As I noted, if I were doing the programming, I’d build in a dead band and correct the slope in software. (Already did, in fact.) But trust me when I say that those are some pretty fancy programming concepts for the kids.

The original IBM PC Games Control Adapter (GCA) used a variable resistance for each axis. The variable resistance was used to determine the period of oscillation of a 555 timer chip (I even have the schematics for the card somewhere; they were published in the Technical Reference.) (Ok, so it may have been the quad version of the 555; is that the 558?). If I remember correctly (and it has been about 25 years since I’ve looked at that thing), the 555/558 was wired as a monostable, and you started it off with a write to the GCA port, and then, the timer elements for each axis timed out as you were reading them in a loop.

Oh, one of the nice things about the IBM GCA card is that it had a prototype wirewrap area included on it, which could hold a couple of TTL circuits. It was pretty easy to toss a couple of latches on there, and use the address decoding on the card itself, to make an 8 bit output port.

One of the nice things about this technique (555 monostable timers) is that you can get as much precision as you want with such a system. It’s basically controlled by the software that reads the timing of the 555 chip (and the reference clock, of course).

Some applications included a joystick calibration routine that you could use to calibrate your joystick to reduce drift and define the limits. Another approach would be to use software to declare anything within so many units of the center as being “zero”, and then to scale the action based on a software defined slope from outside the zero range to the limit.

In any case, that should give you a LOT more resolution than trying to use switches, no matter how many positions the switch has, since any practical switch will be limited to well under 120 contacts on each size of zero (while, with the 555 monostable, and a sufficiently fast clock, you should be able to get many thousands of levels of resolution).

For the old timers here, some of us can remember some of the early joysticks, such as on the Atari games console of the 1970s. These weren’t proportional control, but were simply switch contacts for left/right/up/down. No matter how hard you pushed, the output was either on or off. Trust me when I say proportional control is much nicer.

As for the X-ray controller, I make it a habit of not messing with things that can make you glow in the dark. Of course, we all know that there’s not much danger from an x-ray system as long as it’s powered off (and the capacitors discharged). Still, it’s an interesting piece of equipment.

However, something is confusing me. Most x-ray systems don’t allow you to vary the tube voltage/current; most are fixed. This appears to be a rather specialized x-ray system. Ah, ok, here are some links:

http://www.csb.yale.edu/data_collec/yale/rigaku.html
http://www.mines.utah.edu/pyrite/xray/Rigakurun.html

This appears to have been the control panel for an x-ray diffraction instrument.

Dave