Polar mount, so one axis is declination and one hour angle. Coverage in dec: from the N pole to -30 or so (sometimes therefore looking at the ground). Hour angle -6h to +6h with about 1 minute of time extra (i.e. +-90 degrees).
Each axis has "fast" and "slow" in each direction. Exact speeds TBA, but "fast" is about 15 deg/min, (slow: 1deg/min ?) Hour angle also has "track" (in +ve ha only!) which is at the sidereal rate. Disengaged by clutch.
HA brake is automatic. Dec brake has to be released first (takes a few seconds: big hydraulic thing).
These are controlled by a cabinet full of relays, and I see no reason why we can't get a suitable interface into that. It includes all sorts of obvious interlocks (can't go in more than one direction at more than one speed on each axis ...).
There is an interlock: a "brake off" signal (indicating that the brake is really off, not just that there is a request to take it off) permits the declination motor to start. I think that putting the brake on will automatically turn off the motor as well - though the usual practice would be to turn the motor off first and let the drive slow down by itself.
There is actually a built-in delay, 5 sec or so, between stopping the fast/slow ha slew and starting the track motor.
Two independent sets. The first one permits one to back out; the second can only be overridden by human intervention. In the past these have proved reliable, and Guy cannot recall when the second limit was last tripped.
While we could think of a modern variable-speed controller/motor, we would be limited by funds and (more seriously) manpower to implement it. The disadvantage of sticking to the present scheme is that it makes tracking objects like the Moon a bit more tricky, since they do not move at the sidereal rate.
Both axes may turn independently. There is no impact on safety or significant effect on accuracy.
The system we have is a multipole resolver system, analogue readouts. It is more seriously in need of replacement than the drive system: it relies on transmitters/resolvers with slip-rings which really need a week of dedicated cleaning every year, and this has not been done for a while. I do not think it useful to try to keep it going if we can find an alternative.
On the Ryle telescope we use 16-bit encoders, mostly rather old; I have just bought a set of replacements which urgently need installing. There will therefore be some old encoders which could be fitted, assuming that there is a suitable mechanical mounting. (16 bits = 19 arcsec resolution which is more than enough.) At least this will have a sensible interface for a computer.
Provide a 16-bit parallel interface (in absolute binary) Make provision for a zero point, since you could never install them on the telescope in exactly the correct orientation.
Please don't link to this page. It's in transit to ...
Serial Test utility
Serial Command Format
Scope drive Emulation
'OI' Controller - Top level Schematic