Re: NAN Bug with circular moves

[Jon Elson <jmelson-at-artsci.wustl.edu>]

Ian Wright wrote:

> Hi,
>
> Thanks for the comments Jon, I realised after I sent my last post that I
> obviously loused things up by adding in to the program line a 'z' component
> but no 'k' component. As I said I am still very much a tyro at this and I'd
> forgotten about 'k' so, presumably, the line I fed the program was telling
> the cutter to start from somewhere it wasn't and, by inference of 'k', go to
> where it was. As I see it though, this is still a problem as obviously EMC
> is still a bit light on error checking and I would have expected it to tell
> me I couldn't do this and shut down the machine safely.  I think perhaps in
> my case, the final machine should be fitted with an appliance which slaps me
> around the face with a wet kipper if I try to enter a silly command!!

Were you expecting a 3-axis curved move, or a helical move, with the
Z moving at a steady rate?  I think helical moves are the highest function
defined for EMC so far.  You set the plane for the circular arc with
G17, G18 or G19.  If you are in G17, for circular arcs in XY plane,
then if you specify a Z coordinate, that axis will move linearly.  The
tool will describe a helix around the Z axis.

I don't believe that EMC will accept a command with all 3 IJK
words in it.  There are controls that handle 3-axis circular moves,
but that is a pretty wierd thing to program.

Jon