The following idea is still a seed and may not work, but it may stimulate further ideas from people much smarter than me. Please, comments and suggestions are requested!! I do not take credit for the idea, as it was implemented in some professional observatories in the 80's.
1) Dome azimuth feedback mechanism - Using a CCD Linear Barcode Scanner with a USB interface (an example of which can be seen at
www.skycomp.com.au/product.aspx?id=50562) to read the azimuth of the dome slit. Print a few barcodes using the PC and stick them to the dome at the required interval. One can even use "barcode fonts" to print out the labels. From the limited understanding I have, some scanners' output is the same as a PC keyboards' input, so if the scanner reads a barcode that "says in barcode", say 25 degrees, then the software reads "25" as if it was typed in on a keyboard. This simplicity appeals to me. I am sure there will be a way to "read" this input in software and compare this input to the telescopes' azimuth that can be read in ASCOM. A simple difference calculation can then determine an appropriate length of time to energise the wiper motor. The same idea as Allan's, but not using a laser and sensor.
Another aspect that is appealing is that there is no requirement for the motor to be in sync with the dome azimuth feedback mechanism - the barcodes give the absolute azimuth. Therefore, if the motor "slips" while driving the dome, it makes no difference as it will just make another correction after it is finished until it is within some error tolerance. This is assuming the dome is not being continuously driven (which complicates things significantly as compensation for telescope altitude will have to be made somewhere). Possibly a "dome correction" can be made every minute or so, depending on the slit width.
2) Motors - In my case I am thinking of using two windscreen wiper motors on either side of the dome. This will reduce the tendency for the dome to go off-centre when the motors are activated. Transmission will be by friction, provided by cannibalized "spring clamps" (Bunnings have a range of large clamps) keeping the motors in contact with the dome's inner ring at the appropriate tension. As already mentioned, slippage should not be a problem, as another correction can be issued until the dome is pointing in the correct direction.
The only PC interface that would be required would be something that can provide a single "on-off" capability to the motors using USB (an inexpensive electronics kit possibly?).
Materials List:
a) USB Barcode scanner
b) wiper motor
c) wiper motor control (power supply, start-up etc) - electronics wizard required
d) USB interface to wiper motor control (on/off) - electronics wizard required
e) Software to
1) Read the dome azimuth output from the USB scanner software package
2) Read the telescope azimuth from ASCOM
3) Compare the dome azimuth to the telescope azimuth and issue an "on" command for a duration dependent on the difference.
4) Repeat until within error tolerance
Hopefully there are others here that would be able to refine the idea, point out issues/problems, or shoot it down (saves time going down a dead-end).
Regards
Daven