Search

A new method and principle were presented for the continuous measurement of the depth-dependent reflection coefficient of Rayleigh waves or their pulse interacting with various surface discontinuities. The method was to record the reflection echo amplitudes for varying depth during scanning the Rayleigh wave pulse with a small lateral beam width along a surface discontinuity with a small slope in depth. In the experiments using a declined slot and intensively focused Rayleigh waves, the coefficients were continuously measured with quite good reproducibility in the range of 0.05

Intensive focusing of Rayleigh waves was realized by the fabrication of a curved PZT (Pb(Zr 2 Ti)O 3 ) piezoelectric element having a new shape. Two- and three-dimensional beam profiles of the focusing transducer employing the curved PZT element were obtained by a contact scan on a steel block with flat bottom holes. The beam intensity was higher, about 35 dB, at the focal distance and the signal-to-noise ratio was higher, about 17 dB, than the focusing transducer employing a flexible PVDF (polyvinylidene fluoride) film in the earlier study. The −6 dB and −3 dB lateral widths at the focal point were 0.61 mm and 0.46mm, and the −6 dB and −3 dB axial widths were about 5.5mm and 4.0mm, respectively. The axial stress field showed a good agreement with the overall pattern estimated by a simplified model. The ratio of the detected smallest flaw size to the wavelength of Rayleigh waves for the center frequency was about 0.29. As an application, a single time scan was performed along the flat bottom holes (FBHs) at the focal distance, and an image to show sharp peaks with amplitude depending on the size of the FBHs was obtained. The focusing transducer showed the characteristics of a narrow lateral resolution, high energy, and high signal-to-noise ratio, and was practical for the detection of surface flaws smaller than the wavelength.

In conventional ultrasonic testing of a thin layer with thickness less than about twice the wavelength, successive echo signals become inseparable. Recently, a new time-domain analysis technique to determine the thickness or the wave velocity of a thin plate was developed and through-transmission experiments were reported. We have studied the validity of the technique and its applicability to contact testing under a reflection field system. We found a mistake in the derivation of the reconstructed reflection field equation reported, and a correction has been made. We performed the experiment in a reflection field system using contact type transducers of centre frequency 1, 2.25, and 5 MHz with water layers of thickness ranging from 0.06 to 1.44 mm (0.04 ≤ thickness/wavelength h ∗ λ ) larger than about 0.4, 3% for h ∗ λ larger than 0.8, and 1% for h ∗ λ larger than 2. The time domain technique with our corrected equation was applicable to the evaluation of a thin layered medium under reflection system using contact type transducers.

Eddy current probes were designed to be suitable for the test of fin tubes. Fin tubes, often used for heat exchangers, have uneven outer and inner surfaces to enhance the heat emission. The surface roughness makes it difficult to detect flaws employing eddy current test (ECT). In order to overcome the difficulties we performed two kinds of works, one is the development of ECT probes, and the other is the signal processing including fast Fourier transform and digital filtering. In the development of ECT probes, we adopted empirical design method. Our ECT probes for the fin tubes are inside type. And we are specially concerned about geometric features such as the widths of the coils composing an ECT probe. We fabricated four probes with various coil widths. Eddy current test was performed using those ECT probes on artificially flawed specimens. After analyzing the output signals, we found that, for the effective testing, the width of a coil should be determined considering the pitch of the fins of a tube. And we also learned that the frequency filtering could improve the s/n.

Green's function is used in order to derive the analytical solution for the change in impedance due to a presence of the flaws in a conductor. This solution is applied to a cylindrical flaw and a spherical flaw whose radii are much smaller than the radius of the test coil. For both cases, the change in impedance is obtained within Born's limit.

The impedance of a single-turn coil which surrounds a conducting cylinder with a flaw is calculated employing Green's function technique. The Born approximation is used in order to know how the impedance change due to the presence of a flaw depends on the conductivity and size of a flaw.

The impedance of a test coil was measured as a function of the dc current of an electromagnet, and the determination method of magnetization level with the normalized impedance diagram is discussed. An eddy current test for the outer side of a steel tube wall was also performed. Test results show that the impedance change of the test coil is brought about by the eddy current perturbation and the change of magnetization level. These effects on the flaw inspection are also analyzed.