Physico–statistical models based on the residual defectiveness of structural materials - Probabilistic Safety Assessment for Optimum Nuclear Power Plant Life Management (PLiM) - page 128

Table 5.1 Detection of incomplete fusion defects in a plate of steel
15Kh2MFA with a cladding when using a test sample

Operator
No. 2

Operator
No. 3

Operator
No. 4

Operator No. 1

Defect
size,
mm 2

Number
of defects

Number % Number % Number % Number %

12 9 3 33.3 1 11.1 - - - -

20 12 3 25 2 16.6 (1) - - -

29 15 3 (6) 20 (40) 2(4) 13.3 1 6.6 3 20

63 10 3 (6) 33 (50) 4 40 3 30 3 30

50
(66.6)

113 12 3 (8) 9 75 4 33.3 7 58.3

Average
detectability

34.5 (43.1)% 31 (34.5)% 26.4% 35%

Detection of unac-
ceptable defects

35.2 (44)% 44 (50)% 26.4% 35%

Note : 1. Values in parentheses indicate the number of identified defects of this size, and
detection without taking into account estimates of the size of defects. 2. Operator No. 3
classified a defect with area of 20 mm 2 as unacceptable, i.e. having a large area. 3. The time
spent by operators in inspection: operator No. 1 2 h 30 min; operator No. 2 2 hr 40 min;
operator No. 3 1 hr 15 min; operator No. 4 2 h 05 min.

meas 2

meas 2

S , mm

S , mm

eq

eq

Permissible
according
to agreement

￿ ￿ ￿ ￿ ￿

Permissible
size

S f S f

a) b)

5.8 Results of inspection of the test sample with incomplete fusion
on the side of the base metal (a) and the cladding (b).

control of four NDT inspectors.
In a second experiment on the same test sample, but with a different
composition of NDT inspectors, the results of the inspection on the side
of the base metal and on the cladding side were compared (Fig. 5.8). It
is evident that inspection on the side of the base metal leads to excessive
rejection of defects.