On 9/30/26 05:36, andy pugh wrote:
On Wed, 30 Sept 2026 at 09:22, Luca Toniolo<[email protected]> wrote:
The idea is a reference radius for each rotary axis, set in the INI. For
example, `[AXIS_A] FEED_RADIUS = 25` puts A in the feed calculation as
if it were a length at 25 mm from its centre.
When doing rotary axis work there are two ways that you can set the Z
axis zero (or, more generally, the distance along the tool axis).
You can set zero as the stock surface (this would be common for
engraving work) or at the axis of rotation (this would be usual for
machining cam profiles, etc).
I don't think that your INI proposal addresses either situation.
My home-made rotary's all have a home switch.
What _might_ work would be for the INI file to define the position (in
machine coordinates) of the rotational axes. (or, to be more specific,
the location of the axis of each of the rotary axes)
Then the surface feed speed at the current tool tip can be calculated.
Possibly this mode would need to be switched on/off with a G-code.
G94.1 maybe?
Or, purely in G-code G94.1 Rxxxx could program the current working
radius for angular speed calculation.
CSS mode on a lathe does something similar, but different, in that it
assumes the X diameter is the actual workpiece diameter (this is a
fairly safe bet with a lathe)
Would _I_ use it? Possibly, but also maybe not, only if the CAM system
that I was using knew how to do it would it be truly useful (and CAM
systems can accurately use G93 as they know exactly where the stock
is).
It's an attractive idea, and would possibly simplify trajectory
blending in combined rotary/linear moves. But would also maybe make
LinuxCNC differ ever more from other G-code standards.
There may be more valuable places to focus attention.
Agreed.. Off topic, is there a CAM that's both open src and free? The
lack of which means I write my own g-code. Am I nearly alone in the DIY
arena? For additive work, as in 3d printers, there are many design
kits, I use Openscad due to its gentle hills learning it.
Cheers, Gene
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users
