 · Home· Table of Contents · Fundamental & Applied Research | A study on the couplant effects in contact ultrasonic testing.Young H. Kim and Sung-Jin SongSchool of Mechanical Engineering, Sungkyunkwan University, Suwon, Korea Sung-Sik Lee and Jeong-Ki Lee Korea Inspection & Engineering Co., Ltd., Seoul, Korea Soon-Shin Hong Korea Advanced Inspection Technology, Taejon, Korea. Contact |
Keywords: Ultrasonics, Contact method, Couplant, Glycerine, Apparent Attenuation Coefficient
The amplitude of the back-wall echoes depend on the reflection coefficient on the interface between transducer and test material in the contact pulse-echo ultrasonic testing. The couplant is used to transmit the ultrasonic energy through the interface, and has an influence on the amplitude of the pulse-echo signal. To investigate the couplant effect on the pulse-echo ultrasonic testing, the back-wall echoes are measured by using various couplants made of water and glycerine in a carbon and an austenitic stainless steel specimens. The amplitude of the first back-wall echo and the apparent attenuation coefficient increase with the acoustic impedance of couplant. The couplant having higher value of transmission coefficient is more effective for flaws detection, and the reflection coefficient should be known to measure the attenuation coefficient of test material.
The ultrasonic pulse echo technique has been widely used for detecting flaws in materials and measuring ultrasonic velocity and attenuation coefficient which are the bases for evaluating elastic moduli, characterizing microstructure and assessing mechanical properties (Store 1991, pp. 263-265). This technique uses a broad band transducer contacting with the test material at normal incidence, and the liquid couplants such as water, glycerine, grease and so on, make the ultrasonic energy better to transmit through the interface between the transducer and test material. The couplant is appropriately selected by the testing condition or environment, for example, water is not appropriate to the inspection of carbon steel due to corrosion, and viscous couplants may be required to inspect the inclined place (Golis 1992, pp. 13-18).
The couplant would form a layer between a transducer and a test material. This layer causes the error in the measurement of ultrasonic velocity and attenuation. There were few studies on the couplant effect in the measurement of ultrasonic velocity by pulse echo technique (Vincent 1987). The phase and amplitude of the back-wall echoes is varied by the wavelength and the thickness of layer formed by the couplant and wear plate of transducer. Thus, the same error would be comprised in the ultrasonic velocity and attenuation coefficient measured by the contact ultrasonic testing. Although it has been considering the couplant effect in the ultrasonic velocity measurement, there are few studies on the couplant effect in the ultrasonic attenuation measurement and the flaw detection.
In the present work, it is investigated the amplitudes of the first back-wall-echo and the apparent attenuation coefficient of carbon and stainless steel with varying the characteristic acoustic impedance of couplants made of water and glycerine, in which the thickness of the couplant is constantly fixed by using the aluminium foil.
| (1) |
| (2) |
where Z1 is the characteristic acoustic impedance of medium in which incident and reflected wave are travelling, and Z2 is that of medium in which the transmitted wave is travelling. In contact ultrasonic testing, Eqs. (1) and (2) are generally used to obtain intensity reflection and transmission coefficients between a transducer and a test material. Since a thin layer is formed by using the couplant such as glycerine or grease between the transducer and the test material, the intensity reflection and transmission coefficients become different from Eqs. (l) and (2) in real contact testing. As shown in Figure 1, when a layer of finite thickness is formed between two media, an ultrasonic wave normally incident on the interface from medium 1 generates reflected waves and transmitted waves into the layer. The transmitted wave into the layer is reflected again at the rear interface and transmitted into medium 3. In this case, the intensity transmission coefficient, TI is represented as follows:
| (3) |
Fig 1: Transmission and reflection of plane waves normally incident on a layer..
|
where Z1, Z2, Z3 are the characteristic acoustic impedances of three media, respectively, k2 the wave number in the layer, and d the layer thickness. The intensity transmission coefficient depends on the ratio of the layer thickness to the ultrasonic wavelength, which results from the interference due to the multiple reflections at the two interfaces of the layer (Rose and Mayer 1974).
