Page 34 - 6653

sizes form a scattering field, which is split into 8-10 intervals
and determines the number of sizes in each interval. Under the
abscissa, the size of the parts A, which determine the scattering
field, is broken down at intervals, and along the ordinate axis,
the number of parts of m that fall into a defined interval (the
repetition rate). The mean of the intervals and their
corresponding frequencies are characterized by points on the
graph. After connecting the points with a broken line, a
separation polygon is obtained (Fig. 8.1), which corresponds to
the distribution curve for an infinitely large number of
measurements.
The difference between the largest and smallest
dimensions obtained during the measurement determines the
dispersion of the size, which should not be greater than the
tolerance for processing. If the dispersion of the dimensions
goes beyond the tolerance, this indicates that the processing
errors in a certain part of the details are more permissible (they
are defective).
Random factors can change the shape of the curve
without changing its placement relative to the coordinate axes.
With an increase in actual dispersion of dimensions (for
example, when equipment wears out), the curves become
slower. Systematic factors determine the displacement of the
distribution curve along the axis of dimensions. As the
instrument wears out for some time, the center of error
grouping shifts. If during the processing of the batch of parts
the machine has been adjusted several times, then the
distribution curves are as many as adjustments, but all the
curves are shifted one relative to theother.

31