If the average values obtained for standard stars deviates much from
those in the catalogues, the results obtained from polred, can be
corrected with the programme addcor, and then kokpol can be
run again. The (usually small) corrections for bf addcor are read
from a file cor.add which should contain:
1.4 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
where the first line is the added correction to the zero point of the
instrumental position angle, the second line the wavelength dependent
position angle corrections relative to the V band (=0), and the last
two lines contain respectively the five instrumental polarisation
values for p
and p
, in the order UBVRI. These corrections may
also be input by hand prompted by the programme addcor as it runs.
The constants used by polred can also be updated by the values
used by addcor. This has the advantage that they do not have to be
filled in every time the programmes are run. For observing runs of up
to about two weeks this is OK.
The zero point of the position angle and its wavelength dependent
components are best determined from observations of standard stars
with high polarisation. The corrections given in the constants
file are added to the observed values.
The instrumental polarisation values are determined from observations
of unpolarised standard stars. These data are to be reduced using a
special version of kokpol called kokinst which takes the
rotation from equatorial angle to telescope optics angle into
account. This can be necessary because in alt--az telescopes the angle
between the polarimeter and the telescope changes during the
observations, which when the telescope produces its own polarisation,
will have an effect on the measured instrumental polarisation. At the
NOT this telescope polarisation has been found to be negligibly small
and zeroes can be used in the cor.add file.