Replace change gears with a stepper?

The Electronic Lead Screw is what you are thinking about. Not full CNC but does tapers, threads, etc

I have that same Wabeco D6000 lathe (Mine is a D6000e) and tried to buy a kit but it is no longer made. I would buy a kit is something used or new old stock turned up somewhere..

 
Last edited:
Cadillac STS said:
The Electronic Lead Screw is what you are thinking about. Not full CNC but does tapers, threads, etc

Click to expand...

I was seriously considering doing this to my little South Bend 8x18 when I had it. I was to the point where I had the parts list identified, a friend who was going to do the circuit board layout, and a borrowed in-circuit emulator for debugging my code.

That's as far as I got with it.

Now the little SB is gone and it's replacement has a Norton gearbox on it. It was a good idea while it lasted though. :)
 
Could just put on a simple stepper motor with a driver and a knob for speed control to conveniently adjust the speed of the axis. Pull off one gear to disengage the gear drive but also be able to replace the gear and do the gear change for threading. The stepper setup would free wheel when turned off so the lathe could work as normal too.

You would need to sense the position of the spindle for threading and could add that on with another control board later.
 
Cadillac STS said:
Could just put on a simple stepper motor with a driver and a knob for speed control to conveniently adjust the speed of the axis. Pull off one gear to disengage the gear drive but also be able to replace the gear and do the gear change for threading. The stepper setup would free wheel when turned off so the lathe could work as normal too.

You would need to sense the position of the spindle for threading and could add that on with another control board later.
Click to expand...


I think that's the route I would start with. It's a good proof of concept that just needs finer timing control to make threads later. Well, "just" might be over simplifying a fair bit.. :)

I was looking at the existing sensor on the spindle last night and noticed it's got 2 magnets, offset about 90 degrees, so I should be able to sense direction, speed, and orientation with some accuracy. It's an interesting experiment, but I would want to make sure I can use the machine as designed without too much hassle whatever way I go with it. It makes mounting motors and such potentially more complicated, but it makes it more flexible.
 
ttabbal said:
The plan would be to use a microcontroller to control the stepper and watch the spindle speed to maintain sync.
Click to expand...

Oh, the synchronization of speeds is only part of it. The spindle motor (AC motor) has known speed, but only
on the average. Any variation of the torque applied by the cutting tool becomes a phase error in the motor position,
and the stepper won't follow that kind of variation unless you monitor the rotation with some kind of spindle
rotary position sensor with very fine-grained resolution. Otherwise, every other
motor in the shop will telegraph to your workpiece over the AC power lines.

Oddly, threading of a uniform shaft is one application where it MIGHT work, because the cut is never interrupted
and occurs on a well-centered workpiece. The problems, though, would include a washboard-road kind of
resonance, where a cut pass makes minor corrugations due to steps by the stepper motor, and each subsequent
pass makes the corrugations larger... so, it's hard to predict success. Steppers move in steps; that's
going to encourage chatter, and not the social kind.

Servomotor spindle drive and servomotor carriage drive might work, OR you could use your regular AC
motor to drive the spindle, and gear a light set of gears (a clockwork-sized gearbox) into a synchro
generator, and drive the carriage with a synchromotor. The carriage screw is relatively slow moving,
and coarse, but a modified carriage screw (finer pitch) could be fitted to improve the
accuracy of such a scheme, and the efficiency of the motor.
 
Whitmore, it doesn't require servos although if you have the cash then go for it. This threading 'problem' has been solved many times already, including at least 2-3 commercial units. They all work with steppers, half-step microstepping. Encoders outputting 400-1600 pulses per rev are inexpensive as well.
The German controller that I linked to above is 159 euros for a lot of functionality, including what the original poster is looking to do. It will work with just the Z axis or both Z and X. It requires minimum of 400ppr encoder on the spindle, AMT-103's are less than US$30. The recommended stepper drivers can do 200-51200 steps/rev although anything above 800 is overkill and also reduces your movement speed.

Hacking your own software might be fun, http://www.airspayce.com/mikem/arduino/AccelStepper/supports acceleration and deceleration . Personally I am no longer interested in programming so am saving up for the ELSII to put on my KC1022 (King Canada's G0602 version).
 
Cadillac STS,

That's what I'm doing with my project. I've taken it a step further in that I measure the spindle speed and calculate the proper stepper speed so the feed rate is calibrated in thous/Turn (between .001/T and .010/T).

While it's certain that others have solved the threading problem, as Whitmore points out, it's not simple. That's why you pay the big bucks. Steppers can't change speeds instantaneously. They must accelerate or decelerate under software control. Achieving synchronization with the spindle itself is fairly straightforward, but maintaining sync when factoring in backlash and acceleration is a challenge.

For myself, I'll be happy with not having to listen to the gears whining and wearing out for normal turning. I don't thread very often and when I do I use a hand wheel on the spindle.
 
gzoerner said:
Cadillac STS,

That's what I'm doing with my project. I've taken it a step further in that I measure the spindle speed and calculate the proper stepper speed so the feed rate is calibrated in thous/Turn (between .001/T and .010/T).
Click to expand...

The sensor and drive system should be independent of speed and directly link rotational position of the spindle to that of the lead screw, that way the stepper speed is always correct and variations in speed, including hand winding the spindle, will not result in any errors.
 
You must log in or register to reply here.
Back
Top