MNBrod
commented
6 months ago We are working on a more robust CRR solution, but in general we've found that the red chip is just challenging to clean by nature. The pipeline supports stacking multiple objects to reject CRs, which is the current recommended approach (obviously this only works if you have multiple exposures of the same target in the same configuration).
We're aware of the issue, and working on it.
I've successfully reduced some blue arm data using the current pipeline, however, the red arm data looks quite poor once reduced and stacked. It's quite clear that there are many cosmic rays not removed from the data. See attached video where the data cube is being stepped through in wavelength slices. Essentially all you see is many transient non-physical detections.
Screencast from 19-10-23 16:21:08.webm