[wiki:WikiStart Up to main page]

Instructions for operating the Parkes continuum polarimetry software.

The data are acquired from the pulsar Digital Filter Bank correlator in
"rpfits" format and are written to the area /DATA/PKCCC2_1/corr/dat/.

== Sequence of Operations ==
Three files are necessary before the process can begin.

 1. freqtab.dat lists the centre frequencies (in MHz) of channels in the final output data.
 1. feed.dat is used to transform raw data to Stokes vectors (in the telescope frame). It defines the transformation in terms of reordering and sign changes.
 1. 3c138Flux.dat tabulates the frequency dependence of the flux and polarization characteristics of the polarized calibrator (3c138, B0637-75, ...).

These files should be in the working directory.  Copies of the first two will be assigned to each data file being processed.
The 3c138Flux.dat (or 0637Flux.dat) file will remain in the working directory and is found by the srcnam call in psolve.  

=== Reduce the bandpass calibrator data ===
 1. Load the rpfits file into "parkespol" format.
{{{
pload 2006-11-20_0915_1934-638DEC1.rpf
#
# This will create a directory 2006-11-20_0915_1934-638DEC1/ and 
# with contents
# -rw-r--r--   1 mcc381  atculg  2355200 Feb  1 22:23 data.rpf
# -rw-r--r--   1 mcc381  atculg      204 Feb  2 11:06 feed.dat
# -rw-r--r--   1 mcc381  atculg      646 Feb  2 11:06 freqtab.dat
}}}
The original rpfits file is maintained so you will have two copies of the data.  

 1. Perform the bandpass calibration
{{{
bpass 2006-11-20_0915_1934-638DEC1
}}}

=== Reduce the phase calibrator data ===
 1. Load the rpfits file
{{{
pload 2006-11-20_1401_3c138_Cal_DEC.rpf
}}}
 1. Copy the bandpass tables
{{{
gcopy 2006-11-20_1401_3c138_Cal_DEC 2006-11-20_0915_1934-638DEC1
}}}
 1. Extract the phase vs parallactic angle information from scans of the polarised calibrator
{{{
polcalr 2006-11-20_1401_3c138_Cal_DEC
}}}
 1. Determine polarisation parameters (leakages and phase difference between the signals from the two polarization channels) of the system from the phase data
{{{
psolve 2006-11-20_1401_3c138_Cal_DEC 3c138 /xs
}}}
To get a consistent pair of output files, the procedure needs to be executed twice. An example follows.

{{{
   polcalr 2006-11-20_1401_3c138_Cal_DEC
   psolve 2006-11-20_1401_3c138_Cal_DEC 3c138 /xs
   polcalr 2006-11-20_1401_3c138_Cal_DEC 3c138
   psolve 2006-11-20_1401_3c138_Cal_DEC 3c138 /xs

}}}

Be careful not to perform this procedure more than twice.  If you need to back up in the process remove the scan.dat, phasefit.dat and polparams.dat files from within the calibrator directory.

=== Reduce the astronomical data ===
 1. Load the rpfits file (called source.rpf in this example).
{{{
pload source.rpf
}}}
 1. Copy the bandpass tables and polarisation corrections
{{{
gcopy source 2006-11-20_0915_1934-638DEC1 2006-11-20_1401_3c138_Cal_DEC
}}}
 1. Apply the calibrations and write results as a single-dish FITS file called output.sdfits in this example.
{{{
calscans source output.sdfits
}}}
 1. Alternately, you can supply a list of input files and write out a single sdfits file.
{{{
calscan.pl list ouput.sdfits
}}}
where list is a text file containing a list of the data files to be calibrated.



If calscans is executed successively with the same outfile named it
appends data from each execution into that file.

What calscans does: ''(check that this desription is correct) ''
 1. Reads the rpfits file;
 2. apply the bandpass calibration;
 3. transform to Stokes vectors;
 4. correct for instrumental phase;
 5. reorder the DFB output bins;
 6. properly label scan data with interpolated sky positions;
 7. average into final frequency channels;
 8. apply leakage calibration (if file polparams.dat is present);
 9. write out in sdfits format.

What calscans does not do (but could with some small software changes):
1. Subtract baselines scan data;
2. rotate Q,U from telescope to sky coordinates.