OK, I see that you mean now. As they say, a picture is worth a thousand
words. (Though that does not seem to hold true when writing an essay for
English class!?)

I still think naive cam detector is doing exactly what it is suposed to. If
the maximum deviation of an arc from a strait line is within Q they are
collapsed into a single linear move, which is exactly what I see in your
screen dump.

Kurt

On Thu, Feb 16, 2017 at 2:53 PM, Todd Zuercher <[email protected]>
wrote:

> I didn't expect the Naive cam detection to do this.
> https://s2.postimg.org/ezy3wjfjt/bug.png
> Maybe I am reading the docs wrong.
>
> ----- Original Message -----
> From: "Kurt Jacobson" <[email protected]>
> To: "Enhanced Machine Controller (EMC)" <[email protected]>
> Sent: Thursday, February 16, 2017 2:10:26 PM
> Subject: Re: [Emc-users] G41-42 and G64 Bug?
>
> I don't really know any thing about this, but I'd say that is the expected
> behavior based on the description of G64 Path Blending:
>
> "On G2/G3 moves in the G17 (XY) plane when the maximum deviation of an arc
> from a straight line is less than the G64 P- tolerance the arc is broken
> into two lines (from start of arc to midpoint, and from midpoint to end).
> those lines are then subject to the naive cam algorithm for lines.[1]"
>
> It sounds like your Q value is larger than the radius of your compensation
> lead out, so the arc is broken up into two line segments. Since these line
> segments are within Q of a linear move they are collapsed into one linear
> move, and since this linear move is within Q of the original linear move
> the whole path is collapsed in to a single linear move which is treated
> essentially becomes a long compensation lead out move.
>
> Just thinking out loud here, again, I am a complete noob so take this with
> half a grain of salt.
>
> [1] http://linuxcnc.org/docs/html/gcode/g-code.html#gcode:g61-g61.1
>
> Kurt
>
> On Thu, Feb 16, 2017 at 1:32 PM, Todd Zuercher <[email protected]>
> wrote:
>
> > I was playing arround with G41 and G42 in a simulation and I think I
> might
> > have found a bug with the Naive CAM Detection.
> > The shape I was simulating was a semi circle with the start/end point on
> > the flat side and a single G2 arc move connecting the end points.
> > The simulation behaves strangely. With a tool radius of .125" and .25"
> and
> > staight G64 the tool path precisely followed the path (white and red
> lines
> > over lapped). With G64P0.001, the same. The expected behavior is cutting
> a
> > straight line parallel to the programed line offset the tool radius, cut
> an
> > arc around that end point to connect to the offset path of the commanded
> > arc cut. all good right.
> > With G64P0.001Q0.2 there is odd behavior. It moves in a straight line
> from
> > the beginning of the compensated straight line cut to the begining of the
> > compensated arc cut completely ignoring the compensated end point of the
> > straight cut. I know having such a large Q setting is a little odd. I
> also
> > see simular behavior with G64P0.2 (no Q specified, same as Q0.2) only
> with
> > more rounded corners.
> > I seem to see the odd behavior when ever Q is about 2/3 or more than the
> > tool diameter.
> >
> > --
> >
> > ========================================
> >
> > Todd Zuercher
> > mailto:[email protected]
> >
> > ========================================
> > ------------------------------------------------------------
> > ------------------
> > Check out the vibrant tech community on one of the world's most
> > engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> > _______________________________________________
> > Emc-users mailing list
> > [email protected]
> > https://lists.sourceforge.net/lists/listinfo/emc-users
> >
> ------------------------------------------------------------
> ------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> _______________________________________________
> Emc-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
> ------------------------------------------------------------
> ------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> _______________________________________________
> Emc-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to