QC documentation system: QC procedure qc_DARK.sh for NACO

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 GRAVITY MIDI PIONIER   OMEGACAM VIRCAM [best viewing with style sheets and javascript enabled] qc_DARK.sh qc_DETMON.sh qc_CHECKAO.sh qc_LIMGFLAT.sh qc_TIMGFLAT.sh qc_STDIMG.sh   bottom  plots   |   contact NAME qc_DARK.sh VERSION 1.0 -- 2002-03-12 cloned from ISAAC for NACO 1.1 -- 2005-08-17 incorporated from old NACO; new name qc_DCAL.sh 1.2 -- 2015-09-03 use AB instead of DATE and PF as input 1.3 -- 2017-08-30 docu reviewed SYNTAX SHELL CALL processQC from $DFS_PRODUCT/DARK/$DATE: qc_DARK $AB INSTRUMENT NACO RAWTYPE DARK PURPOSE The QC scripts calclates a lot of further QC paramaters per detetcor quadrant and creates a plot the procedure calls DARK_1night.sh PROCINPUT NACO_IMG_DARK_AVG product; QC1TABLE trending | table(s) in QC1 database: naco_dark TRENDPLOT trending | HealthCheck plot(s) associated to this procedure: trend_report_DAILY_DARKRON_HC.html | trend_report_DARKEXTRA_HC.html QC1PAGE trending | associated documentation: dark_qc1.html QC1PLOTS top qc_DARK.gif box1: comparison master (red) <-> raw (blue); upper panel: middle row and middle column, full pixel range, in ADU; lower panel: middle row and middle column, central 100 pixels, in ADU box 2: structure description; all columns (top) and all rows (bottom) are collapsed and displayed; box 3: histogram: all pixel values in log frequency histogram; raw: blue, master: red, reference: green; there should be a dominant peak with Gaussian shape (which translates to parabola shape in this log freq diagram) top label: name of analyzed master bias frame qc_DARK_diff.png full display of difference of the master with corresponding reference frame QC1PARAM The NACO dark recipe generates seven QC parameters: QC.RON1 , read out noise based on the first two raw frames QC.RON2 , read out noise based on the raw frame #2 and #3 QC.DARKMED , median dark level of the product QC.NBCOLPIX , number of cold pixel in the cold pixel map product QC.NBHOTPIX , number of hot pixel in the hot pixel map product QC.NBDEVPIX , numer of deviant pixel in the deviant pixel map This QC script retrieves a lot of QC parameters per detector quadrant and stores them into the QC1 DB Justification for this is the wild history of the NACO detector performance. Since 2015-01-01, NACO is operated with the third detector. Generally the eclipse commands are used to get properties from the master dark. like statistical noise, fixed pattern, large scale gradient corrected values, odd-even column effects, contamination etc. UL stands for upper left quadrant, LR stands for lower right quadrant ALGORITHM Description of algorithms: sigma_master: statistics (std_dev) in quadrants (stcube) sigma_raw: statistics (std_dev) in central window X1024..1124, Y1998..2098 (MIDAS: stat/ima, bins=1, outputr(4)) median_master: statistics (median) of full frame (MIDAS: stat/ima, bins=0.1, exc=100,300) struct_row/col: average/row or col; statistics with exc=mean+/-2 ADU; struct_row/col is the std_dev CERTIF Reasons for rejection: - unusually high dark level or RON level (unless representative); - unusual structure (compare panel 2 plots with older/reference version); - non-Gaussian shape of histogram; - DATANCOM lower than usual (2-3), especially if extra N=3 version exists Also full display of the local difference, i.e. of the master with closest in time conterpart is monitored. COMMENTS The script calls gnuplot for graphics and ESO eclipse for fits manipulation top [2017-09-12T08:53:08]         created by qcDocu v1.1.2, a tqs tool