Measurement of ultrasonic scattering attenuation in austenitic stainless steel welds: Realistic input data for NDT numerical modeling

Multipass welds made of 316L stainless steel are specific welds of the primary circuit of pressurized water reactors in nuclear power plants. Because of their strong heterogeneous and anisotropic nature due to grain growth during solidification, ultrasonic waves may be greatly deviated, split and attenuated. Thus, ultrasonic assessment of the structural integrity of such welds is quite complicated. Numerical codes exist that simulate ultrasonic propagation through such structures, but they require precise and realistic input data, as attenuation coefficients. This paper presents rigorous measurements of attenuation in austenitic weld as a function of grain orientation. In fact attenuation is here mainly caused by grain scattering. Measurements are based on the decomposition of experimental beams into plane-wave angular spectra and on the modeling of the ultrasonic propagation through the material. For this, the transmission coefficients are calculated for any incident plane wave on an anisotropic plate. Two different hypotheses on the welded material are tested: first it is considered as monoclinic, and then as triclinic. Results are analyzed, and validated through comparison to theoretical predictions of related literature. They underline the great importance of well-describing the anisotropic structure of austenitic welds for UT modeling issues.     

Grain fragmentation in ultrasonic-assisted TIG weld of pure aluminum

Under the action of acoustic waves during an ultrasonic-assisted tungsten inert gas (TIG) welding process, a grain of a TIG weld of aluminum alloy is refined by nucleation and grain fragmentation. Herein, effects of ultrasound on grain fragmentation in the TIG weld of aluminum alloy are investigated via systematic welding experiments of pure aluminum. First, experiments involving continuous and fixed-position welding are performed, which demonstrate that ultrasound can break the grain of the TIG weld of pure aluminum. The microstructural characteristics of an ultrasonic-assisted TIG weld fabricated by fixed-position welding are analyzed. The microstructure is found to transform from plane crystal, columnar crystal, and uniform equiaxed crystal into plane crystal, deformed columnar crystal, and nonuniform equiaxed crystal after application of ultrasound. Second, factors influencing ultrasonic grain fragmentation are investigated. The ultrasonic amplitude and welding current are found to have a considerable effect on grain fragmentation. The degree of fragmentation first increases and then decreases with an increase in ultrasonic amplitude, and it increases with an increase in welding current. Measurement results of the vibration of the weld pool show that the degree of grain fragmentation is related to the intensity of acoustic nonlinearity in the weld pool. The greater the intensity of acoustic nonlinearity, the greater is the degree of grain fragmentation. Finally, the mechanism of ultrasonic grain fragmentation in the TIG weld of pure aluminum is discussed. A finite element simulation is used to simulate the acoustic pressure and flow in the weld pool. The acoustic pressure in the weld pool exceeds the cavitation threshold, and cavitation bubbles are generated. The flow velocity in the weld pool does not change noticeably after application of ultrasound. It is concluded that the high-pressure conditions induced during the occurrence of cavitation, lead to grain fragmentation in a pure aluminum TIG weld during an ultrasonic-assisted TIG welding process.     

Ultrasonic reflection properties of planar defects within austenitic welds

Reflection of ultrasonic beams from the faces of planar weld defects is affected by the anisotropic nature of the weld material. The simple laws of geometrical reflection no longer hold and energy may be strongly reflected into unexpected directions. This has important consequences for ultrasonic inspection techniques relying on specular signals for defect detection. In this Paper a ray tracing approach is used to study, theoretically, the expected patterns of defect reflection behaviour for several weld types and several defect locations and orientations. In general we find that when the ultrasonic beam is generated in the austenitic weld material without first passing through the base metal then the strongly reflected specular signal travels in very unexpected directions. If, however, the ultrasonic beam travels into the weld metal via the base metal then reflected signals usually occur close to the expected directions, except for vertically polarized shear waves. Furthermore, for a given weld type it is obviously useful to minimize ultrasonic ray paths within the weld material, to minimize both attenuation and curvature of ray paths.     

