Likewise only a thought experiment from this end.

I favour using an old PC running Linux and whatever I/O cards are
available for the purpose. If you network the device into the ROW you
will have time of day and if you need the accuracy, the actual sun
motion numbers.

You can also use a old DOS version as that allows direct access to ports
using debug. (Finally found a use for a 286! I use to have one moving a
stepper motor ez-al array some years ago. Used 3 bits each axis from the
printer port)

From memory most rotators have a pot feedback that determines where it
is pointed. This means you need some kind of analogue input to the PC. A
joy stick port maybe? Don't forget that a sound card input could also be
used once you remove the AC coupling capacitor to its A/D.

So after calibrating the system for azimuth vs input volts you have a
lookup table that you can base your command on. You can use serial or
parallel port pins to actuate the motor.

Problem of course is that the PC uses a considerable amount of the
energy gained from the array just to run itself! You could of course
automatically sleep and wake it every 30 minutes or so!

Another possibility - Your rotator controller either has a direct
feedback to a position knob or a display and CW/CCW switch? Should be
pretty easy to use the feedback knob/display meter volts/comparator to
actuate the drive. Just create an analogue lookup table with a simple
4-8 bit TTL/CMOS counter and a stack of resistors and you are away! It
would even reset itself for the start of the new day. Only real problem
is where you source the counter clock from.

Cheers Bob VK2YQA

JERD wrote:
I am wondering if anyone has managed to convert/use a standard antenna
rotator for tracking purposes? I know that very expensive antenna rotators
are available for this use but hey there is no fun or challenge in that!