Freescale Semiconductor, Inc.
PREDICTION
In control assignable
causes eliminated
TIME
TIME
Out of control
(assignable causes present)
SIZE
Process “under control” – all assignable causes are
removed and future distribution is predictable.
SIZE
?
?
?
?
?
?
?
?
?
?
?
Lower
Specification Limit
Upper
Specification Limit
PREDICTION
In control and capable
(variation from random
variability reduced)
TIME
TIME
SIZE
In control but not capable
(variation from random variability
excessive)
SIZE
Figure 2. Impact of Assignable Causes on
Process Predictable
Figure 3. Difference Between Process
Control and Process Capability
At Motorola, for critical parameters, the process capability
is acceptable with a Cpk = 1.33. The desired process
capability is a Cpk = 2 and the ideal is a Cpk = 5. Cpk, by
definition, shows where the current production process fits
with relationship to the specification limits. Off center
distributions or excessive process variability will result in less
than optimum conditions
improved stability, the interval may be increased. The data
collected must be carefully documented and maintained for
later correlation. Examples of common documentation
entries would include operator, machine, time, settings,
product type, etc.
Once the plan is established, data collection may begin.
The data collected will generate X and R values that are
plotted with respect to time. X refers to the mean of the
values within a given subgroup, while R is the range or
greatest value minus least value. When approximately 20 or
more X and R values have been generated, the average of
these values is computed as follows:
SPC IMPLEMENTATION AND USE
DMTG uses many parameters that show conformance to
specification. Some parameters are sensitive to process
variations while others remain constant for a given product
line. Often, specific parameters are influenced when
changes to other parameters occur. It is both impractical and
unnecessary to monitor all parameters using SPC methods.
Only critical parameters that are sensitive to process
variability are chosen for SPC monitoring. The process steps
affecting these critical parameters must be identified also. It
is equally important to find a measurement in these process
steps that correlates with product performance. This is
called a critical process parameter.
Once the critical process parameters are selected, a
sample plan must be determined. The samples used for
measurement are organized into RATIONAL SUBGROUPS
of approximately 2 to 5 pieces. The subgroup size should be
such that variation among the samples within the subgroup
remain small. All samples must come from the same source
e.g., the same mold press operator, etc.. Subgroup data
should be collected at appropriate time intervals to detect
variations in the process. As the process begins to show
=
(
+ X
+
3
...)/K
+
X
R
X
2
X
= (R1 + R2 + R3 + ...)/K
where K = the number of subgroups measured.
The values of
X and R are used to create the process
control chart. Control charts are the primary SPC tool used
to signal a problem. Shown in Figure 4, process control
charts show X and R values with respect to time and
concerning reference to upper and lower control limit values.
Control limits are computed as follows:
R upper control limit
R lower control limit
X upper control limit
X lower control limit
UCL
LCL
D4 R
D3 R
R
R
UCL
LCL
A2 R
A2 R
X
X
X
X
1–14
www.motorola.com/semiconductors
Motorola Sensor Device Data
For More Information On This Product,
Go to: www.freescale.com
MOTOROLA [ MOTOROLA ]
