still working on things.
Is it useful to use overscan to try to remove frame-to-frame offsets in "noise".