Re: HARDINGE HNC RETROFIT threading

Hi Jon:

Well, the simple answer is that the spindle speed variations aren't
that large or rapid in practice to cause a problem; the acceleration
available to the Z axis usually constrains the process, rather than
the amount of data from the spindle.

I was surprised at this too when I first started experimenting with
CNC threading algorithms, so I can understand the motivation behind
your questions. Sync-up time is just a function of how much the Z
needs to accelerate to pick up the thread. One rev is often enough
from a dead stop, but I often go with 2 or 3 just to be sure at
higher speeds.

For example, on my mini-lathe I use a 90VDC variable speed motor
(not the one it ships with). Although I've always considered
it "squishy" as far as maintaining speed goes - when I measured it I
found it stays within about 5% or so under a healthy cutting load.
The thread is already a few turns in by the time I can hear the
difference, as the spindle and drive inertia tend to damp the
response. This is at 200-300 rpm or so.

About 6 ppr is where I calculated the "empty resolution" limit to
be, and so I have a 6 ppr encoder on there for when I add more
sophistication to the algorithm. Since 1 pulse works well enough for
now, I haven't bothered. Having more pulses is better of course, but
deceptively few are required.

In testing, I found that it's actually not that bad to just sample
the speed once and run down the part without checking the spindle
again at all! Not a great way to do it and the results kind of
sucked after the first 3/4" or so, but it's really not as
temperamental a process as one would think.

Besides, there's no reason to run slowly for threading on a CNC.
Computers have quick enough reflexes.

Dave Kowalczyk
Everett WA
http://www.dakeng.com <-- TurboCNC software

--- In CAD_CAM_EDM_DRO@y..., Jon Elson <elson@p...> wrote:
{snip}
>
> One pulse per rev? How long does it take for the Z axis to sync up
to
> the spindle? What about speed variations when running at low speed
with
> high spindle load? This is a common ocurrence on the last few
passes of
> a coarse thread. How quickly can the Z axis adjust when the
spindle speed
> changes?
>
> Jon