On 7 October 2011 14:55, Viesturs Lācis <[email protected]> wrote:

> Do You mean that G17.1 will not work??? That should select UV plane
> and let do arcs.

Not according to
http://www.linuxcnc.org/docview/html/gcode_main.html#sub:G2,-G3:-Arc

> Is there _any_ solution for that?

If the CAM can convert arcs to short line segments, then it can be
done that way (and emc2 will blend the path).

The problem is largely that I, J and K set X, Y, Z offsets. You could
reasonably expect G17.1 G3 R to do a UV arc, though, but would you
want to do simultaneous arcs in XY and UV?
In that case, you might want to run arcs in both XY and UV, so perhaps
a G3.1 is needed? But then how do you define the arc centre in a
combined G3 and G3.1 move, on the same line?

I don't even think we can do 3D arcs, G3 X2 Y3 Z5 I0 J2 K 4 might
unabiguously define an arc in 3D space, but I don't think EMC2
supports that format. Arcs are 2D with an optional perpendicular
movement. Of course, then it becomes very difficult to define
clockwise and anticlockwise.

Your case is simpler, you don't seem to want different axes to do
different things at the same time. The solution might be to use M1xx
codes to switch control modes, so that the G-code always uses XYZ, but
that controls both axes, or one, or the other.

I think you would need to either only allow a switch with both axes at
their home position (seems wise anyway) or possibly to write a HAL
component to allow bumpless transfer (ie synchronise the stepgens at a
controlled rate while lying about feedback until they match).

-- 
atp
"Torque wrenches are for the obedience of fools and the guidance of wise men"

------------------------------------------------------------------------------
All of the data generated in your IT infrastructure is seriously valuable.
Why? It contains a definitive record of application performance, security
threats, fraudulent activity, and more. Splunk takes this data and makes
sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-d2dcopy2
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to