Re: [CAD_CAM_EDM_DRO] Question about angular and linear feedrates

Dan Statman wrote:

> I would like to know how most (all) control programs will handle the
> feedrate for the following moves:
>
> A-axis is angular with feedrate expressed in degrees/sec
> X-axis is linear with feedrate expressed in inches/min
>
> Assume that I am always starting at X0 and A0 for these moves.
>
> If I command: G01 A180 F1.0 it should take 180 seconds to make the move.
> If I command: G01 X0.5 F1.0 it should take 30 seconds to make the move.
>
> My question is how long should the following move take starting at A0 X0:
>
> G01 A180 X0.5 F1.0

This can't be answered with the information you give. You need to know the
radius of the part at the location of the cutter tip. What is most important is
the feedrate across the part surface at the cutting edge. If the Y and Z axes
are in such position that the radius is 1", then the first command should
take 3.14 * 2 (diameter) / 2 (half revolution) / 1.0 (IPM) = 3.14 minutes!
The second command is all linear, so 30 sec is correct.

Your 3rd command should tale slightly OVER 3.14 minutes, as it is slightly
longer due to the combination of the two motions. But, if the radius of the
cutting edge from the centerline of the A axis was greater, the move should
take longer, since more surface is covered.

Unfortunately, the current version of EMC, and a number of other programs
that started life as 3-axis cartesian programs, don't have this calculation in
them. Real 4+ axis machining programs usually have this information provided
in some manner. 4 and 5-axis machining centers usually have the rotary
axes on the spindle, and so the machine control has all the distances between the
joints and the spindle programmed into it for proper coordination of tool offsets
when the rotary axes move. On a machine where the part is moved by rotary
axes mounted to the table (this can be true of trunnion-type machines, too)
this info is not necessarily known by the CNC control, but it CAN be made
available. If one were to force the constraint onto the part program setup that
Y and Z zero were such that the spindle centerline and tool tip were exactly
intersecting the A axis centerline, then you could make the correct calculations
of of feedrate based on the actual surface being defined by the toolpath.

This is generally how I would program such a part, anyway, as any other scheme
would lead to big headaches, I think.

Jon