[PIPE2D-776] Unexplained BAD_DATA in extracted spectra

Wed, 17 Mar 2021 13:45:06 +0000

Using /projects/HSC/PFS/Subaru/CALIB-SuNSS-bad to extract the spectra for 46241 r1 results in missing spectra from some rows of the detector, I think, certainly the same index in the pfsArm.flux data. E.g. 46241 r1 the mask for [1751, 1787, 1816, 1832, 1841, 1869, 1874, 1915, 2016] are all set to 256 (NO_DATA).

See

The data looks fine for that row, and there's nothing obviously wrong with the fiberTrace – and I don't know how that could give problems for a particular pixel for every extracted spectrum.

Using /projects/HSC/PFS/Subaru/CALIB-SuNSS I don't see the problem. The CALIB-SuNSS-bad calibs are not good, they were made with only two 10s fibre trace exposures, but I think we need to know what went wrong. I'd also like to see a bit set for why extractions fail in addition to NO_DATA (which can mean other things too).

PFS-JIRA

[PIPE2D-776] Unexplained BAD_DATA in extracted spectra