Re: CNC lathe threading?

[Matt Shaver <mshaver-at-erols.com>]

On Tuesday 11 March 2003 22:22, you wrote:
>    I'm holding in my hand a rod with 1/2" of m12x1 threads cut in 6
> passes.

Two things:

1. I was at NIST a week or so ago, delivering a CNC Sherline mill and rotary 
table (I think the rotary axis bug will be fixed soon now). One of the topics 
that came up in converstion was guys like Barry. Since you can get the EMC 
anonymously, no one has a clue how many EMC users are out there, and every 
once in a while guys like Barry will unexpectedly pop up. They have no prior 
history of posts to the list (that I could find anyway), but instead of 
simple beginner level questions they exhibit a considerable knowledge of the 
EMC's code. In this case, enough understanding to pull of an impressive hack 
like this one! I'm always totally stunned when this happens... I wonder how 
many guys like this are out there that never say anything?

2. I wanted to exhibit rigid tapping at NAMES, but with 6 weeks to go, I don't 
know if there's time. I promised to fit the NIST Sherline with a spindle 
encoder that we could read through a Kulga DRO board, as well as figure out a 
way to reverse the spindle motor. The sequence of operations would be:

a. Position X & Y over the hole to tap.
b. Start the spindle & wait for the speed to stabilize.
c. If the spindle isn't going too fast, sync the Z axis to the spindle encoder 
starting on an index pulse. This will cause the Z axis to move toward the 
workpiece.
d. At some Z depth, cut the spindle power and wait for it to stop rotating.
e. With Z still synced, start the spindle in reverse causing the tap to be 
withdrawn from the hole.

We need to be able to put the tap back in the hole without cutting the threads 
out of tolerance. We also need to be able to do this with regular AC motor 
driven spindles, not servo motors. It may require (later on) the ability to 
brake the spindle in cases of tapping a blind hole, either mechanically, or 
by injecting a little DC current into the windings. I don't really know that 
much about how to achieve this, only that large, multi-horsepower machines 
don't have DC servo spindle motors, and since they can do rigid tapping, it 
must be possible.

Since this is a more difficult problem, my hope is that lathe threading will 
almost fall out of this as a byproduct of developing the rigid tapping 
technology (the exception being the interpreter differences).

Later on, we can figure out how to abstract the encoder counting hardware 
support from the core threading/tapping functions. This would enable 
experimentation with different encoder types/resolutions as well as 
alternative encoder/computer interface schemes.

Matt