mirror sites: PL (internal link) HQ     [?]

Quality Control and
Data Processing


CAL | HC | refs | QC
QUALITY CONTROL
    HOME
MUSE QC
Trending & QC1
   wave
Pipeline
MUSE ...
QC links:
MUSE: Wavelength solution and resolution
Resolution and Central Wavelength | Wavelength Calibration Fit | Flux of the Arc Lines

 
HC PLOTS
Resolution in Nominal(N) mode
Resolution in Extended(E) mode
Resolution per channel
Wavelength calibration fit N mode
Wavelength calibration fit E mode
Flux of arc lamps
QC1 database (advanced users): browse | plot

Arc lamp calibrations are taken regularly in one instrumental setup: INS.MODE=WFM-NOAO-N, INS.OPTI1.NAME=Blue, DET.READ.CURNMAE=SCI1.0. Each set of wavelength calibrations consists of fifteen input raw frames (multiple of 3) from different arc lamps: Ne, Xe, and HgCd (5 input frames per lamp). Whenever science observations are executed in a different instrumental setup, the corresponding arc lamp frames are taken as well, e.g, in the AO (Adaptive Optics) mode.


Arc lamp exposure. Example of the input arc frame - the Ne lamp exposure, for CHANEL 10 (EXTNAME=CHAN10). Each of the 48 slices (vertical stripes) contains spectrum of the Ne lamp. The full input arc frame contains 24 of such extensions - one extension for each of 24 channels of MUSE.


Resolution and Central Wavelength
Resolution and Central Wavelength | Wavelength Calibration Fit | Flux of the Arc Lines

QC1_parameters

FITS key QC1 database: table, name definition class* HC_plot** more docu
QC.WAVECAL.SLICE24.RESOL muse_wave..slice24_resol Mean spectral resolution R determined in sliceHC [docuSys coming]
QC.WAVECAL.SLICE1.RESOL muse_wave..slice1_resol Mean spectral resolution R determined in sliceHC [docuSys coming]
QC.WAVECAL.SLICE24.WLEN muse_wave..slice24_wlen [Angstrom] Wavelength associated to WLPOS in slice HC [docuSys coming]
QC.WAVECAL.SLICE1.WLEN muse_wave..slice1_wlen [Angstrom] Wavelength associated to WLPOS in slice HC [docuSys coming]
*Class: KPI - instrument performance; HC - instrument health; CAL - calibration quality; ENG - engineering parameter
**There might be more than one.

Trending

The spectral resolution averaged across all detectors (CHANELs), including rms, is monitored for central slice24 and border slice1. We monitor it separately for the data taken in the WFM, Nominal and Extended modes (both NOAO and AO). Also, the resolution is monitored for each detector. Only data taken in the read-out mode (DET.READ.CURNAME) SCI1.0 are selected.

Scoring&thresholds Resolution and Central Wavelength

The thresholds are used for the HC settings - the wavelength calibrations taken in the WFM-NOAO-N and WFM-NOAO-E modes. The lower thresholds for resolution are relatively tight to be sensitive to a sudden degradation. The central wavelength is very stable in general, so the thresholds are not so tight to avoid false alarms.

History

Date event
February 2014 wavelength calibrations are taken regularly

Algorithm Resolution and Central Wavelength

To compute the wavelength solution, arc lines are detected at the center of each slice (using threshold detection on a S/N image) and subsequently assigned wavelengths, using pattern matching to identify lines from the input line catalog. Each line is then traced to the edges of the slice, using Gaussian centering in each CCD column. The Gaussians not only yield center, but also centering error, and line properties (e.g. FWHM).


Wavelength Calibration Fit
Resolution and Central Wavelength | Wavelength Calibration Fit | Flux of the Arc Lines

QC1_parameters

FITS key QC1 database: table, name definition class* HC_plot** more docu
QC.WAVECAL.SLICE24.FIT.NLINES muse_wave..slice24_fit_nlines Number of arc lines used in calibration solution fit in sliceHC [docuSys coming]
QC.WAVECAL.SLICE1.FIT.NLINES muse_wave..slice1_fit_nlines Number of arc lines used in calibration solution fit in sliceHC [docuSys coming]
QC.WAVECAL.SLICE24.FIT.RMS muse_wave..slice24_fit_rms [Angstrom] RMS of the wavelength calibration fit in slice HC [docuSys coming]
QC.WAVECAL.SLICE1.FIT.RMS muse_wave..slice1_fit_rms [Angstrom] RMS of the wavelength calibration fit in slice HC [docuSys coming]
*Class: KPI - instrument performance; HC - instrument health; CAL - calibration quality; ENG - engineering parameter
**There might be more than one.

Trending

The number of identified arc lines used in calibration solution fit, as well as the fit rms, averaged across all detectors (CHANELs), is monitored for central slice24 and border slice1. The parameters are followed as a function of time and as a function of ambient temperature (INS.TEMP4.VAL, qc parameter "temp_ambient"). Only data taken in the WFM-NOAO-N, WFM-NOAO-E modes and read-out mode (DET.READ.CURNAME) SCI1.0 are selected.

