Re: Elastic belt stretch, lead screw twist, and far end encoders

--- In CAD_CAM_EDM_DRO@y..., "Ed Koffeman" <edk@k...> wrote:

> > In other words, the shaft
> > twist is present as an error in all cases: When the encoder is at

the pulley

> > end, the table will experience error from shaft twist when the

nut is far

> > from the encoder, and vice versa when the encoder and table are

at opposite

> > ends.
> > Comments?
>
> It seems to me that if the encoder is on the undriven end, it will

have zero screw twist error, since there is no load on the screw

> between it and the nut on the carriage.
>
> Ed Koffeman

I think your right Ed. The location of the table should be
slightly better known, as the twist is not in the equation with a far
end encoder. On machines with lead screw diameters that are small
relative to their loading, it could matter. In the example machine,
0.0001" of error on a heavy, probably roughing, cut, I could ignore.
The belt would seem to be the biggest contributor.
That might be another lesson here. Especially if you use a
stepper driven machine, without feedback. Maximizing the pulley
diameter, and or belt width would reduce the maximum belt stretch, as
the EA plot has a fairly steep slope in the beginning. This means
that the stretch increases fairly rapidly with increasing load. This
might also mean that light cuts are better on a stepper machine for
obtaining accuracy, since this stretch is not taken into account. A
direct mounted motor should have a better chance here, as it would
not have the stretch to contend with.

Brad Heuver