Some CCD detectors suffer from bad rows or columns.
These typically present anomalously high counts (usually identical for each pixel in the row/column). A facility is provided to check frames for this problem, any such pixels are then flagged as having bad data quality and omitted from consideration in the subsequent processesing.
As the presence of bad rows/columns is normally a static process (i.e. all images from a particular chip exhibit the same bad row/column), the package only examines one frame at the start of the reduction. This should be the object frame, as the check will append a data quality array to the frame, and it needs to be attached to the object frame to be 'seen' by subsequent processing.
Checking for bad row/columns is disabled by default and may be selected using the parameter TUNE_FCHECK.
An additional parameter allows for 'saturated' pixels to also be flagged as bad, by setting the parameter TUNE_SATRTN to the threshold saturated pixel value.
Martin Clayton