WHT rotator tracking
The rotators on which the Cassegrain and prime-focus instruments are mounted,
and the Nasmyth derotators, are expected to track accurately enough
to ensure < 0.2 arcsec movement of star images at the edge of the field of view.
At Cassegrain and at Nasmyth, and for prime-focus imaging, there is no
guiding in rotation (this is provided only for AF2 at prime focus).
A series of 20 300-sec ACAM images taken over a period of 2 hours
in on 17 February 2016 give a useful indication of rotator tracking
accuracy at Cassegrain.
During this period, the positional angle (PA) of the rotator varies from
12 to 48 deg, at speeds up to 3 deg per minute.
Throughout this period, any exposure-to-exposure rotation of the field of view
relative to the detector changed the positions of stars at the edge of the
field of view, relative to those in the centre, by < 0.05 arcsec rms.
ACAM has an 8-arcmin field of view, so this
corresponds to a physical rotation error < 0.05 / 240 = 2e-4 radians,
i.e. < 0.01 deg rms.
While a rotator may track the sky well, the instruments mounted on it
may flex as they are rotated relative to the gravity vector,
often resulting in net displacements of the detector relative to the
light path. Sometimes, this flexure is continuous, sometimes it's
a one-off flop as the instrument passes through a certain orientation.
In the above data, the measured positions of the stars jump by ~ 0.3
arcsec as the Cassegrain mount PA passes through 90 deg.
Such effects are often seen when observing near the meridian,
where the rate of change of Cassegrain rotator PA is fastest
(for an alt-az telescope).