考虑晶粒分布的多晶体材料超声散射统一理论

现有超声散射统一理论可通过多晶体材料的微观结构和力学特性,实现全频域范围内衰减和相速度的正演建模,但其忽略晶粒尺寸分布的影响,进而降低了正演模型的计算精度.本文对不均匀介质的波动方程进行二阶Keller近似,用全频域格林函数推导介质中的平均波;以截断对数正态分布描述晶粒分布,构建加权的空间相关函数;结合材料的弹性模量协方差,建立含晶粒分布的超声散射统一理论,揭示晶粒分布对超声散射的影响规律;制备304不锈钢试块并开展超声散射实验.结果表明考虑晶粒分布特性后,纵波衰减谱和相速度谱相对于实验结果的相异性降低约49%和64%,横波衰减谱和相速度谱相对于实验结果的相异性降低约12%和4%.可见,本文的统一理论模型能有效修正晶粒分布导致的衰减谱和相速度谱偏差,为晶粒分布反演评价提供理论基础.     

焊缝散射对板中超声水平剪切导波直线合成孔径阵列成像的影响

为探讨焊缝散射与板中超声导波阵列成像之间的关系,提出了将散射渡越时间矩阵作为分析工具用来开展焊缝散射对板中超声水平剪切(Shear horizontal,SH)导波直线合成孔径阵列成像影响的研究。运用散射因子和散射渡越时间矩阵描述焊缝散射,结合分析超声SH导波直线阵列信号所特有的圆锥曲线特征和合成孔径成像算法,从导波阵列信号和阵列成像机理出发,重点研究了焊缝散射因子和散射渡越时间矩阵在阵列成像中的作用。结果显示,焊缝散射易造成同一成像散射体反射声波的阵列信号形成\     

Ultrasonic characterization of shrinkage microporosity in aluminum castings

Shrinkage microporosity in cast aluminum was characterized utilizing the frequency dependence of ultrasonic attenuation caused by scattering from the pores. Measurements were made with the plate specimen immersed in water, and, by using a focused transducer, spatial resolution of about 2 mm was obtained. An accurate measure of attenuation was obtained by comparing the specimen’s ultrasonic signal with that from a pore-free reference specimen. Although the attenuation could be fitted using a single spherical pore size, better fits were obtained by assuming a lognormal distribution of spheres. Pore volume fraction inferred from the lognormal fits overestimates the actual volume fraction, determined from density measurements, by the same factor for all volume fractions. The actual volume fraction is overestimated by more than 100%, due to the complicated, nonspherical pore shapes, and must be taken into account to obtain accurate values of porosity. The strong correlation (r²=0.97) between ultrasonic and density-derived volume fractions permits reliable, nondestructive laboratory measurements of porosity.     

Theoretical assessment of the errors involved in ultrasonic location and sizing of molten weld pools

Ultrasonic transit times may be used to locate the interface between molten weld metal and parent plate as a weld is formed. These transit times will lead to errors in interpretation of the location of the weld interface if account is not taken of the variation of ultrasonic velocity with temperature. We assess the magnitude of these errors using a theoretical model based on the Green's function to obtain the temperature distribution at any time and then convert the temperatures into elastic constants through empirical fits to high temperature data. A ray tracing method and a semi-analytical approach are used to estimate the effects of the changes in elastic constants on the ray paths and transit times. From these theoretical predictions we conclude that the location errors incurred by assuming ultrasonic velocities appropriate to cold metal are less than 0.5 mm for a 3.5 mm radius weld in austenitic steel or in iron. Experimental values tend to be larger than this, possibly suggesting that the solid-liquid interface is not as perfect as in the theoretical model. Under all of the conditions investigated, predicted errors are smaller with compression waves than with shear waves.     

The propagation of ultrasound in an austenitic weld

The propagation of ultrasound through an austenitic weld is investigated experimentally as well as in a numerical simulation. The weld is insonified at normal incidence to the fusion line with a longitudinal contact transducer. In order to experimentally trace the ultrasound through the weld, slices of different thicknesses from the original weld have been fabricated. Through-transmission A-scans have then been produced for each weld slice and compared with the corresponding numerical simulation. A comparison of the direction of ultrasound propagation through the weld for the two approaches shows quite good agreement. However, attenuation due to scattering at grain boundaries in the weld is poorly modelled in the simulation. In order to improve this, a better model of the weld is needed.     