If the layer thickness is very thin compared with wavelength, i.e. k2d<<1, then cosk2d»1 and sink2d»k2d and also if the acoustic impedance of the layer is smaller than that of media 1, 3, i.e. Z2 < Z1, Z3, then Eq. (3) simplifies to
| (4) |
where A and B are constant given by the following.
|
When the ultrasonic testing is performed by using a normal beam longitudinal mode transducer which is directly contacted with the surface of test material using couplants such as glycerine or grease, the couplant forms a very thin layer much smaller than the wavelength and its characteristic acoustic impedance is smaller than those of transducer and test material. Since the acoustic impedances of the transducer and test material are fixed in the ultrasonic testing, the intensity of the transmitted wave is influenced by the wavelength of transducer, the thickness and the characteristic acoustic impedance of the couplant being used. Eq. (4) shows that the smaller thickness of a layer, the stronger intensity of a transmitted wave.
2.2 Couplant Effect on The Back-wall Echoes and Attenuation Coefficient
The amplitude of ultrasonic wave decreases as the propagating distance increases. The amplitude of ultrasonic wave which has propagated the distance of x is represented as
| V(x)=V0 e -ax | (5) |
where, a is the attenuation coefficient of a material.
To measure the attenuation coefficient of material, the amplitude of back wall echoes reflected from the bottom surface of a specimen is compared. As shown in Fig. 2, the wave incident on a front surface of specimen is reflected at the bottom of specimen contacting with air, and the reflected wave returns into the transducer, and some of it reflect again at the boundary between the transducer and test material and propagate into the specimen. Thus the back-wall echoes will appear on CRT screen of ultrasonic testing equipment and the amplitude of those gradually decrease as transit time corresponding to the propagating distance increases. If the back-wall echoes were measured by the contact pulse-echo method in the specimen of the thickness h, the amplitude of back-wall echoes were influenced by the thin couplant layer, as referred in the preceding as well as the material attenuation. Since the intensity, I, is related to the amplitude, V, as IµV2, the intensities of the first and second back-wall echoes can be represented as follows,
| I1 = I0T21e-4ah I2 = I0T21R1e-8ah | (6) |
where, TI and RI are the intensity transmission and reflection coefficients in Eq.(4). We assumed that intensity reflection coefficient at interface between specimen and air is 1, and neglected the diffraction effect and attenuation in the layer formed by couplant.
Fig 2: back-wall echoes from a plate.
|
In contact ultrasonic testing, the attenuation coefficient is generally obtained by the ratio of the amplitudes of the first back-wall echo to that of the second back-wall echo. It is called the apparent attenuation coefficient, which is presented in ASTM E 664-78 as follows,
| (7) |
in which the couplant effect is neglected. Because the thin layer is formed by the viscous couplant or the rough surface of specimen in contact ultrasonic testing, the amplitude of the back-wall echoes is influenced by the transmission and the reflection coefficients as represented in Eq. (6). Briks et al (1991, pp.377-390) proposed the attenuation coefficient should be revised as follows:
| (8) |
Substituting Eq.(8) into Eq.(7), the apparent attenuation coefficient is related to the real attenuation coefficient of materials as follows:
| (9) |
Since the intensity reflection coefficient is less than 1, log RI is negative. Therefore, the apparent attenuation coefficient is larger than the real attenuation coefficient of material.
