The principle of operation of TurPol is illustrated in Figure 2.1. The
diaphragm in the focal plane of the telescope has two apertures, one
passing the star's light plus the sky background, the other one passes
background light equal to that going through the stellar aperture. A
rotating chopper opens and closes alternate apertures, illuminating
the photo-cathode of the photo-multiplier tube consecutively with the
star and background light. A field lens produces an image of the
telescope exit pupil on the photo-cathode, thereby avoiding image
motion due to guiding errors. This is of importance because the
sensitivity of a photo--cathode is generally not uniform over its
surface, and any movement would give rise to spurious intensity
variations.
Figure: Principle of operation of TurPol.
The chopper simultaneously chops the light from two IR LEDs onto two
photo-diodes. This signal is used to synchronise the electronics in
such a way as to accumulate the star plus sky and sky alone signals into
two separate counters. Thus, sky subtracted photometry is done.
By inserting a plane parallel calcite plate into the beam before the
focal plane, polarisation measurements can be made. As can be seen
from Figure 2.1, the calcite plate splits the incoming beam into two
relatively displaced but parallel components, the so--called ordinary
and extra--ordinary rays, which are orthogonally linearly
polarised. The calcite plate is constructed and placed so that the
ordinary and extra--ordinary beams pass centrally through the two
apertures. Each aperture now passes the light from the star and the
sky, and by measuring their relative intensities, after passing
through a wave-plate, the degree and angle of polarisation of the star
light can be determined. In TurPol the intensities are measured at
eight rotational positions of the half wave plate.
One important advantage of this method is that the possible sky
polarisation is cancelled out, albeit at the cost of having to
subtract twice the sky intensity. Since the sky background can be very
highly polarised and variable, this is a small price to
pay. Polarimetry of faint stars in the presence of moonlight is only
possible due to this cancellation method, as is the high precision
( 0.005%) obtained on bright stars.