Re: [CAD_CAM_EDM_DRO] Re: Help needed with G code

Mike Chaney wrote:

>David...
>
>You're right! It does sound funny, but I'll give it a go as soon as I've had my
>breakfast. It does make sense in that, as you point out, all the information
>for a curve is contained in just one parameter of the end point and either the
>centre coordinates or the radius. If it works for YZ, it would be interesting
>to see if this can be done in the XY plane, where I now have enough experience
>to program with confidence.
>
>I'd like to use R rather than I and J but it appears that it is not supported by
>CNCPro. Pity.
>
>Am I correct in thinking that, in the YZ plane, I relates to the Y axis and J to
>Z - and that K is only used in true 3D systems?
>
>

This is absolutely wrong in the standard RS-274D language, although any
CNC control program can make up their own version of G-code, and many
do diverge from the published standard in small and sometimes large ways.

The standard definition for arc moves using relative offset coordinates is
that I is the X offset to the center of the arc, J is the Y offset, and
K is the
Z offset. In general, most controls don't support arc moves except in on
of the 3 planes, XY, YZ and ZX. There are more sophisticated controls
that can do arbitrary arc moves in any plane.

A lot of controls do not support arc moves with the R radius, unfortunately.

So, to follow the language spec, an XY plane move may only have I and/or J.
A YZ arc may only have J and/or K. A ZX arc may only have I and/or K.

The plane of arc moves is specified by :
G17 XY plane
G18 ZX
G19 YZ

Jon