Imagine that your rotator is reset by you each sunny morning, pointing
east
at the rising sun. You flip a TRANSFER switch and westward rotator
power
is then applied through the contacts of a normally-closed (NC) relay . At
that moment, however, the relay is held open with coil current provided by
the amplified signal of a photosensor which is collocated with the
collector
panels. It is aimed in the same direction as the panels, but it has
blinders on it so it has only a narrow field of view in the east-west
direction. Thus, the photosensor is illuminated because it is pointing in
the direction of the sun. First thing in the morning, with the panels and
the photosensor pointing at the sun, nothing happens when you throw the
TRANSFER switch.

But wait ...
As the sun rises, it gradually moves to a point where it is NOT
illuminating
the photosensor. The relay drops and energizes the rotator. The rotator
moves but an instant later, the whole rig is again aimed at the sun, which
illuminates the photosensor. The relay re-energizes, removing power from
the rotator and movement stops. This process might repeat ten or twenty
times a day, depending on the angle of view of the photosensor.

Surely it is much simpler than that, as long as you are only interested in
azimuth and not elevation. The trouble with photo-sensors and the like is
that there will be a problem when the sun is behind cloud, and that will
cause all sorts of 'hunting' problems.

The good thing about the sun is that you always know where it will be at any
time of the day, and the rate at which it moves, so all you have to do is
arrange a mechanism that rotates by 15 degrees an hour. If you are only
pointing solar panels the you could step in at say 15 or 30 minute
intervals, rather than have it continuously tracking.

73
Jeff