SIZING OF SMALL SURFACE-BREAKING TIGHT CRACKS BY USING LASER-ULTRASONICS

Size: px

Start display at page:

Download "SIZING OF SMALL SURFACE-BREAKING TIGHT CRACKS BY USING LASER-ULTRASONICS"

Katherine Morton
5 years ago
Views:

1 SIZING OF SMALL SURFACE-BREAKING TIGHT CRACKS BY USING LASER-ULTRASONICS M. Ochiai, T. Miura, H. Kuroa, S. Yamamoto, an T. Onoera Toshiba Corporation, Yokohama, Kanagawa, Japan Abstract: On the nonestructive testing, not only etection but also sizing of crack is esirable because the crack epth is one of the most important parameter to evaluate the impact of the crack to the material, to estimate crack growth an ultimately to preict lifetime of the component. Moreover, accurate measurement of the crack epth optimizes countermeasures an timing of repairs, an eventually reuces total cost for plant maintenance. Laserultrasonic is a technique that uses two laser beams; one with a short pulse for the generation of ultrasoun an another one, long pulse or continuous, couple to an optical interferometer for etection. The technique features a large etection banwith, which is important for small efect inspection. Another feature of laser-ultrasonics is the remote optical scanning of generation an etection points, which enables to inspect components in narrow space an/or having complex shapes. A purpose of this paper is to escribe the performance of a laser-ultrasonic testing (LUT) system on stress corrosion cracking (SCC) inspection. We have evelope a new technique for sizing shallow cracks, say mm, base on the laser-inuce surface wave an its frequency analysis. First, sizing capability of the system will be emonstrate by using an artificial surface-breaking slot having epth of 0-2mm in a stainless steel plate. Evaluate epths show goo agreement with the machine slot epths within the accuracy of about a few hunre micrometers. Then, SCCs in a stainless steel plate are examine by using the system. Depth of SCC is evaluate every 0.2mm over the crack aperture length. The evaluate epths are compare with the epths measure by the estructive testing. Introuction: Laser-ultrasonics has brought practical solutions to a variety of nonestructive evaluation problems that cannot be solve by using conventional ultrasonic techniques base on piezoelectric transuction [1,2]. Laserultrasonics uses two lasers, one with a short pulse for the generation of ultrasoun an another one, long pulse or continuous, couple to an optical interferometer for etection. Laser-ultrasonics allows for testing at a long stanoff istance an inspection of moving parts on prouction lines. The technique features also a large etection banwith, which is important for numerous applications, particularly involving small crack etection. In laserultrasonics, a pulse power laser is usually use to generate ultrasonic waves. When a laser pulse is irraiate onto a sample surface, an acoustic pulse is generate ue to thermoelastic or ablative interaction between laser an the material. Ablation process achieve by the irraiation of a high power laser pulse is more suitable to obtain intense ultrasonic signals. This metho of excitation simultaneously generates a various ultrasonic moes; surfaceskimming longituinal waves (P), Rayleigh waves (R), bulk longituinal waves (L) an bulk shear waves (S). These ultrasonic waves are etecte by another laser combine with an optical interferometer as a micro isplacement of the surface. One obvious application of laser-ultrasonics is nonestructive testing (NDT) of surface-breaking cracks an burie efects. Crack etection using R-wave, signal amplitue of which is the largest among the excite waves, ha achieve success. Cooper et al. [3, 4] showe that small slits having a epth of the orer of 100 m are etectable with laser-inuce R-wave in the ultrasonic pulse-echo measurements. In orer to measure a epth of the slit, one possible technique base on ultrasonic moe conversion at an ege of the slit has been suggeste; this technique however requires rather complicate ultrasonic propagation analysis incluing moe-conversion an fairy sensitive etection of weak moe-converte ultrasouns. On the other han, the laserinuce bulk ultrasouns have been use in a non-contacting thickness gauge an elamination etection [5], rarely for small crack inspection, because the signal-to-noise ratio (SNR) of the bulk waves are relatively poor to be use for micro crack etection. In this paper, we will review a recent evelopment in laser-ultrasonics for crack inspection on inustrial materials. Frequency analysis of R-wave is use to measure the crack epth [6]. The techniques are applie on a stainless steel plate with stress corrosion cracking (SCC). Results: An experimental setup, schematically shown in Fig.1, was use to investigate basic performances of laserinuce ultrasonic testing. Laser pulses from a Q-switche N: YAG laser with a maximum energy of about 100 mj/pulse in 10 ns pulse uration an a wavelength of 532 nm, was launche onto the surface. These energies are quite intense but are still below the threshol for optical fibre elivery [7]. The laser pulses were focuse into a small spot having a iameter of 400 µm to generate R-wave. The generate ultrasouns were etecte as micro surface isplacements using a confocal Fabry-Perot interferometer (CFPI) [8]. The etection system ha a broaban frequency response extening from about 1 MHz to 100 MHz. The signal from the interferometer was

