Had some problems with machining in corners, tool wobbling. Trying to learn more about it I stumbled upon document with constant tool engagement angle. Concept is keep length of tool peripheral which is the length teeths machine material constant. There is a problem this distance increase in corners but it also change if following a curved wall. Anyone no if tool engagement angle should be different for different chip thicknesses? This is of particular interest on outside of curved walls as chip thickness increase in this case but might get a little bit more optimal cutting also on inside.

Following a straight wall is simple. Have figured the mathematics following a curved wall both on inside an outside including the case then tool radius is slightly larger than wall radius. Also wrote a short document about it if anyone is interested. Tool radius slightly larger than wall radius might for example happen starting with a drilled using a helix to make a larger hole and this case there is a minimum tool engagement angle. Also wrote a small program in Octave to calculate tool path for this case if anyone is interested? Should be rather simple to convert to C if needed as "fprintf(...) function is used to write the file.

Keep constant tool angle in inner corners or at least put some kind of upper limit if ordinary arcs are is however harder. If I got right I already figured the gradient then machining into an orthogonal wall have not been able to get it on a form so that calculate arcs for inner corners is possible. Then walls are not always orthogonal and may also be curved. Guess maybe the curved case could be solved by a worst case line once figured out entering a wall at any angle. Manually made some tool to take a short cut in corners and seems work but use computer to calculate tool path preferably as close to optimal is a better option. Anybody have seen or know more about the mathematics behind it?


Nicklas SB Karlsson



_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to