The ratio is approximately 8:1. There is a belt that takes power off the driven pulley right above and coaxial to the spindle. This is connected to a step pulley on the 1J head or the driven pair of the varispeed pulley of the 2J. For high-speed range, this pulley directly drives the spindle through a toothed clutch between the pulley and the assembly below it, which contains the female spline that drives the spindle. For low-speed range, this clutch is disengaged by slightly raising the pulley assembly through a cam ring. The belt drives a pinion at the rear of the belt housing, roughly coaxial to the motor shaft. The pinion has a shifter fork and is raised to engage the bull gear. the bull gear has the internal splines that drive the spindle. Note that the spindle will run the opposite direction when in back-gear mode.
So, to shift to low-speed mode, the pulley is lifted by the cam ring to disengage the clutch, and the pinion is rased to engage the bull gear.
The Vari-Speed head works similiarly except that the bull gear does the shifting. The gear has clutch teeth on its top and for high range it is forced upward by springs to engage the clutch teeth on the end of the pulley shaft. To engage low range the gear is forced downwards to disengage the clutch and to mesh with the pinion. The pinion is driven from the pulley shaft at reduced speed through the cogged timing belt.
I'm thinking about fitting a tachometer to the machine and that the easiest way is to monitor the speed of the belt driven pulley (not the v pulleys). This I think is permanently rotating. I then divide by whatever when in low speed.
I put a gear sensor in the head, and it picks up the bull gear's teeth. The bull gear has 81 teeth, IIRC. It works quite nicely. I used a commercial ratemeter made under the Durant label. But, most of the modern ratemeters made by Red Lion and similar industrial control makers can be programmed for # of teeth on the gear that is being sensed.
This is actually harder than it sounds. The spindle itself is WAY deep in the interior, surrounded by things that either slide up and down or rotate at different speeds. The most exposed part that turns always at spindle speed is the bull gear. On the 2J there is a complication, the bull gear slides up and down. So, you might need two sensors, one in the slow position, one in the fast. The entire area around the driven pulley is filled up with that pulley, and the spindle slides up and down 5 inches. So, I can't see how you'd get a hall sensor between the pulley ID and the spindle OD, There's REALLY little room there. If you could run the sensor on a wire all the way down inside the quill, that might work. But, then what do you attach it to? You don't want to be drilling HOLES in the quill! And, I think the wire would get pinched or caught in the works when you retract the quill.
But, maybe taking the speed off the big toothed pulley and switching ratios when the speed selector is shifted is a good way to do it.
The hole in the top of the spindle that the drawbolt goes into. Poor communications on my part. Sorry. I guess I should have said "hole in the top of the spindle the draw bolt goes into"
Generally..most drawbolts are a bit undersized in OD..and if you took a thin piece of tubing and turned it od/id to hammer into the hole..you could hook up a hall effect sensor to sense something on the tube as it spins. Just not so tall that you cant get to the draw bolt..which of course needs to pass through the tube.
"A prudent man foresees the difficulties ahead and prepares for them; the simpleton goes blindly on and suffers the consequences."
I'm minded to use two gear-tooth-detecting hall sensors monitoring the bull gear and somehow 'OR'ing these into the input of counter.
Given time, I'd also like to use a proximity sensor close to either the top bearing or the big pulley for feedback to an inverter. I think then the G code for lo-speed would ask for a spindle speed 8.3x more than 'usual'.
What do you think?
Also does the bull gear run in an oil bath? Could too much oil affect the hall sensors?
At least on my CNC designed J2 head (quill is surrounded by a hollow ball screw), when the quill goes down, the top of the spindle disappears into the housing, so the tube would have to be at least as long as the quill feed, and the drawbar would have to be lengthened accordingly.
You might check whether the top of the spindle stays put when you feed your quill down.
On the 2J head the pinion gear is stationary and turns all the time. A hall effect sensor there should work. You would have to change the pulses/rev factor for high and low range though. That might be simpler than sensing the movable bull gear, although the bull gear always turns at spindle speed