Physico–statistical models based on the residual defectiveness of structural materials - Probabilistic Safety Assessment for Optimum Nuclear Power Plant Life Management (PLiM) - page 130

Table 5.3 The detection of the total number of defects in the
control of different methods

Parameter Method No. 1 Method No. 2 Method No. 3

N det 11 5 4

K d 0.81 0.38 0.307

K ex - 3.6 1.92

Note. N det is the number of detected defects coincident on the coordinate with the certificate
of the sample; K d is the coefficient of defect detection having the form of the ratio of the
number of detected defects, which coincide with the certificate for the sample, to the number
of defects embedded in the sample, K ex is the coefficient of excessive rejection in inspection.

levelwith a once- reflected beam. Sensitivity setting for inspection with the
converter with the prism angle of 40 o was performed in a 3.5; 2.0 mm notch
in the base metal in the direct beam. The lowest level recorded in inspection
with the transducer with a prism angle of 50° is shown in Table 5.2.
The lowest level recorded in inspection with the transducer with the
prism angle of 40 o corresponded to a signal from a level of 8 dB.
The inspection No. 2 (Fig. 5.9, curves 2 and 3) was developed by one
of the institutes servicing thermal power engineering. Curves 2 and 3 were
obtained by the method No. 2 by different groups of NDT inspectors: curve
2 by the developers of the method, curve 3 by the NDT inspectors from
one of the nuclear power plants.
The method No. 2 is described in the document MU 34-70-023-86. The
acoustic customised transducer produced a slowly diverging acoustic beam
of the shear wave with the main parameters: frequency ƒ = 1.3 MHz; input
angle α 0 = 48 o , width of the angular capture A = 7, the amplitude of the
echo signal at 6 dB from the cylindrical side reflector with diameter of
6 mm SOP No. 2.
The weld was inspected with a standard flaw detector UD 2-12 with an
acoustic beam reflected once from the inner surface of the sample from
twosides of the weld.
Curves 4 and 5 (see Fig. 5.9), obtained by the computational method, 12
integrally reflect the state of inspection of pipelines DN 500 at the nuclear
power plant in 1988 and 1991, respectively.
In assessing the detectability of defects by various techniques it was
found that the slag and cracks, as well as the total number of defects are
detected more efficiently by the method No. 1 (Table 5.3).
Additional studies were subsequently carried out using an automated
ultrasound inspection system 'Sumiad' together with X-ray inspection.
Automated ultrasonic inspection was performed by standard technology
using the Sumiad automated inspection system developed by Technatom
(Spain) with printout of the test results. The general view of the apparatus,
mounted on a test sample, is shown in Fig. 5.10. Sensitivity was set in

￿ ￿ ￿ ￿ ￿