Re: Threading under EMC

[Chris Stratton <stratton-at-mdc.net>]

Ray wrote:

> I'm not going to be much help with the details of this but I've got three
> lathes out in the garage that will need it soon.  I have had some
> experience with machines where a servo is not properly responding to rigid
> tapping or threading and am of the conclusion that both position and
> velocity of Z need to be coupled directly to changes in the position of the
> spindle for this to work satisfactorily.

I've been looking at the code more in depth and understand a bit more
of it now, unfortunately, I'm seeing two major modes of operation both
of which will probably need support:

For mills doing rigid tapping, you have to servo the spindle since you
have to be able to back out of the hole.  So you want to keep running
a servo loop for at least the spindle axis.  A bit of recoding could
replace the trajectory-planned and cubic-interpolated Z axis desired
positions with ones derived directly from the spindle rotation.

For lathes doing single point threading, I'd really like to support a
mode where the spindle is not servo'd, but simply put in a slow enough
speed for the Z axis to keep up.  This requires that we stop trying to
control the state of the machine, and instead become willing to
respond to what it is doing beyond our control.  Trajectory planning
and interpolation would have to be suspended for the Z axis and
instead we would need to calculate Z positions based on sampled
spindle rotation at the full servo rate.  For taper threads we might
need to run a sort of trajectory calculation driven by spindle
rotation rather than time intervals.

For either mode, we need to synch up the Z position to the initial
spindle rotation.  If we simply start throwing spindle-locked Z
positions into the PID loop we may get some nasty accelerations out.
Best way to synch up would probably be to calculate an acceleration
profile that has us catch up with the spindle in speed and position at
the same time. 

I hear PVC pipe is nice for practicing manual single point threading...

Chris

> I set up a repo Mazak lathe recently that had a program in it for hard
> tapping a 1/4-28 in aluminum.  They ran with spindle speed set to 4000
> rpm.  We ran a few trials, being skeptical and the spindle speed just
> reached 4k when the decel took over.  In this case, there was no constant Z
> speed.
> 
> I agree that rigid threading may require us to increase the precision of
> our axis tuning but I believe that 
> 
> On Sun, 27 May 2001, Chris wrote:
> > Am I right in thinking that rigid threading modes are not currently
> > supported in EMC?  As far as I can tell, there is a cycle for tapping 
> > (I'm really intersted in single point threading on the lathe) but
> > it depends on a canonical function "SET_SPEED_FEED_SYNCH" which
> > doesn't do anything yet?
> 
> Yea I've read something about tapping but with no speed feedback from the
> spindle this is worse than buying a lottery ticket.
> 
> <s> 
> > Alternately, the interpolater could be altered to ignore the commanded
> > feedrate and simply generate points based on the actual spindle
> > rotation.  The problem is that EMC's PID servo loop as most actually
> > use it is really a P & Feedforward algorithm, and we can't know
> > exactly where the spindle will be several trajectory (or is it servo?)
> > cycles from now.  Sure, we can initially plug in numbers based on the
> > projected spindle rotation (it shouldn't change much) so the FF
> > calculations can work and then use the actual spindle position for the
> > P calculation - but will the loop close following errors
> > satisfactorily this way?
> <s>
> 
> My thinking is that this is the way both tapping and threading ought to
> be done.  
> 
> Thanks for your interest in this step forward with the EMC.  Matt and
> several of us at NAMES were working on priorities for advance and this was
> near the top  Let me know if there is anything that I can do to help.  I'd
> be happy to set up the feedback and break a few taps with my Grizzly to test
> this code out.
> 
> Ray
> 
> 
> 


-- 
Christopher C. Stratton, stratton-at-mdc.net
Instrument Maker, Horn Player & Engineer
22 Adrian Street, Somerville, MA 02143
http://www.mdc.net/~stratton
NEW PHONE NUMBER: (617) 628-1062 home, 253-2606 MIT