Scoring&thresholds Wavelength Calibration Fit

The thresholds are used for the Health Check settings - the wavelength calibrations taken in the WFM-NOAO-N and WFM-NOAO-E modes. The lower thresholds for number of arc lines used in the fit are relatively tight to be sensitive to a sudden degradation.

History

Date event
February 2014 wavelength calibrations are taken regularly

Algorithm Wavelength Calibration Fit

To compute the wavelength solution, arc lines are detected at the center of each slice (using threshold detection on a S/N image) and subsequently assigned wavelengths, using pattern matching to identify lines from the input line catalog.


Flux of the Arc Lines
Resolution and Central Wavelength | Wavelength Calibration Fit | Flux of the Arc Lines

Flux of each wavelength calibration lamp, of the Ne, Xe, and HgCd lamp is monitored.

QC1_parameters

FITS key QC1 database: table, name definition class* HC_plot** more docu
QC.WAVECAL.SLICE20.LAMP3.LINES.PEAK.MEAN muse_wave..slice20_lamp3_peak_mean [count] Mean peak count level of lines of lamp 3 (Ne) above background in slice20 HC [docuSys coming]
QC.WAVECAL.SLICE20.LAMP4.LINES.PEAK.MEAN muse_wave..slice20_lamp4_peak_mean [count] Mean peak count level of lines of lamp 4 (Xe) above background in slice20 HC [docuSys coming]
QC.WAVECAL.SLICE20.LAMP5.LINES.PEAK.MEAN muse_wave..slice20_lamp5_peak_mean [count] Mean peak count level of lines of lamp 5 (HgCd) above background in slice20 HC [docuSys coming]
QC.WAVECAL.SLICE20.LAMP3.LINES.PEAK.MAX muse_wave..slice20_lamp3_peak_max [count] Peak count level of lines of lamp 3 (Ne) above background in slice20HC [docuSys coming]
QC.WAVECAL.SLICE20.LAMP4.LINES.PEAK.MAX muse_wave..slice20_lamp4_peak_max [count] Peak count level of lines of lamp 4 (Xe) above background in slice20HC [docuSys coming]
QC.WAVECAL.SLICE20.LAMP5.LINES.PEAK.MAX muse_wave..slice20_lamp5_peak_max [count] Peak count level of lines of lamp 5 (HgCd) above background in slice20 HC [docuSys coming]
*Class: KPI - instrument performance; HC - instrument health; CAL - calibration quality; ENG - engineering parameter
**There might be more than one.

Trending

We monitor flux of each MUSE arc lamp: Ne, Xe, and HgCd, in selected slice20, averaged across all IFUs. It is in units of electrons (due to FITS standard regulations 'counts' is used instead).

Scoring&thresholds Flux of the Arc Lines

The thresholds are used only for the lamp_peak_max parameters. The upper limit is set to warn of possible saturation of the arc lines.

Although good indicator of the flux extrema, this parameter may not always be sensitive enough. Since the median tries to discard all bad (saturated pixels, i.e. with raw ADU>65500) and the image combination of the 15 exposures has done the same before, it's much less likely to contain (near-)saturated pixels. Also, often lines are not saturated over the whole width, because of the poor sampling on the CCD, and so values ~8000 counts below the 65500 limit naturally occur, even when lines are saturated. And finally, the saturation may occur in different slice than slice20 (probably has to do with the graded AR coating of the CCD and the resulting sensitivity to a given wavelength).

History

Date event
2016-03-11 Xe lamp replaced and aligned
2016-05-19 Ne lamp replaced and aligned
2016-06-05 HgCd lamp replaced; flux appeared too high, so the exp_time was decreased (from 0.8s to 0.65s), PPRS-065580
2016-07-02 Testing lamps with different UIT caused saturation of some HgCd lamp frames(three frames at ~71 ADU)
2017-04-10 After recent adjustment of the arc lamps and integration times the thresholds were modified
2017-07-15 Xe lamp was replaced (PPRS-069401)
2017-10-10 New version of the MUSE pipeline adopted, v.2.2. The number of arc lines used for fitting in AO data decreased; This is because not all the detected arcs are used; Some are dropped from fitting; Scoring thresholds adjusted.
2017-10-22 Integration time of the Ne arc lamp was reduced by 20% to avoid saturation of the strongest lines (PPRS-070494)
2018-06-18 HgCd arc lamp was replaced (PPRS-072811)
2018-10-27 the Pen-ray lamp Xe was replaced (PPRS-074217)
2018-11-27 Ne lamp was replaced (PPRS-74524)

Algorithm Flux of the Arc Lines

The maximum flux is defined as a peak of all lines in a given lamp, in a spectrum that is a median of the three middle CCD rows of slice20.


Send comments to <qc_muse@eso.org> powered by QC [webCMS v1.0.1]