Re: EMC Bug???

["Les Watts" <leswatts-at-alltel.net>]

Jon, I have been running a lot of contouring carved sign code
with lots of g2 and g3 moves and have not noted any problems.
In fact some of the code generated by the cam program has
more short g1 point to point moves than I would like... it can
get a little choppy. Of course I am running at high speeds and
accelerations.

I am starting to yearn for a post that converts most motion to
circular interpolation so I can run even faster. I am also looking
closely at the cubic spline part of the code... I would love to have
that smoothing more EXCEPT at certain hard corners. It might
if I run the trajectory planner faster. It is now at 10 ms with
servo update at 500 microseconds... 20:1.

But if you have some code that demonstrates a quirk-
yes I would like to run it and see.

Les
Leslie Watts
L M Watts Furniture
Tiger, Georgia USA
http://www.alltel.net/~leswatts/wattsfurniturewp.html
engineering page:
http://www.alltel.net/~leswatts/shop.html

----- Original Message -----
From: "Jon Elson" <elson-at-pico-systems.com>
To: "Multiple recipients of list" <emc-at-nist.gov>
Sent: Thursday, August 29, 2002 2:03 AM
Subject: Re: EMC Bug???


>
> Matt Shaver wrote:
>
> > Yep, this is a known bug and should (I hope...) be fixed in the CVS
version,
> > or for that matter, the feb2002 version on Sourceforge. Fred Proctor,
Ray, &
> > I looked at this exact bug at N.A.M.E.S. in April and Fred had a fix for
it
> > (which I really hope has gotten into the CVS by now!).
> >
> > I'll try to verify whether it's fixed now & hopefully you can update
your
> > system.
>
> Speaking of bugs, has anyone looked into the quirk in circular
interpolation?
> I ran some programs that run fine on my 20-Dec-1999 version (and I'm
pretty
> sure on my 15-Mar-2000 version, too).  But, these programs, which contain
> simple contiguous 90 degree circular arc segments have a velocity hitch at
> the point where it progresses to the next arc segment.  Velocity seems to
drop
> to 50% or less of the velocity it runs the arc at.  It appears as though
it is
> decelerating much more than is required, quite possibly all the way to
zero,
> as if it were in exact stop mode, which it was not.
>
> This is on the 2-12 BDI version.  If someone needs the program, I can
provide
> a program that demonstrates the quirk.
>
> Thanks,
>
> Jon
>
>