i ran across the degrees axis feedrate bug also. the problem is that gcode treats the abc axes as another set of linear axes, but that is not really what you mean when trying to coordinate a movement in a combined linear/rotary motion. it's a little like treating helical feedrate as if it was with respect to the plane of arc interpolation only, so that low helix angles result in hugely scaled feedrates. in the case of a rotation axis, disregard for the radius from the center of rotation ruins what would ideally be locally linear scales for motion.
i see only two approaches: the rotation axes are purely for indexing between various angular offsets without any coordination of other movements (i.e. stay clear of spinning mechanism), or incorporate the absolute cartesian positions of the rotation axes into a calculation of resulting movements. the latter has got to be the origin of the 'teach in jog' programming mode on some controllers. at any rate, perhaps at an unknown rate, is it correct to treat the machine unit of linear movement as just equal to one degree of angular movement by definition? --- On Thu, 6/7/12, andy pugh <[email protected]> wrote: > From: andy pugh <[email protected]> > Subject: Re: [Emc-users] cylindrical coordinate kinematics > To: "Enhanced Machine Controller (EMC)" <[email protected]> > Date: Thursday, June 7, 2012, 2:59 AM > On 7 June 2012 05:11, Ralph Stirling > <[email protected]> > wrote: > > I'm putting together a little 3-axis system with a > horizontal rotary > > table (like a C axis), an X axis, and a Z axis. I'd > like to program > > it as XYZ, though, so I need some kinematics to convert > XYZ to XZC > > You might not need kinematics. For a constant diameter it is > simply a > case of setting the C-axis scale to suit the radius. You can > then wire > the Y-axis commands directly to the rotary actuator and run > G-code in > YXZ space. > > If the radius changes during a job then it isn't quite than > simple and > you probably do need a simple kinematics module. That will > require the > Z-axis to be accurately homed to absolute position relative > to the > axis of rotation. > > One wrinkle will be the axis speed limits, as your Y-axis > physical > speed limit is much higher at large Z than at small Z. > > -- > atp > If you can't fix it, you don't own it. > http://www.ifixit.com/Manifesto > > ------------------------------------------------------------------------------ > Live Security Virtual Conference > Exclusive live event will cover all the ways today's > security and > threat landscape has changed and how IT managers can > respond. Discussions > will include endpoint security, mobile security and the > latest in malware > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Emc-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-users > ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
