Stuart, Hmm. This doesn't sound right, and I haven't noticed that behavior ( not that I'm saying it doesn't happen). You can't just say...
G1 G91 X.5; G1 G91 Y.5; With G41 / G42 turned on? I disagree on your machining strategy, but agree that if the above generates an error, then the behavior is incorrect. The only place size of radius relative to tool should matter to the controller is on entry / exit of G41 / G42 compensation ( lead in, lead out ) for acceptable reasons. This is true on any controller I've used and is usually a PITA until you learn how to make that particular controller happy. One way to get around it on pockets is to some dramatic compensation move above the part, then spiral in to the center. Jared On Wed, Apr 15, 2020, 1:42 PM Stuart Stevenson <[email protected]> wrote: > Jared, > > A milled pocket will have a fillet in the corners at least equal to the > radius of the cutter used but that does not make it a requirement to > describe a fillet to be able to cut it. I like to use G41/G42 during even > the roughing of a pocket. When I rough a pocket using cutter comp in > LinuxCNC I must describe a fillet for EVERY change of cutter direction. > This is not necessary for efficient roughing and is a pita. > > Regards > Stuart > > On Wed, Apr 15, 2020, 11:24 AM Jared McLaughlin < > [email protected]> wrote: > > > I know I'm jumping in the middle of the conversation, but I can't see > > what other behavior would be desired. > > > > On Wed, Apr 15, 2020 at 11:30 AM Stuart Stevenson <[email protected]> > > wrote: > > > > > > I have tried, without success, to convince a developer to add a switch > to > > > the config file to enable/disable the "feature". I don't like this > > > behavior. I would always run with the feature disabled. > > > > > > On Wed, Apr 15, 2020, 8:26 AM Reinhard <[email protected] > > > > > wrote: > > > > > > > Greetings, > > > > > > > > On Mittwoch, 15. April 2020, 15:00:38 CEST Stuart Stevenson wrote: > > > > > It sounds to me like you will run into a problem with G41/G42 ... > > > > > When using G41/G42, if the tool radius (plus any diameter/radius > > > > > modification for roughing) is not smaller than the programmed > radius > > of > > > > the > > > > > corner LinuxCNC will give you an error code and will not run. > > > > > > > > Well, that is correct and expected behaviour. > > > > > > > > Of cause, if faking tool properties, linuxcnc should use the > resulting > > > > properties to (re-)calculate the new toolpath. If that would not be > the > > > > case, > > > > the support for faking tool properties would be useless. > > > > > > > > But tests with G43.1 gimme the hope, that linuxcnc already takes the > > fake > > > > values for path calculations. > > > > I will test that, when I have a new dev-box up and running. > > > > > > > > Reinhard > > > > > > > > > > > > > > > > > > > > > > > > > > > > _______________________________________________ > > > > Emc-developers mailing list > > > > [email protected] > > > > https://lists.sourceforge.net/lists/listinfo/emc-developers > > > > > > > > > > _______________________________________________ > > > Emc-developers mailing list > > > [email protected] > > > https://lists.sourceforge.net/lists/listinfo/emc-developers > > > > > > _______________________________________________ > > Emc-developers mailing list > > [email protected] > > https://lists.sourceforge.net/lists/listinfo/emc-developers > > > > _______________________________________________ > Emc-developers mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-developers > _______________________________________________ Emc-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-developers
