QC documentation system: QC procedure std_arg.prg for GIRAFFE

This documentation is intended both for QC scientists and SciOps astronomers (who may want to ignore the technical information displayed in grey).
 
HOME
 
CRIRES
FORS1
FORS2
KMOS
NACO
 
GIRAFFE
UVES
XSHOOTER
 
ISAAC
SPHERE
VIMOS
VISIR
 
HAWKI
MUSE
SINFONI
 
AMBER
MIDI
PIONIER
 
OMEGACAM
VIRCAM
[best viewing with style sheets and javascript enabled]
bias.prg dark.prg iflat.prg
flat.prg nflt.prg wave.prg simlamp.prg
std_ifu.prg std_arg.prg science.prg
 
bottom  plots   |   contact
NAME std_arg.prg
VERSION 1.0 -- copied from science.prg (May 2005)
1.1 -- adapted to new CCD, scoreQC included (2008-05-28)
1.2 -- efficiency now separate product (2008-07-17)
SYNTAX MIDAS
CALL processQC -a ab_name
INSTRUMENT GIRAFFE
RAWTYPE STD_ARG (Argus standard star)
PURPOSE compares raw and product STD files for FLAMES_GIRAFFE QC assessment; QC1 values are calculated, a QC1 plot is generated, displays are generated. The QC1 data are fed into the QC1 database.
PROCINPUT STD_RBNSPECTRA, EFFICIENCY_CURVE
QC1TABLE trending | table(s) in QC1 database:
giraffe_std
TRENDPLOT trending | HealthCheck plot(s) associated to this procedure:
../FULL/trend_report_SYSTEM_EFFIC_HC.html | ../FULL/trend_report_SYSTEM_EFFIC_WLN_HC.html
QC1PAGE trending | associated documentation:
std_qc1.html
QC1PLOTS
top

std_arg1.gif
box1:crosscut (middle row) through the raw file (red: mbias level); there should be a few fibres illuminated
box2a:histogram (log) of raw file (as another check for signal)
box2b:raw-bias, should show parabolic peak at 0, not shifted
box3:extracted spectrum (not flattened), all fibres collapsed
box4:efficiency curve, before (black) and after (red) sky correction; the sky correction factor is displayed at bottom
std_arg2.png
full display of raw frame; box: area displayed in std_arg3.png

std_arg3.png

closeup of raw frame, a few fibres should display a signal

std_arg4.png
reconstructed image displaying the STD star

std_arg5.png
fibre tracing plots (there are actually 4), with SKY and SIMCAL fibres marked in blue
QC1PARAM QC1 parameters written into QC1 table:
QC1_db names: num_fib | mean_signal | obs_targ_name | airmass | seeing | mean_effic | qc_numsat
FITS key names: PRO.SLIT.NFIBRES | [calculated] | OBS.TARG.NAME | TEL.AIRM.START/END | TEL.AMBI.FWHM.START/END | [calculated] | [calculated]
QC1 parameters calculated by the pipeline: mean_signal | mean_effic | qc_numsat
ALGORITHM mean_signal, mean_effic: stat/ima; qc_numsat: count all pixels > 60.000
CERTIF Reasons for rejection:
- scoring alerts, in particular saturation
- no target visible in reconstructed image
COMMENTS Scoring: mean_signal | mean_effic | num_fib | qc_numsat
ABs fail if target not in catalog
section MARK_TRACINGS is called in science.prg
top [2015-10-29T10:06:35]         created by qcDocu v1.1.2, a tqs tool