Operating Manual

189188
In practice the following rule of thumb is
applied to the detection of planar defects-
with a high probability:
“a defect is detectable if the angle between
the X-ray beam and the defect is
approximately 10° or less”.
The value of 10° is based on decades of
practical experience but does not gua-
rantee detection. The rule is visualised
in the graph of figure 15-17. To detect
2D defects with unknown orientations
and exceeding 10°, a multi-angle tech-
nique would be required, which in
general is impractical. As a measure to
enhance detection of lack of fusion
defects in critical welds, sometimes two
additional shots are made in the direction of the weld preparation. Apart from orientation,
the detectability of 2D defects is also dependent on the type of defect, its height and its ope-
ning (width). Lack of side wall fusion (LOF) - dependent on the welding method - will in
general be easier to detect than a crack because LOFis often accompanied by small 3D slag-
type inclusions.
In fact, LOF defects are in practice often
detected because of the presence of these
small inclusions rather than because of
the LOF itself. Such secondary small
defects do usually not occur along with
cracks, although their character might be
erratic and their small facets under diffe-
rent angles - some of them just in line with
the X-ray beam - might help to detect them
as illustrated in figure 16-17. From experi-
ments it is known that, if more than 1 to
2% of the material in the line of radiation
is missing, a defect is detectable.
3D CT for sizing of defects in (welded) components
To know the through-thickness size of a defect can be of paramount importance to calcula-
te the strength of a cracked component, its remaining strength or its fitness for purpose.
Traditional single-shot (and even multi-shot radiography) is unable to measure through-
thickness height of planar defects.
Even detection itself is not always easy, as the previous paragraph describes. Therefore
sizing of defects once detected with radiography is often done by ultrasonics, with accepta-
ble sizing accuracies for engineering critical assessment (ECA) calculations.
The application of ultrasonics on welds requires that both the material and the weld are
acoustically transparent. This requirement is often not met in welded or cast austenitic
materials as used in nuclear power plants.
In such cases 3D Computer Tomography
(3D CT) can provide a solution. In secti-
on 17.3, CT systems for low to moderate
energies and with extreme small foci are
described to inspect small components
with low radiation absorption.
For sizing and sometimes detection of
defects in welds or (cast) stainless steel,
high energy 3D CT systems have been
developed which are able to accurately
size randomly oriented cracks or other
(planar) defects with a minimum width
as small as 20 microns.
Such systems use common high-energy
X-ray tubes with common focus dimen-
sions of a few mm
2
, in combination with line detector arrays. Figure 17-17 schematically
shows the set-up for a series of exposures to create a 3D CT image. Synchronised and simul-
taneous movement of the X-ray source and the detector causes only a particular volume of
the material (slice) to be “in focus”. All information from the adjacent area is out of focus
and does not contribute to the image of the defect. Many of such focus areas (volume pixels
or “voxels”) are stored, for instance a few hundreds per slice. If the detector has sufficient
length, it does not need to move (“virtual movement”).
In performing a CT scan, the X-ray beam goes through a wide range of angles including the
angle(s) of the defect. Numerous slices are made, together resulting in a large number of
voxels. On the basis of these many voxels the image of the entire defect can be reconstructed,
including its position, orientation and depth location in the component. This can be done
with considerable accuracy, typically 1mm or better; accurate enough for calculation purpo-
ses. Figure 18-17 shows a crack in an austenitic weld and next to it its CT reconstruction. Figure
19-17 shows another example of a CT image obtained with this system. The image
shows two
planar defects in a V-shaped weld with a clear indication of their orientation and size.
Fig. 15-17. Graphical presentation of “rule-of-thumb”
Fig. 16-17. Crack facets creating the defect image
Fig. 17-17. Schematic of 3D CT defect detection and sizing
GOOD
DEFECT CONTRAST
POOR
X-ray beam
Film
Unfavourable
orientations
invisible
on film
Facets creating
the image
Crack image
Movement
X-ray positions
Out of focus
Area of focus
Detector positions
Open planar
defects
Tight planar
defects
Angle O/
Rule-of-thumb
Crack with facets
Fig. 18-17. Crack in austenitic weld
CT reconstruction of the crack
Macro of crack in a weld
Fig. 19-17. CT reconstruction of two planar
defects in a weld