Hi digman:
With the WestAmp drivers, the axis limit switches connect to the drivers themselves. When either of the limits of a given axis are tripped, they cut the power to that direction of travel; then the axis comes to a stop, and you can only jog the opposite direction with that axis to come off the limit switch. The WestAmp driver manual has the schematics, and explains the action pretty well. This was an industrial standard at that time.
The analogue drivers are technically obsolete, given the newer digital AC drivers, and brushless AC motors, but if yours are still good, you can have a very useful machine that works. WestAmp has been out of business since the late 90's, but they can still be repaired, and used amps are available on the internet. It I had a big, high tech business with manufacturing deadlines, using old equipment would be a risky endeavor. We are talking home shop, or small business.
The reference switch is also a switch just like the limit, but is positioned to trip slightly before the limit, which is mounted at the right end of the X axis table. When you are referencing, from the Mach3 program control, the machine moves until the ref switch trips, then the machine reverses, and moves very slowly until the "Z" channel of the encoder is seen. The machine stops, and then backs off something like .050 or .100 and stops. The point the Z channel went active, is the machine "Zero". Since the encoder is connected directly to the motor, and the motor is directly connected to the Ball screw, this "ref" point is very accurate and repeatable. You can come back to a work zero position, after the machine has been turned off, and then restarted, to within tenths, .0001 or .0002. This is pretty useful, and is absolutely reliable; at least it has been for my machine, and setup. The "Z" channel on the encoder is only one digital pulse signal per encoder revolution. The controller only see's the reference signal when you are in REF mode; in normal running the reference switch is ignored, and the machine can only see the limit switch.
Each axis has 3 micro switches, 2 limits, and 1 reference.
The spindle is controlled by Mach3. The 24v relay, tripped by the G-code reading an M3, output signal of Mach3, in turn trips the spindle motor AC contactor. My machine has a 5hp AC motor, with a vari-speed pulley. I plan to install a VFD sometime in the future, but have not gotten around to it.
The controller box has inputs, and outputs. In the case of the dspmc, they list each and every pin, and what the corresponding Mach3 I/O pin # is.
My machine originally had a bank of 8 24v relays, that actuated the various things, like the Driver power supply, spindle cw, spindle ccw contactors, coolant motor, mist solenoid, lube pump, etc. I did put in a new relay bank, but it connects to the same original AC wires in the AC cabinet. The original relay bank had some very obscure wires that were very difficult to trace, but I could discern what they controlled by tracing the AC wire side, going back into the AC Mag cabinet. This is where you have to get into the wire looms, and laboriously follow each little wire to make your schematic. I had some factory schematics for the 24v things, but nothing that showed the entire relay<>AC path. You have all the original factory drawings, which will be a big time saver.
The old controller end of things is going to be removed. The new controllers will have schematics showing all the inputs and outputs, so you just have to make the new connections. What you are doing is basically giving the machine a new "brain", which is far better, than the original collection of circuit boards. What used to take an entire circuit board, with 2 or 3 different voltages, and a big 50 pin edge connector, is now on a chip the size of a quarter. The computer will connect to the controller with a single ethernet cable, and a pendant will connect to the computer with a single USB cable. The dspmc controller is a small box, with 5 db25 plugs, an ethernet socket, and a 2 pin connector for the power.
Things really have improved, and made electronics doable by anybody with some time, patience, and a few bucks. In the 1980's CNC controls were proprietary, factory creations, with unique circuit boards that cost thousands of dollars, and failed frequently. Now, nearly the entire electronics in a CNC machine are "off the shelf" items. When you finish you mill, you will have a CNC machine that you can fix yourself, and make as good as you want.
John