IPC-TM-650 EN 2022 试验方法1.pdf - 第49页

Note: Note: shall shall Figure 5-1 Ma intai n ing Perpendicularity throughout the Microsection Process 90º NO T PERPENDICULAR PERPENDICULAR IPC-TM-650 Page 2 of 8

Note:

IPC-MS-810

ASTM E 3

Material in this Test Methods Manual was voluntarily established by Technical Committees of IPC. This material is advisory only

and its use or adaptation is entirely voluntary. IPC disclaims all liability of any kind as to the use, application, or adaptation of this

material. Users are also wholly responsible for protecting themselves against all claims or liabilities for patent infringement.

Equipment referenced is for the convenience of the user and does not imply endorsement by IPC.

Page 1 of 8

Note:

shall

Figure 5-1 Maintaining Perpendicularity throughout the

Microsection Process

90º

NOT PERPENDICULAR PERPENDICULAR

IPC-TM-650

Page 2 of 8

A graph showing only the relationship between the measurement variability and the spec limits is shown below. Here the

normal curve represents only the variability of a given measurement. The curve clearly consumes half the spec width. Place-

ment of the curve in the spec interval is arbitrary. The curve is formed by drawing it with a total width of six times the

Sometimes there is only one spec, and the table above will be blank. For those situations, the next table compares measure-

ment variability to the total observed variation.

The numbers in the table are again percentages, but this time percentages of total variation. In this example, the measure-

ment variation is 23% of the overall total observed variation. The same guidelines apply, except measurements would con-

sume less than 10% of the total variation. Unacceptable measurements would consume more than 30%. This example falls

in the gray zone in between; not good enough, with improvement recommended.

The ﬁnal metric in the scorecard provides a measurement tolerance. The stated value forms a 99% conﬁdence interval. The

tolerance is shown in this example:

A ‘‘conﬁdence interval’’ is one of those strange statistical concepts that ought to be easy to describe, but somehow become

garbled when translated from equation to English. Let us try to state it clearly:

We make a measurement to estimate the actual, but unknown, performance point of the product. There is a 99% chance that

the actual performance will be within ± the measurement tolerance of the measurement.

In this example, any given gasket measurement will have a tolerance of ± 0.111 mm. The real gasket thickness is within ±

0.111 mm of our measurement.

The ﬁnal step is to determine lessons learned from the MSA and document any changes to the test procedure. If the evalu-

ation indicates the test procedure needs to be improved, these improvement projects should be undertaken as soon as pos-

sible.

0.2 0.4 0.6 0.8 1.0 1.2 1.4

Spec

limits

Measurement

error

Comparison to Total Variation

% Repeatability 2.76653 Percent of total variability used by repeatability error

% Reproducibility

2.69817

Percent of total variability used by reproducibility error

% Measurement R&r

5.46469

Percent of total variability used by measurement error

Compared to the total variability, the precision of

this measurement:

Measurement Tolerance

Units

On the basis of this study, there is a 99% chance that

this interval around the measured product performance

will contain the actual product performance level.

+/- 0.111243938