WFI Weekly & Daily Health Check Instructions


This page contains instructions on how to run the Weekly and Daily WFI Health Checks at the 2p2. NOTICE: as of 2004-08-26 the health check reductions are done from the e2p2ops account of w2p2off, and the sciops account of epu. There is no need to position oneself on the directory where the images are stored. Just run it from any e2p2ops console. In any case, the images and reduction are done in the directories /data/reduced/WFI/DHC and /data/reduced/WFI/WHC.


On the oh machine, in Service account:

  • Start up OT
  • From the WFI CALIBRATION queue choose either the weekly_health_check or the daily_health_check OBs.
    NOTE: The weekly generates 35 images (5 bias and 30 betas), and the daily only 4 (2 bias and 2 betas).
    Notice that the beta-light images are now called WFI_FlatLampIma.fits.
    Make sure that all the images go to the same "date" directory, that is, do not start these OBs if the UT will cross the 12:00 UT time during the test.

On the w2p2off machine as user e2p2ops:

  • Type export MID_WORK="/data/reduced/WFI/midwork"
  • From any workspace start the midas session typing wfiStartHCMidas.
    This creates a MIDAS console on the first display window (workspace Health Check), and a graph and image window on the Health Check workspace of the second display window.
  • Daily Health Check: from any directory type wfiReduceHC daily date bias1 nbias beta1 nbetas T
    • daily is the string that indicates that this is the daily health check.
    • date is the date of the test in the format used to name the directories in /data/raw e.g, yyyy-mm-dd.
    • bias1 is the frame number of the first WFI_Dark.fits (0 for the first bias of the day).
    • nbias is the number of bias frames taken.
    • beta1 is the number of the first WFI_FlatLampIma.fits file (0 for the first beta of the day).
    • nbetas is the number Beta Lights frames taken.
    • T is a single character indicating that we want extensive output (T or F for debugging).
    • Example: "wfiReduceHC daily 2004-08-26 1 2 3 2 T", for the case in which the first bias is stored in WFI_Dark.1.fits, and the first beta-light in WFI_FlatLampIma.3.fits.

  • or Weekly Health Check: from any directory type wfiReduceHC weekly date bias1 nbias beta1 nbetas T :
    • Example: wfiReduceHC weekly 2004-08-26 19 5 34 30 T"

  • Check the values of RON, GAIN, BIAS and Flat Field Levels against the reference values (listed in the midas display). If everything is O.K. continue with the transfering and publishing od the data. If not O.K. solve the problem and repesat the test, (i.e. FIERA Crash can result in zero bias and Flat Field levels).
  • Run the script: wfiDHC2epu [epu password] or wfiWHC2epu YYMMDD [epu password] for the daily and weekly health checks respectively

  • If the daily health check script fails you need to run in the midas console the command indisk/fits mosaicNew.fits mosaicNEW.bdf (notice the change from small caps to capital in New), and then repeat the wfiReduce command.
  • If the weekly health check fail change directory to /data/reduced/WFI/WHC and remove all files of the types: ccd*cat, ccd*bdf, and *YYMMDD*, then run the wfiReduceHC again.

On epu as user sciops:

  • Run the script to update the web pages: wfiUpdateWHC WFI or wfiUpdateDHC
  • Check the output for each chip paying particular attention to:
    1.- Is the Bias level within 2 sigmas ?
    2.- Is the intensity level of the last point following the line from the previous points ?
    3.- Is the gain within 2 sigmas ?
    4.- Is the RON within 2 sigmas ?
    5.- Is the shutter error within 2 sigmas ?
    6.- Is the linearity error OK ?
    If the answer to any of the above questions is NO, run the test again next day.

    Notice that the Beta Light output light depends on temperature ~ -0.3% / �C. So you would expect some variation in intensity if the test was done at too extreme temperature.