| couplant | Water: glycerine | density(kg/m3) | velocity(m/s) | impedance (x 106 kg/m2 s) |
| Cl | 10:0 | 998 | 1490 | 1.49 |
| C2 | 9: 1 | 1018 | 1533 | 1.56 |
| C3 | 8: 2 | 1039 | 1586 | 1.65 |
| C4 | 7: 3 | 1065 | 1637 | 1.74 |
| CS | 6: 4 | 1089 | 1685 | 1.84 |
| C6 | 5: 5 | 1113 | 1739 | 1.94 |
| C? | 4: 6 | 1141 | 1790 | 2.04 |
| C8 | 3: 7 | 1173 | 1836 | 2.15 |
| C9 | 2: 8 | 1195 | 1863 | 2.23 |
| ClO | 1: 9 | 1224 | 1895 | 2.32 |
| Cl1 | 0:10 | 1260 | 1920 | 2.42 |
| Table 1: couplants and their acoustic impedance | ||||
In the contact pulse-echo ultrasonic testing using the normal beam longitudinal mode transducer, 11 kinds of couplants made of water and glycerine are prepared for investigation of the couplant effects. The densities and the ultrasonic velocities of the used couplants are measured and the characteristic acoustic impedances of those are calculated with the measured densities and ultrasonic velocities. They are listed in Table 1. The longitudinal wave velocities measured by the pulse-echo overlap method are 5,960 m/s for carbon steel and 5,600 m/s for stainless steel. The measurement of the echo signals in test materials are performed with the normal beam longitudinal mode transducer of diameter 12.7 mm and center frequency 5 MHz (Ultran WC 50-5). The transducer was directly contacted with the surface of test material using the prepared couplants. An ultrasonic pulse generator and receiver (JSR PR-35) was used to excite the transducer and receive the pulse echo signals. The measured signal is sent to the digital oscilloscope (Lecroy 9410) set up to display 4 back-wall echoes on CRT screen. As the thickness of couplant may be varied due to its viscosity, we have made it constant by using an aluminium foil (thickness 16m m) between test material and transducer.
Figures 3 and 4 show the back-wall echoes on CRT screen of ultrasonic flaw detector using pure water and glycerine couplants in the carbon steel. In order to keep first back-wall echo on 80% of full screen, pure water couplant requires 5 dB higher gain compared to the case of glycerine couplant. In addition, decreasing rate of the back-wall echoes are different with each other and the second back-wall echo using glycerine decreases more rapidly than that using water. The amplitudes of the first echo in carbon steel and stainless steel versus the impedance of couplants is shown in Figure 5. The amplitudes of the first echo using the pure glycerine is about three times as large as that using pure water. This means that amplitude and decreasing rate of the back-wall echoes are influenced by the couplant. As represented by Eq.(6), the intensity of the first echo increases with the transmission coefficient, and that of the second echo and other consecutive echoes decrease as the transmission coefficient increase. The intensity transmission coefficient increases with the characteristic acoustic impedance as represented in Eq. (4). The amplitude of the first back-wall echo in the carbon steel is larger than that in the stainless steel, because the attenuation coefficient of stainless steel is larger than that of carbon steel.
Fig 3: Back-wall echoes using water couplant in the carbon steel.
| 
Fig 4: Back-wall echoes using glycerine couplant in the carbon steel.
|
The apparent attenuation coefficient, a ', obtained from the amplitude variation of the back-wall echoes versus the characteristic acoustic impedance of couplants is shown in Figure 6. The apparent attenuation coefficient is increased with acoustic impedance of couplant and those using the pure glycerine are about two times as large as that using the pure water. As represented in Eq.(9), the apparent attenuation coefficient is influenced by the intensity reflection coefficient at the boundary of the transducer and test material.
Since the intensity reflection and the transmission coefficients have the complementary relation, the increasing intensity transmission coefficient causes the amplitude of the first back-wall echo and the apparent attenuation coefficient to increase. To be the better flaw detection, the amplitude of the first back-wall echo should be large, which is caused by the increasing transmission coefficient, but the error of the attenuation coefficient increase with the transmission coefficient.
Fig 5: the amplitude of 1st back-wall echo versus the acoustic impedance of couplants.
|
Fig 6: apparent attenuation coefficients versus the acoustic impedance of couplant. |
The back-wall echoes in carbon and stainless steel are measured by using the various couplants made of water and glycerine. The amplitudes of the first back-wall echo and the apparent attenuation coefficients increased with the characteristic acoustic impedance of the couplants. The amplitude of the first echo and the apparent attenuation coefficient measured by using pure glycerine is about three and two times, respectively, as large as that using pure water. The larger the transmission coefficient at the boundary formed between transducer and test material, the higher the sensitivity of the flaw detection becomes in the normal beam contact ultrasonic testing. Since the apparent attenuation coefficient is influenced by the reflection coefficient at the boundary between transducer and test material in the contact ultrasonic testing, the real attenuation coefficient should be corrected.
| © AINDT , created by NDT.net | |Home| |Top| |