Filtern
Dokumenttyp
- Beitrag zu einem Tagungsband (4) (entfernen)
Schlagworte
- Phased array (4) (entfernen)
When dealing in ultrasonic testing with inhomogeneous and anisotropic material structure such as
diverse types of components made from austenitic or nickel based cast, which are currently used
for modern power plant concepts, data interpretation is quite difficult. For better understanding of
the complex interaction between the sound field and the component under test, the mathematical
modeling of sound propagation in solids is a substantial task to increase the probability of
detection of relevant defects.
First we present a mathematical approach for modeling the three dimensional transient
particle displacement as a function of time in each point in a half space excited by an impulsive
point load at the surface. The transient ultrasonic field of a rectangular array element is calculated
with this approach by a point source synthesis. Based on this solution we model the wave
propagation of a phased array transducer by time delayed superposition of the wave field of the
transducer elements.
Next we use an electrodynamic technique to visualize the grazing sound field at the surface
of a test block radiated by a phased array probe. By detecting the grazing beam at the samples
surface with a small electrodynamic probe, we measured the particle displacement as a function of
time. It allows for measuring the displacement in all three spatial directions. This comprises the
detection of the horizontal and vertical particle displacement with respect to the surface and thus
also the detection of longitudinal and shear waves is possible.
The calculated and measured wave fields will be compared for different delay laws in
isotropic and transversely isotropic media. The results support the theoretical activities to model
the wave propagation and to find optimal testing parameters for different components and
configurations.
The spatial resolution in ultrasonic testing as a wave phenomenon in nature is limited to a
certain fraction of the wavelength, usually defined to be close to the Rayleigh criterion. In case of
complicated reflector surfaces – such as stress corrosion cracking - this limitation prevents an exact
visualization of the defect shape. There exist a few approaches to improve the spatial resolution, whose
reconstruction quality all in common also depend on the achieved signal-to-noise ratio of the raw data. In this
work we present a specific SAFT analysis, in which a high number of different angles of incidence produce a
sufficiently high number of different reflections at the crack edges resulting in an improved SNR. In doing
so, we reconstruct a coherent crack structure. First investigations were made at artificially simulated crack
configurations with different contours and curvatures in flat and cylindrical test blocks. The measurements
results visualized by representative scans – show details of crack design and crack orientation. We also will
present a comparison of the SAFT analysis between modelling and phased array measurements.
Within a know-how transfer project funded by the government conventional ultrasonic
technique was replaced by phased array technique for automated round-bar testing. Instead of
applying a great number of conventional probes to achieve acceptable volume coverage we used
curved linear arrays. The benefits of phased array technique such as programmable skew angles,
beamforming and beam positions, led not only to a significant decrease in inspection time, but also the
number of probes could be substantially reduced . Finally, the testing parameters for a large range of
bar-diameters could be adapted by software control instead of time-consuming mechanical
replacement. The probe-design was carried out by a proprietary modelling program. Both the
theoretical calculations as well as the latter experimental verifications revealed significant advantages
of curved arrays versus the planar types. A radial oriented probe offers perfect adaption to the
cylindrical shape of the specimen allowing wide variations of the sound field. Thus beam direction,
beam size and beam position could be optimized with respect to a minimum of inspection cycles, as
inspections have to be executed in-line during the production. A number of laboratory tests were
carried out on special test components. In order to achieve an optimal performance of the reference
rod we implemented three different types of reference reflectors: (i) flat-bottom-holes with diameters
of 0.8 mm and 1.2 mm, (ii) side-drilled-holes with a diameter of 0.7 mm for the detection of
volumetric flaws, and (iii) notches with a depth of 0.2 mm and 0.5 mm for the detection of surfaceoriented
defects. All laboratory tests were carried out with the COMPAS-XXL inspection system, a
proprietary development of BAM.