176 Chap.

7 / Computer Numerical Control

(c) Motor speed = table velocity (feed rate) divided by screw pitch, corrected
for gear ratio:

rg fr 515002
Nm = = = 416.667 rev/min
p 6.0
Comment: Note that motor speed has the same numerical value as in
Example 7.1 because the table velocity and motor gear ratio are the same.

7.4.3 Precision in Positioning Systems

To accurately machine or otherwise process a work part, an NC positioning system must

possess a high degree of precision. Three measures of precision can be defined for an NC
positioning system: (1) control resolution, (2) accuracy, and (3) repeatability. These terms
are most readily explained by considering a single axis of the positioning system, as de-
picted in Figure 7.14. Control resolution refers to the control system’s ability to ­divide the
total range of the axis movement into closely spaced points that can be distinguished by the
MCU. Control resolution is defined as the distance separating two adjacent a­ ddressable
points in the axis movement. Addressable points are locations along the axis to which the
worktable can be specifically directed to go. It is desirable for control resolution to be as
small as possible. This depends on limitations imposed by (1) the electromechanical com-
ponents of the positioning system and/or (2) the number of bits used by the controller to
define the axis coordinate location.
A number of electromechanical factors affect control resolution, including screw
pitch, gear ratio in the drive system, and the step angle in a stepper motor for an open-
loop system or the angle between slots in an encoder disk for a closed-loop system. For an
open-loop positioning system driven by a stepper motor, these factors can be combined
into an expression that defines control resolution as
p
CR1 = (7.17)
nsrg

Distribution
of mechanical
Addressable errors
points
Desired
position

Linear
Accuracy Repeatability axis
CR = 3
= +3
2
Control resolution = CR

Figure 7.14 A portion of a linear positioning system axis, with

definition of control resolution, accuracy, and repeatability.
Sec. 7.4 / Analysis of Positioning Systems 177

where CR1 = control resolution of the electromechanical components, mm (in);

p = leadscrew pitch, mm/rev (in/rev); ns = number of steps per revolution; and
rg = gear ratio between the motor shaft and the screw as defined in Equation (7.6). The
same expression can be used for a closed-loop positioning system.
The second factor that limits control resolution is the number of bits used by the MCU
to specify the axis coordinate value. For example, this limitation may be imposed by the
bit storage capacity of the controller. If B = the number of bits in the storage register for
the axis, then the number of control points into which the axis range can be divided = 2B.
Assuming that the control points are separated equally within the range, then

L
CR2 = (7.18)
2B - 1
where CR2 = control resolution of the computer control system, mm (in); and L = axis
range, mm (in). The control resolution of the positioning system is the maximum of the
two values; that is,
CR = Max 5CR1, CR2 6 (7.19)
A desirable criterion is CR2 … CR1, meaning that the electromechanical system is
the limiting factor that determines control resolution. The bit storage capacity of a mod-
ern computer controller is sufficient to satisfy this criterion except in unusual situations.
Resolutions of 0.0025 mm (0.0001 in) are within the current state of CNC technology.
The ability of a positioning system to move the worktable to the exact location de-
fined by a given addressable point is limited by mechanical errors that are due to various
imperfections in the mechanical system. These imperfections include play between the
screw and the worktable, backlash in the gears, and deflection of machine components.
The mechanical errors are assumed to form an unbiased normal statistical distribution
about the control point whose mean m = 0. It is further assumed that the standard devia-
tion s of the distribution is constant over the range of the axis under consideration. Given
these assumptions, nearly all of the mechanical errors (99.73%) are contained within {3s
of the control point. This is pictured in Figure 7.14 for a portion of the axis range that in-
cludes two control points.
These definitions of control resolution and mechanical error distribution can now be
used to define accuracy and repeatability of a positioning system. Accuracy is defined under
worst case conditions in which the desired target point lies in the middle between two ­adjacent
addressable points. Since the table can only be moved to one or the other of the addressable
points, there will be an error in the final position of the worktable. This is the maximum
possible positioning error, because if the target were closer to either one of the addressable
points, then the table would be moved to the closer point and the error would be smaller. It
is appropriate to define accuracy under this worst-case scenario. The accuracy of any given
axis of a positioning system is the maximum possible error that can occur between the desired
target point and the actual position taken by the system. In equation form,
CR
Ac = + 3s (7.20)
2
where Ac = accuracy, mm (in); CR = control resolution, mm (in); and s = standard
deviation of the error distribution. Accuracies in machine tools are generally expressed
for a certain range of table travel, for example, {0.01 mm for 250 mm ( {0.0004 in. for
10 in) of table travel.
178 Chap. 7 / Computer Numerical Control

Repeatability refers to the ability of the positioning system to return to a given ad-
dressable point that has been previously programmed. This capability can be measured
in terms of the location errors encountered when the system attempts to position itself at
the addressable point. Location errors are a manifestation of the mechanical errors of the
positioning system, which follow a normal distribution, as assumed previously. Thus, the
repeatability of any given axis of a positioning system is {3 standard deviations of the
mechanical error distribution associated with the axis. This can be written as
Re = {3s (7.21)
where Re = repeatability, mm (in).

Example 7.3 Control Resolution, Accuracy, and Repeatability in NC

Suppose the mechanical inaccuracies in the open-loop positioning sys-
tem of Example 7.1 are described by a normal distribution with standard
deviation = 0.005 mm. The range of the worktable axis is 1,000 mm, and there
are 16 bits in the binary register used by the digital controller to store the
programmed position. Other relevant parameters from Example 7.1 are the
following: pitch = 6.0 mm, gear ratio between motor shaft and screw = 5.0,
and number of step angles in the stepper motor = 48. Determine the (a) con-
trol resolution, (b) accuracy, and (c) repeatability of the positioning system.
Solution: (a) Control resolution is the greater of CR1 and CR2 as defined by Equations
(7.17) and (7.18).

p 6.0
CR1 = = = 0.025 mm
nsrg 4815.02
1,000 1,000
CR2 = 16
= = 0.01526 mm
2 - 1 65,535
CR = Max50.025, 0.015266 = 0.025 mm

(b) Accuracy is given by Equation (7.20):

Ac = 0.510.0252 + 310.0052 = 0.0275 mm
(c) Repeatability Re = {310.0052 = {0.015 mm

7.5 NC Part Programming

NC part programming consists of planning and documenting the sequence of processing

steps to be performed by an NC machine. The part programmer must have a knowledge of
machining (or other processing technology for which the NC machine is designed), as well as
geometry and trigonometry. The documentation portion of part programming involves the
input medium used to transmit the program of instructions to the NC machine control unit.
The traditional input medium dating back to the first NC machines in the 1950s is 1-in wide
punched tape. More recently, magnetic tape, floppy disks, and portable solid-state memory