At the moment all documentation of our CCD production process is done manually (on paper) and ad hoc. There is no central archive of information characterizing each chip we have released for scientific use, and exactly how each chip was produced. The design work for such a database is 90% complete, but we have not identified funding to transfer the paper records into electronic form. Since many of the records are handwritten, in log books, this issue cannot be resolved by simple scanning and ICR.
However, for DEIMOS to be an adequately documented and supported instrument, it would be good to know about at least the set of chips which will be used in its detector mosaic. Chips finished and packaged in our CCD Detector Lab are labelled (on the underside of the package) with a unique ID code. This ID code should be associated, via an online database, with at least the chip's known defect list. Preferably we should also document the provenance of the silicon, the run from which the chips were selected, the process to which they were subjected before packaging, etc. But at the very least the known defect list should be preserved.
Once we know the "names" of the 8 chips inside DEIMOS, their locations in the mosaic should be remembered, and each time chips are moved around (due to failure or other causes) the change of seating arrangement should be documented.
A quick comparison of the known defect list against a flat and zero field for a given chip which we believe is in socket "A" should yield one or the other of two valuable pieces of information: (1) is/isn't this really the chip that is in socket A, and (2) has this chip's defect list grown since we last stored it? If so, we can update the defect list and monitor the degradation of the chip's performance over time.
de@ucolick.org