2 converte to a igital waveform. Each waveform, representing the surface isplacement, was store into external memory for later signal processing. Trigger signal synchronize with the laser irraiation was also fe in orer to ientify the accurate time of ultrasouns generation. x y z Generation Laser 3-D Sample Scanning Stage Lens Optical Fibre (outgoing) Detection Laser Confocal Fabry-Perot Optical Fibre (incoming) Optical Detector Data Acquisition an Signal Processing Sync. Fig.1 Experimental setup for laser-ultrasonic A CFPI consists of two spherical mirrors having a high reflectivity arrange with their concave surface facing each other an separate by a istance, h, nearly equal to their curvature raius, as shown in Fig.2(a). The optical arrangement well known as an etalon has a sharp spectral response, as illustrate in Fig.2(b). The proceure of measuring small ultrasonic vibration using the CFPI is as follows: (1) A single-moe laser beam is irraiate onto a vibrating surface. (2) The original frequency, f 0 (=c/λ 0, c: velocity of light, λ 0 : wavelength of light), of the reflecte or scattere laser beam from the surface is moulate to f 0 ±f ue to the Doppler effect. (3) The frequency-shifte laser beam illuminates the CFPI maintaine at point A in Fig.2(b). (4) The output of the CFPI, which contains the signal, I, proportional to the velocity of the surface as illustrate in Fig.2(b), is etecte by a photo-etector. In aition, to stabilize the separation of the CFPI strictly at point A in Fig.2(b), an active control system base on the feeback control using a piezoelectric isplacer (PZD) is incorporate.

3 (a) mirror mirror to a photo-etector (b) incient laser beam h PZD Feeback Signal Transmission I 0 +I I 0 (D.C.) A I 0 -I Frequency of light f 0 -f f 0 +f Fig.2 (a) Basic arrangement of the CFPI an (b) an example of a spectral response curve f 0 Discussion: It is well known that surface wave travels only through the surface layer which is as thin as one wavelength of itself. Therefore, most of R-wave having higher frequency is reflecte, elaye an moe-converte by the etaile geometry of a small crack. The lower frequency R-wave penetrating eeper layer is not so sensitive to the geometry; therefore, it is easier to travel over a complicate shape crack to the other sie. Since the laser ultrasonic technique allows generation an etection of wie frequency ban R-wave, it woul be a suitable tool for analyzing frequency response of a crack by comparing incient, reflecte an transmitte R-wave. Figure 3 compares the observe waveforms transmitte from cracks having epths of from 0.2 mm to 1.5 mm. The pulsewith of transmitte R-waves tens to be wier with an increase of crack epth. This result shows that a surface slit behaves as a low pass filter (LPF) to the broaban R-wave. The cut-off frequency of the LPF epens on the slit epth. To etermine the transfer characteristics of slits having a epth of, G, the 50th orere moving-average (MA) moel, 50 T ( t) = G ( m) I ( t m), m= 0 is applie. Here I(t) an T(t) are the incient an transmitte SAW signal, respectively. The response times of calculate transfer functions, G, are shown in Fig.3.

4 Crack epth (measure) (mm) Stress corrosion cracking EDM slot Crack epth (nominal) (mm) Fig.3 Crack epth measure by laser-inuce surface waves Conclusions: We have reporte that a laser-ultrasonic system couple with signal processing of frequency analysis is capable of proviing valuable information on actual crack inspections. The system provies very accurate epth of shallow cracks. It shoul be note that these excellent results were le by several features of laser-ultrasonics. Laser-ultrasonics is not only a technique of interest for the non-contact an/or remote inspection but also offers many other attractive features, such as: (1) ultrasonic far-fiel is easily obtaine, (2) wie banwith an wiely iverging ultrasoun can be use, (3) lower frequency ultrasoun is available even when the generation spot is very small whereas it is unavailable using piezoelectric transucers, (4) small laser spots allow higher spatial an temporal resolution, an (5) complex shapes, which one often encountere in inustrial inspection, can be scanne easily. It shoul be mentione that there are still a few outstaning issues respecting the laser-ultrasonic NDT. The technique base on the frequency analysis of R-wave, at least in the case that the present specification is use, cannot measure the epth eeper than a few mm. The penetration epth of R-wave is limite within onewavelength of itself; it therefore means that very low frequency, e.g. 100 khz or less, shoul be use to measure eeper epth. References: [1] Scruby, C. et al., Laser-ultrasonics: techniques an applications, Aam Hilger, Bristol, UK (1990).

5 [2] Monchalin, J. -P., Review of Progress in Quantitative Nonestructive Evaluation, 12A, 495, Plenum, NY (1993). [3] Cooper, J. A. et al., IEEE Trans. UFFC, UFFC-33, 462 (1986) [4] Suh, D. M. et al., Journal of Nonestructive Evaluation, 14, 4, 201, (1995) [5] Monchalin, J. -P. et al., Avance Performance Materials, Kluwer Acaemic Publishers, 5, 7 (1998) [6] Ochiai, M., et al., J. At. Energy Soc. Japan, 43, 3 (2001) (in Japanese) [7] Schmit-Uhlig, T., et al., Eur. Phys. J. AP., 9, 3, 235 (2000) [8] Monchalin, J.P., et al., IEEE Trans. Ultrason. Ferroelectr. Frequency Contr., UFFC-33, 485 (1986)

AN ACTIVELY-STABILIZED FIBER-OPTIC INTERFEROMETER FOR

AN ACTIVELY-STABILIZED FIBER-OPTIC INTERFEROMETER FOR LASER-ULTRASONIC FLAW DETECTION S.G. Pierce, R.E. Corbett*, and RJ. Dewhurst Department of Instrumentation and Analytical Science UMIST P.O. Box 88