NIGHT/EXPID/TILEID = 20241014/257978-80/27338 : stuck fiber on bright star

[araichoor]

tileid=27338 has the stuckFIBER=1604 (LOCATION=3398) on a mag~6 star:

this is the star ("high-proper motion", don t know if that matters): https://simbad.cds.unistra.fr/simbad/sim-id?Ident=TYC++1218+1667+1&NbIdent=1

that messes up the PETAL_LOC=3 redshifts:

I think it s just bad luck, as this positioner has been moved this summer..

the observers did flag those exposures as "bad":

22:09 [05:09] Exp. 257978- Bad - Split 1 of tile 27338, transparency 95%, seeing 1.2 arcsec, survey speed 15%, positioner RMS final move is borderline (5.05), guiding is ok. One warning for high amp read noise in Z. Group of fibres with high flux (1603-1606) which show step changes in flux in uncalibrated spectra. 1604 is very bright object. CCD images for camera 3 show this might be affecting more fibres, especially in B. (SO)
Tile 27338, Exptime: 1304.60, Airmass: 1.37, Sequence: DESI, Flavor: science, Program: BRIGHT

22:33 [05:33] Exp. 257979- Bad - Split 2 of tile 27338, transparency 95%, seeing 1.2 arcsec, survey speed 15%. Same issues as exposure 257978. (SO)
Tile 27338, Exptime: 1013.82, Airmass: 1.27, Sequence: _Split, Flavor: science, Program: BRIGHT

22:52 [05:52] Exp. 257980- Bad - Split 3 of tile 27338, transparency 95%, seeing 1.2 arcsec, survey speed 20%. Same issues as exposure 257978. (SO)
Tile 27338, Exptime: 912.30, Airmass: 1.21, Sequence: _Split, Flavor: science, Program: BRIGHT

the spectro. pipeline still processed those.
here s a zoom on the Z vs. FIBER plot; should we just declare e.g. FIBERS=1500-1749 as bad? or do we want to define a finer mask?

[geordie666]

@abhi0395, @akremin: Can we mark all of Petal 3 as bad for exposures 257978, 257979 and 257980 and re-process tile 27338. Thanks!

[akremin]

I've marked petal 3 as bad for all three exposures, purged, rerun, and updated tsnr and nightqa. Back to surveyops for QA.