|NDT.net July 2005 Vol. 10 No.7|
Non-destructive testing of sheet product inner surfaces using longtitudinal surface acoustic wavesS. Sajauskas, A. Valinevičius, L. Mieutavičiute
Department of electronics engineering, Kaunas University of Technology
Student u 50, LT-51368 Kaunas
*Corresponding Author Contact:
Email: email@example.com, Internet: www.ktu.lt/ultra/
AbstractThe paper presents experimental investigation results of nondestructive ultrasonic testing using secondary longitudinal surface waves (LSAW II). Analysis of real LSAW signals, generated and received by angle transducers, was performed and causes of lateral interference were investigated, optimal angle transducer constructions were offered. Real angle transducer parameters and levels of LSAW II signals, reflected from artificial defects in duralumin sample, were evaluated. It was concluded by experimental investigations that LSAW II signal magnitude and collateral interference level ratio reaches 24 dB. This gives opportunity to perform sufficiently reliable shell product inner surface nondestructive testing using secondary longitudinal surface waves
IntroductionExcitation of secondary longitudinal surface acoustic waves (LSAW II) on the opposite side of the plate during propagation of primary LSAW (LSAW I) [2, 3] on the first surface is one of the exceptional features of LSAW [1-3]. That permits analysis and detection of surface defects located on inaccessible inner surface of the sheet products, such as tanks, reservoirs, boilers and also nuclear reactor components [4, 5], using secondary LSAW. There are some reservations about application possibilities of the secondary LSAW in non-destructive tests under real conditions, considering natural attenuation of LSAW, highly exceeding attenuation of Rayleigh waves. The purpose of this work was to investigate experimentally the parameters of the real secondary LSAW signals and to determine the potential limits for shell product inner surface analysis.
Investigation methodology and equipmentIt was experimentally determined that LSAW are excited most effectively using variable angle prismatic transducers [6, 7]. The earlier described digital ultrasonic defectoscope was used in experiments; it incorporates 2-4 MHz variable angle prismatic piezoelectric transducers . The secondary LSAW (LSAW II) excitation and investigation scheme for this case is shown in Fig. 1.
The angle emitter matched for the first critical angle JIcr is excited using the pulse generator and it generates LSAW I in the test sample. Bulk transversal waves (BTW) propagate and at the same time generate LSAW II on the opposite side. During investigation these waves can be picked up by the angle receiver placed on the inner surface of the product. Notches of 1 mm width and of the different depth (from 2 to 4 mm) are made on the inner surface in order to simulate the defects. The receiver placed on the bottom position is turned around by 180 ° (Fig. 1, dashed line) in order to receive the LSAW II, reflected from the notch. Features of LSAW signal pick-up using angle transducer Reception of the LSAW signals by the variable angle transducer is associated with a series of problems. One of them is the duplication of the acoustic pulse excited by a piezoelement generating the plane bulk longitudinal wave (BLW). Duplication of the acoustic pulse is due to the small LSAW excitation angle
where c'L and c''L are velocities of bulk longitudinal wave in the angle transducer prism and the sample respectively. For example, the LSAW is generated in duralumin ( c''L =6320 m/s) products using the transducer with the prism, manufactured of plexi glass (c'L =2730 m/s), JIcr=25.6 °. In the case when the dimension of the piezoelement generating the plane BLW d >> lL , where lL is the length of the longitudinal waves in the prism, conditions may arise for excitation of piezoelement repeatedly by the reflected BLW pulse (Fig. 2). Back reflexion range
The LSAW pulses of an identical shape are generated because of repeated excitation of the piezoelement by the BLW pulse, reflected from the surface of the sample (Fig. 3). Besides, the magnitude of the second LSAW pulse may be sometimes even higher than of the first, although it is created by the pulse reflected from the surface of the test sample, consequently weaker than the first BLW pulse. This paradox is explained in the following way. First, when piezoelement is excited by the electric pulse the it emits the acoustic pulse, shape of which is determined by piezoelement pulse response. Thus, shape of the BLW pulse reflected from the surface ideally matches the pulse response of the piezoelement; for this reason efficiency of the repeated excitation (acoustic) is maximal. Second, if the transducer incidence angle J is not exactly equal to JIcr, then the LSAW II pulse may be emitted at the angle, closer to JIcr, therefore, it may be stronger than the first one.
