Manufacturing Engineering, 6/2002 Pre-cutting Prep. Pp60-66
Checking
the Volume Charles
Wang President, Optodyne Inc. Compton, CA Recently there has been a lot of
discussion about the validity of a
ball bar test, and the usefulness of the
laser interferometer. Some
say these instruments don't
give enough information about the
machine's performance. The laser interferometer
checks each axis for position accuracy, but the results are
independent of one other, Because the laser interferometer
checks only the
end-points. It does not check the machine under true dynamic conditions or the
actual tool path.
Two forms of ball bar. They are
the telescoping ball-bar and the laser. Both check the machine
dynamically.
The circular tests show how the X and Y axes work together to move the machine
in a
circular path. Path data, plotted in
polar coordinates, show any deviation the
machine makes
from a perfect circle. The shapes are diagnosed and correlated to performance
features such as
servo mismatch, backlash, reversal spikes, squareness error, cyclic error, stick
slip, and machine
vibrations. The telescoping ball-bar is a one-dimensional measurement.
Basically, only the
radial changes along angular positions are measured. Angular positions are not
measured but are
calculated by assuming the machine feed rate is constant. Because the
telescoping ball bar
systems normally work with radii of 150-300 mm and feed rates of 6-10 m/min,
they cannot follow
circular tests with smaller radii, and the errors they detect usually trace back
to problems with the
machine's geometry rather than its controller. These
machine errors are larger than those produced by the control loops and
consequently hide them. Because the telescoping
ball-bar uses a transducer
for the measurement, periodic
calibration is
necessary, and the unit is sensitive
to temperature changes. The telescoping
ball-bar has
a cable between the transducer (inside
the telescoping bar) and the electronic
processor that makes
a circular path with multiple revolutions difficult.
