VYSOS5 setup notes from Eugene Magnier (2009.01.19)

Software Installation & Data Injection

• I build the IPP software, fixing some compile-time errors caused by a more strict C compiler (did not allow function prototypes without arguments: Foo() -> Foo(void))

• I installed the ippMonitor software, but needed to get DB.php installed before it could be used (done by Martin P.)

• I created the basic ippadmin database using the ippadmin tool.

• I set up a test database, vysostest, and one for simtests called ipptest.

• I injected all images from 20080920 (the directory provided by Martin for testing):

ipp_serial_inject.pl --telescope vysos --camera vysos --workdir /home/ipp/data/work --end_stage reg --dbname vysostest *.fts

(same command for the science images in /home/ipp/data/20080920 and the subdirs with the detrend images)

Detrend Creation

• build a bias:

dettool -pretend -dbname vysostest -definebyquery -workdir /home/ipp/data/work/detrend -inst VYSOS5 -det_type BIAS -select_exp_type BIAS

• build a flat:

dettool -pretend -dbname vysostest -definebyquery -workdir /home/ipp/data/work/detrend/VYSOS5.FLAT.PSr.1 -inst VYSOS5 -det_type FLAT -filter r -select_exp_type FLAT -select_filter PSr

• what sort of shutter does VYSOS-5 have? do we need a shutter correction frame? If so, we'll need to get flats with a range of exposure times.
  • It is very obvious from the flat-field residuals that the shutter correction is important. There is a significant blade structure visible in the sequence of dawn flats. That data probably can be used to generate the shutter correction.

• It seems there is a bug in the flat-field normalization: the flat-field is being set to a mean of about 5, rather than 1

• I'd recommend changing the OBSTYPE LIGHT to OBJECT (much more standard).

• I've set the VYSOS5 camera to use only BIAS and FLAT corrections, and no MASK. When we get a bit more familiarity with the camera, we can try a mask and a dark. 2009.01.29 : I've turned off the flat-correction as well until we can fix the flat-field normalization bug.

Science Analysis

• the images can be flipped by the readout software to keep the orientation the same for both east and west of the mount. This is identified in the header by FLIPSTAT having the value "Flip/Mirror"

• to run the science analysis, we need to supply as astrometry model. This can be supplied on the command-line for psastro testing (add the argument -astrommodel (file.asm). However, for pipeline processing, it needs to be available automatically. This can be done by inserting it in the detrend database. I've done this for vysostest using a shell script (/home/ipp/data/detrend/register.astrom.sh), but the only interesting lines in that script are:

dettool -dbname vysostest -register_detrend -filelevel fpa -inst VYSOS5 -telescope VYSOS5 -det_type astrom -use_begin 2008/01/01 dettool -dbname vysostest -register_detrend_imfile -det_id 3 -class_id fpa -uri file:/home/ipp/data/detrend/vysos5.asm dettool -dbname vysostest -updatedetrun -state stop -det_id 3

To do this in another circumstance, it is necessary to know the next detrend db det_id, or to modify the script to grab if from the output of the first command (we don't currently have a standard script to do this step since it is rare).

• I have copied the basic (bright) astrometric and photometric reference catalog to centaurus:/home/ipp/data/catdir.synth.bright.