The second pulse of the BLW signal may be eliminated increasing piezoelement distance from the radiating prism plane, using standard double transducers with an acoustic delay and an additional prism (Fig. 4). However, in such case transducers sensitivity would decrease and transducers dimensions would increase due to the increased BLW attenuation in the prism.
Another lateral effect, natural to investigations of LSAW, is associated with back radiation (pick-up) of the angle type transducer. It was found experimentally that the angle type transducer radiates (receives) LSAW not only in the forward direction, but also in the reverse direction. The measured ratio of transducer sensitivities in the forward and the reverse directions is 23.5 dB. Collateral forward LSAW pulses can be seen on oscilloscope display due to reception in the reverse direction, when the LSAW signals, reflected from the defect, are received using the experimental set-up, shown in Fig. 1. They are observed in the time domain earlier than the LSAW pulses reflected from the defect, which have traveled a longer distance (Fig. 5).
Especially negative factor is that collateral pulses are the first on the time axis. Interpretation of a complex signal can be complicated and may lead to non-destructive testing errors, if collateral reverse reception signals are significant in a magnitude. In such cases unwanted reverse reception signals can be eliminated using the angle transducer with two identical piezoelement (Fig. 6) and introducing the reception channel, using electronic gate blocking for a time interval
where Dt1L and Dt2Lare respectively the first piezoelement (emitter) radiated and the second piezoelement (receiver) received BLW pulse delay times in the prism; Dt1T and Dt2T are respectively the first and the second piezo-crystal BTW pulse delay times in the sample; DtIIPIB is the delay time of the LSAW II signals, when they propagate between the points A and B.
Results of experimental investigationsSet-up used for experimental investigations is shown in Fig. 1. After determining the value of the first critical angle JIcr=25,6°, two identical fixed angle transducers were made where the piezoelement are acoustically damped. The transducer maximum sensitivity frequency is 2,5 MHz. LSAW-II signal excited by one of these transducers and received by another on is shown in Fig. 7. Fig.7. LSAW II signal received by the angle transducer (1) at 2.5MHz, propagating in the forward direction.
When pick-up transducer is turned around by 180 °, the LSAW II signals reflected from the artificial defect (1 mm width, h=2 mm depth notch) are registered by it, and the different distances from the defect are selected (Fig. 8). The first two lower magnitude pulses are collateral LSAW II signals, propagating in the direction of the defect and received by the reverse side of the angle transducer. When the transducer is moved closer to the defect, the LSAW II signal reflected from it moves along sweep origin direction, and when L=0 its magnitude is maximal (Fig 8, b). When the transducer is behind the defect, there is no LSAW II signal, reflected from the defect; only duplicated collateral pulses are left (Fig. 8, d).
When the inner surfaces of sheet type products are investigated using LSAW II, these waves have to be excited and received up from the outer surface side by the same angle transducer or using the double transducer (Fig. 4, 6). Scheme of such investigation is shown in Fig. 9, and the LSAW II signal reflected from the defect in Fig. 10.
In this case there are no signals received by the reverse side of pick-up element. Although LSAW II signal and the hardware interference ratio (16...24 dB) is less than during investigation using two separate transducers (~40 dB), it is still sufficiently large to register the defect reliably. In the sheet sample with the thickness D =10 mm, due to multiple signal reflections between the walls and due to small BTW attenuation in duralumin, the received acoustic signal is strong and consists of long series of LSAW II waves, reflected from the defect (Fig. 11). This indicates, that it could be possible to use higher than 2.5 MHz frequency angle transducers for sheet products, which would allow to increase the investigation accuracy and to detect smaller defects.
It should be noticed, that only a part of LSAW II energy is reflected from 2 mm depth notch (defect), and the magnitude of the received LSAW II signal is significantly greater (~3 dB) when defect is deeper (h=4 mm) (Fig. 11, b). That confirms LSAW property to propagate in a near-surface layer.
Since the angle type transducer radiates directional LSAW, it is naturally that the magnitude of the signal reflected from a regular shape defect (smooth notch) depends on the defect orientation with respect to the LSAW propagation direction. Therefore, detection probability of defects of such nature is higher for the defects, orientation of which can be foreseen (e. g., for defects along welding juncture). The described investigations were performed by the angle transducer, directional characteristic width of which in the plane of sample surface measured at the 3 dB level was 3...40.