Re: Question about angular and linear feedrates

Although we're looking at the simple example here that Dan has asked
about, ideally the controller should also be able to deal with more
complex moves such as a situation where the Y,Z and A axes all move
but the toolpoint stays stationary or hardly moves in respect to the
workpiece, only altering the relative angle between the tool and
workpiece.
Of course, if the controller handles the G93 command then the
calculations can be pre-processed and the controller dosn't need to
do any further work. Even if the controller dosn't handle G93 the
feedrate can be suitably adjusted if you know how the controller's
angular / linear feedrate relationship works. However, to be accurate
these calculations all need to know where the toolpoint is in
relation to the centre of rotation of the A-axis to work out distance
and from there feedrate (or the inverse feedrate), so a value is
still needed somewhere along the line.
The same applies to 5-axis control, where the offset between the
toolpoint and the centre of rotation of the B-axis is required.

What I don't understand is why some controllers don't clamp the
angular feedrate to the maximum set value when doing mixed linear /
angular moves. This wouldn't give perfect feedrates, but would at
least be useable without doing aditional calculations.


Here's another good document on the subject, page 49;
http://www.isd.mel.nist.gov/personnel/kramer/pubs/RS274NGC_3.pdf

Rab

--- In CAD_CAM_EDM_DRO@y..., Brian Pitt <bfp@e...> wrote:
> lets keep it simple
> if the control dosent have a register for the A-axis diameter offset
> it should at least be able to do an interpolation based on the user
units
> of both axes
>
> in the example Dan gave I'm guessing the software did the position
> interpolation right and the tool path was a smooth helix from
> X0,A0 to X.5,A180 but for the velocity interpolation only used the
X-axis
> value and ignored the larger value of the A-axis movement and
> forced the A-axis to move too fast
>
> if the control does have an A-axis diameter offset it can use that
> to calculate the feedrate but if it dosent or if that offset is set
to zero
> then the control should calculate the feedrate with the values its
given
>
> Brian
>
> On Monday 05 August 2002 01:42, you wrote:
> > What the control really requires is the relative position of the
zero
> > point to the centre of rotation of the A-axis so it can calculate
the
> > actual coordinates and distances travelled by the toolpoint for
all
> > subsequent movements.
> > If the zero point is directly above the centre of rotation and
almost
> > touching the work surface of a cylynder then that of course is the
> > radius. I've termed this Initial Tool Offset around A-axis (ITOA)
to
> > hopefully save confusion unless anyone has a better term.
> > I assume that the zero point for 4-axis work is normally set at or
> > above the centre of rotation of the A-axis. Is it likely that the
> > zero point would ever be offset in the Y-axis ?
> >