lamp flux [ADU/s] in D2 flat for 1.0x11 slit vs. time (100k/hg, 1x1)
maximum counts [ADU] in D2 flat for 1.0x11 slit vs. time (100k/hg, 1x1)
maximum counts [ADU] in D2 flat for any slit slit vs. time (100k/hg, 1x1)
maximum counts [ADU] in D2 flat vs. slit width (100k/hg, 1x1)
lamp flux [ADU/s] in QTH flat for 1.0x11 slit vs. time (100k/hg, 1x1)
maximum counts [ADU] in QTH flat for 1.0x11 slit vs. time (100k/hg, 1x1)
maximum counts [ADU] in QTH flat for any slit vs. time (100k/hg, 1x1)
maximum counts [ADU] in QTH flat vs. slit width (100k/hg, 1x1)
-arc_lamp:
normalized average flux of selected ThAr lines in FMTCHK frame
number of selected ThAr lines in FMTCHK frame
-source_spectrum:
current flat / reference flat = a + b (lam - lam0), current flat / reference flat = a + b (lam - lam0)
For this QC, we follow this algorithm:
We fit the flat field spectrum via the following linear formular:
current(lambda) = reference(lambda) ( a + b ( lambda - lambda_0 )
where lambda_0 is the center wavelength of the considered wavelength range, parameter a is the relative overall flux of the spectrum, corresponding to the lamp power and parameter b is a first order chromatic deviation (the slope between the reference flat spectrum and current flat spectrum). The reference spectrum is from 2015-10-24.
The plot will show both parameters a and b, of which paramter b is the important one.
On 2015-10-24, there is a=1 and b=0 by definition.
E.g. a = 1.22 means the current flux calibrated flat spectrum is 22% brighter than the flux calibrated reference spectrum.
A negative b (e.g. b = -1e-4) means the spectral slope of the current spectrum is less steep, hence blueer than the reference spectrum. In the trending plot, increasing b means the spectrum becomes redder, decreasing b means the spectrum gets more blue.
SLIT:
rms of the slit function, any slit, only 1x1 binning
illumination gradient of the slit function, any slit, 1x1 binning only,
error in slit illumination gradient b, any slit, 1x1 binning only
rms of the slit function in IFU mode, 1x1 binning only,
illumination gradient b of central slice, 1x1 binning only, error in b, 1x1 binning only
(September 12, 2021: Dave removed 4 and 5 above. Think they are only needed for IFU mode)
The median of the 21 central rows of the master flat is taken and a cross section over one order is extracted. (see plot tutorial for details). For flats in slit mode an analysis region is determined exluding both edges. For flats in IFU mode the analysis region is confined to the central slice.
Boxes 1 and 4 show the rms of the analysis region.
The analysis region is fit by a linear function
y = a + b * x
where b represents the illumination gradient over the slit. This QC parameter is given as gradient in box 2 for slits and in box 5 for the central slice in IFU flats.
The uncertainty of the gradient b, called grad_err, is given in boxes 3 and 6.
Thresholds are based on the experience gained during the 2017-02 events in the UVB-arm
stdev: The standard deviation over a sample of five raw frame median counts. This QC parameter is sensitive to fluctuations in the illumination level., The standard deviation over a sample of five raw frame mean counts. This QC parameter is sensitive to fluctuations in the illumination level., The standard deviation over a sample of five raw frame maximum counts. This QC parameter is sensitive to fluctuations in the illumination level and/or to fluctuations of the jnumber of bad pixel, The maximum over a sample of five raw frame maximum counts. This QC parameter is sensitive if only frame of the sample saturates or has more counts
thespacedoctor
commented
3 years ago
Here the QCs for Lamp, Arcs, and flat fields
-arc_lamp:
-source_spectrum:
rms of the slit function in IFU mode, 1x1 binning only,illumination gradient b of central slice, 1x1 binning only, error in b, 1x1 binning only(September 12, 2021: Dave removed 4 and 5 above. Think they are only needed for IFU mode)