let the machine keep track of where the intial part origin goes when rotary axis moves with a macro that contains trig, the rotation centers of the machine setup, and g92s. then just do all the programming based on that single part origin. the gcode is then portable between different machine/rotary setups. sounds like the g68 might work like that. it works well for parts attached anywhere to a single rotary axis, using the rotary axis to index different angled faces. the programming is easy with the exception of remembering to choose an appropriate rapid plane for each face.
--- On Thu, 4/19/12, Stuart Stevenson <stus...@gmail.com> wrote: > From: Stuart Stevenson <stus...@gmail.com> > Subject: Re: [Emc-users] Nonplanar arcs > To: "Enhanced Machine Controller (EMC)" <emc-users@lists.sourceforge.net> > Date: Thursday, April 19, 2012, 9:13 PM > I don't remember the complete syntax > and symbols used but on my fanuc 15m > control G68 sets the rotation angle of one rotary axis. You > can use two G68 > lines to rotate two rotary axes. The regular 2D code then > works at the > angle described by the G68 definitions. It takes some > thought to get the > rotations correct. It takes some thought and attention to > follow the xyz > zero position. G69 cancels the rotations. > On Apr 19, 2012 10:58 PM, "charles green" <xxzzb...@yahoo.com> > wrote: > > > ..suppose you had a five axis millimg setup with the > normal xyz plus > > alpha-beta rotation of the cutter rotation axis about a > shperical center. > > then suppose that to take advantage of these > spindle axes, you wanted to > > mill a planar facet on a part that was tipped at say > five degrees to the x > > and five degrees to the y. the facet contains > pockets with circular > > features that run normal to the tipped face, or > circular bosses with tap > > bores running along their axis. > > > > it would be nice to rotate the entire coordinate system > to align with some > > of the features of interest on a part that is otherwise > already set up and > > positioned in the machine. > > > > another solution would be to accurately rotate the part > about a and b and > > c axes, and just keep track of what that does to the > xyz zero on the part, > > and then all the 2d gcode still applies. this is > probably the more common > > of the five axis styles. > > > > maybe g16 a b c is an arbitrary interpolation plane for > arcs, where abc > > are rotations relative to the abc origin (= the usual > xyz directions)? > > xyzijk would have the usual meanings. > > > > maybe g15 p q r is an arbitrary x'y'z' coordinate > system, where pqr is a > > unit vector relative to the usual xyz unit > vector? xyzijk would be in the > > sense of the rotated x'y'z' coordinate system, and g16 > would act on top of > > this new coordinate system. > > > > maybe g14 selects no preferred plane, and disallows > g2/3 until g16-19 are > > set. maybe it also sets x'y'z' = xyz > coordinates? > > > > > > --- On Thu, 4/19/12, Chris Radek <ch...@timeguy.com> > wrote: > > > > > From: Chris Radek <ch...@timeguy.com> > > > Subject: Re: [Emc-users] Nonplanar arcs > > > To: "Enhanced Machine Controller (EMC)" <emc-users@lists.sourceforge.net > > > > > > Date: Thursday, April 19, 2012, 11:57 AM > > > On Thu, Apr 19, 2012 at 09:45:42PM > > > +0300, Viesturs L??cis wrote: > > > > > > > > Uhhh, You are right, halfcircles. All three > points are > > > on a straight > > > > line, around which the arc can freely rotate. > I guess > > > that this is > > > > special case (is there any other?), > > > > > > That is just the worst problem. Your system > doesn't > > > uniquely identify > > > any arc. For every start, center, end points > there are > > > a pair of arcs > > > that share the points. This is why we have > > > G2/G3. If you don't have > > > a normal vector you can't say which way is > clockwise, so > > > G2/G3 don't > > > make sense. > > > > > > This is also a problem you get when you specify > the > > > arbitrary plane > > > with three points, as was proposed by Ian M. > > > > > > The correct solution is probably to specify the > plane's > > > normal vector. > > > > > > While it's entirely possible to do, I doubt anyone > would > > > ever use this > > > feature if someone did the work to implement it. > > > > > > Chris > > > > > > > > > ------------------------------------------------------------------------------ > > > For Developers, A Lot Can Happen In A Second. > > > Boundary is the first to Know...and Tell You. > > > Monitor Your Applications in Ultra-Fine > Resolution. Try it > > > FREE! > > > http://p.sf.net/sfu/Boundary-d2dvs2 > > > _______________________________________________ > > > Emc-users mailing list > > > Emc-users@lists.sourceforge.net > > > https://lists.sourceforge.net/lists/listinfo/emc-users > > > > > > > > > > ------------------------------------------------------------------------------ > > For Developers, A Lot Can Happen In A Second. > > Boundary is the first to Know...and Tell You. > > Monitor Your Applications in Ultra-Fine Resolution. Try > it FREE! > > http://p.sf.net/sfu/Boundary-d2dvs2 > > _______________________________________________ > > Emc-users mailing list > > Emc-users@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/emc-users > > > ------------------------------------------------------------------------------ > For Developers, A Lot Can Happen In A Second. > Boundary is the first to Know...and Tell You. > Monitor Your Applications in Ultra-Fine Resolution. Try it > FREE! > http://p.sf.net/sfu/Boundary-d2dvs2 > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users > ------------------------------------------------------------------------------ For Developers, A Lot Can Happen In A Second. Boundary is the first to Know...and Tell You. Monitor Your Applications in Ultra-Fine Resolution. Try it FREE! http://p.sf.net/sfu/Boundary-d2dvs2 _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
