schlafly
commented
2 years ago This was also caught by the autofocus analysis. I don't know if we have access to that information in the file headers, but if we did, that would be a useful alert to have. @crockosi might know.
The other hint we had here was that the RMS numbers were all high. We might make a warning for those when they're larger than 0.1 (guessing!), or turn them yellow. IIRC the petal gets dropped when RMS are >0.2, but >0.1 has been indicative of out-of-focus.
Finally, as usual in this case, we could imagine keeping the tile. The GFAs are on the edge and the focus was best in the center, so all of our fibers got more effective time than we expected. We haven't been keeping them because they're obviously problematic, our effective times won't be accurate, and they're very rare. I don't really want to change that, but we could consider it.
Out of focus plot like this one (tilei=23108, exposure=128784) is a useful plot to be included in the tile-qa plot.
One problem is that those tile qa pages are already quite packed, so we would need to make room for this one.
Comment from @schlafly: "I think I might replace the turbulence plot with the out-of-focus plot, and maybe reduce the turbulence plot to a single RMS that gets printed somewhere."
I was initially thinking to remove the sky map, and just report the Galactic latitude (and maybe distance to the Sagittarius Stream?) Along with the tile ra, dec I guess that the people doing the tile vi are familiar enough with all that to figure out where the tile falls in the footprint without the actual plot.