C.2 Methods of calculation
The following two methods illustrate ways to calculate inaccuracy of a reading:
Method C.2.1
Method C.2.1 calculates the inaccuracy of a measurement by adding the inaccuracies of all the influencing parameters.
Method C.2.2
Method C.2.2 is a method for calculating the inaccuracy, /g, of a measurement result, MR, in accordance with ISO 14253-2 where the measurement result is equal to the reading, R, ± the inaccuracy, /g.
MR = R±lg
where
К shall be chosen for a level of confidence, e.g.:
К = 1 for 68 % level of confidence;
К = 2 for 95 % level of confidence;
К = 3 for 99,8% level of confidence;
oi is the uncertainty for each parameter obtained:
either by statistical approach;
or by other methods, e.g. standards, specifications, analysis;
/ is the different parameters which have been considered independent (for instance; surface condition, linearity, repeatability).
Statistical distribution:
uniform or rectangular law: $ = 0,6 a;
Gauss law: o; = 0,5 a,
where
a is the accuracy of the result.Table C.2 — Illustrates the application of methods C.2.1 and C.2.2 where the test object is a steel plate 10 mm thick
surface roughness Ra= 6,3 pm
|
Parameter |
Group |
Factors |
Measuring conditions |
Estimated inaccuracy mm |
|
|
|
|
|
|
Method C.2.1 |
Method C.2.2 |
|
Test object |
Material |
Composition |
Ferritic steel |
0 |
0 |
|
Structure |
Fine grained |
|
|
||
|
Anisotropy |
|
|
|
||
|
Surface condition |
Cleanliness |
|
|
|
|
|
Roughness |
Surface Ra = 6,3 pm |
0,006 3 |
0,003 2 |
||
|
Surface profile |
Flat |
|
|
||
|
Coating |
Coating |
Not coated |
0 |
0 |
|
|
Paint |
Not painted |
|
|
||
|
Surface treatment |
Un-treated |
|
|
||
|
Geometry |
Non-parallelism |
Parallel faces |
0 |
0 |
|
|
Curve radius |
No curvature |
|
|
||
|
Range |
Negligible attenuation |
|
|
||
|
Reference |
Calibration Method |
Uncertainty of calibration method |
Reference block Same material / 5 step calibration |
0 |
0 |
|
Reference block |
Thickness and velocity uncertainty |
Thickness uncertainty: 0,01 mm Velocity uncertainty: ± 30 m/s |
0,05 |
0,025 |
|
Table C.2 — Illustrates the application of methods C.2.1 and C.2.2 where the test object is a steel plate 10 mm thick, surface roughness Ra=
6,3 pm (concluded)
|
Measuring |
Equipment |
Resolution |
Digital instrument resolution: 0,01mm |
0,01 |
0,006 |
|
Cable length |
Fixed length |
0 |
0 |
||
|
Drift of instrument |
Stable instrument |
0 |
0 |
||
|
Time of flight |
Accuracy of time measurement: 10 ns |
0,03 |
0,018 |
||
|
Linearity |
1 % of the maximum range (manufacturer's data) |
0,1 |
0,05 |
||
|
Measuring |
Operation |
Trigger point |
Constant amplitude |
0 |
0 |
|
V-path |
Single element probe |
0 |
0 |
||
|
Phase shift |
No phase shift |
0 |
0 |
||
|
Repeatability |
Operation |
Coupling |
Coupling error included in method |
|
|
|
User training |
Qualified operator |
0,1 |
0,05 |
||
|
Miscellaneous |
Temperature |
Variation of sound velocity |
Measurement at room temperature, negligible variation |
0 |
0 |
|
|
|
|
|
|
|
|
|
Global error |
0,306 |
0,135 |
||
Annex D
(informative)
Measuring technique selection
Figure D.1 — Flow chart for manufacturing inspection
Figure D.2 — Flow chart for manufacturing inspection
Figure D.3 — Flow chart for in-service inspection
Figure D.4 — Flow chart for in-service inspection
Bibliography
[1] EN 473, Non-destructive testing — Qualification and certification of NDT personnel — General principles