Polydisperse particle size characterization by ultrasonic attenuation spectroscopy in the micrometer range

The theoretical advantages of ultrasonic attenuation spectroscopy for particle size are currently not fully utilized. Especially in the region of larger particles, there is a lack of experimental confirmation of applicable models which may be used to infer particle sizes from measured attenuation spectra. With the present work, an attempt is made to supply experimental data, obtained with a commercially available ultrasonic attenuation spectrometer, and model calculations, which are based on the resonant scattering theory. It is shown that measured attenuation results for various combinations of disperse and continuous phase for both polydisperse emulsions and suspensions are reproducible by calculation. The approach is further examined for suspensions of porous particles. Here, the resonant scattering approach is combined with the Biot model for poroelasticity to obtain attenuation results with several fractions of titania aggregates, differing in particle size and pore diameter. The results indicate that the theory of resonant scattering is a valid approach if applied to particle size characterization in the large particle limit.     

Quantitative evaluation of ultrasonic C-scan image in acoustically homogeneous and layered anisotropic materials using three dimensional ray tracing method

Quantitative evaluation of ultrasonic C-scan images in homogeneous and layered anisotropic austenitic materials is of general importance for understanding the influence of anisotropy on wave fields during ultrasonic non-destructive testing and evaluation of these materials. In this contribution, a three dimensional ray tracing method is presented for evaluating ultrasonic C-scan images quantitatively in general homogeneous and layered anisotropic austenitic materials. The directivity of the ultrasonic ray source in general homogeneous columnar grained anisotropic austenitic steel material (including layback orientation) is obtained in three dimensions based on Lamb’s reciprocity theorem. As a prerequisite for ray tracing model, the problem of ultrasonic ray energy reflection and transmission coefficients at an interface between (a) isotropic base material and anisotropic austenitic weld material (including layback orientation), (b) two adjacent anisotropic weld metals and (c) anisotropic weld metal and isotropic base material is solved in three dimensions. The influence of columnar grain orientation and layback orientation on ultrasonic C-scan image is quantitatively analyzed in the context of ultrasonic testing of homogeneous and layered austenitic steel materials. The presented quantitative results provide valuable information during ultrasonic characterization of homogeneous and layered anisotropic austenitic steel materials.     

声速和衰减系数与含蜡原油特性的关系

当含蜡原油的温度低于蜡的析出温度后,蜡晶将不断析出并相互交联成网状结构,若网状结构达到一定强度,原油将失去流动而凝结。在该过程中不仅含蜡原油的流动特性发生变化,原油的声速和衰减系数也发生变化。本研究结果表明：含蜡原油的超速及衰减系数均随温度降低而增大;含蜡原油在凝结温度前后表现出不同的声速与温度的对应关系;含蜡原油在凝结后的声速的温度的关系同原油的含蜡量有关;含蜡原油的衰减系数在接近凝结温度时急剧     

The effect of grain size distribution on the frequency dependence of the ultrasonic attenuation in polycrystalline materials

An expression has been derived for the effect of the grain size distribution in polycrystalline materials on the frequency dependence of the ultrasonic attenuation. It has been shown that two specimens with the same mean grain size can have significantly different ultrasonic attenuations if their grain size distributions are different and that no unique solution in terms of the grain size exists for a particular ultrasonic attenuation against frequency curve. Qualitative agreement has been found between the theory and some of the experimental data available.     

Studies of sound transmission in various types of stored grain for acoustic detection of insects

In developing passive acoustic systems for detecting insect infestations in bulk-stored grain, it is advantageous to understand the transmission of sound in the grain between the insects and the sensors. In the work presented here grain is shown to be a strong acoustical absorber, the attenuation coefficient increasing roughly as the square root of the frequency. Tests with soft wheat immersed in three different gases: air, argon and carbon dioxide, support an earlier conclusion that sound is transmitted principally through the gas in the passageways between the grain kernels. The speed of sound and the attenuation coefficient were measured as a function of frequency for six different types of grain: hard and soft wheat, brown rice, soybeans, corn and sorghum. It was determined that sound is transmitted over longer distances in grains with a larger inter-kernel spacing, such as corn and soybeans. Grain depth, up to several meters, appears to have little effect on sound transmission.     

Shape effect of elongated grains on ultrasonic attenuation in polycrystalline materials

Longitudinal and transverse wave attenuation coefficients are obtained in a simple integral form for ultrasonic waves in cubic polycrystalline materials with elongated grains. Dependences of attenuation on frequency and grain shape are described in detail. The explicit analytical solutions for ellipsoidal grains in the Rayleigh and stochastic frequency limits are given for a wave propagating in an arbitrary direction relative to ellipsoid axes. The attenuation exhibits classic frequency dependence in those frequency limits. However, the dependence on the grain shape in the stochastic limits is unexpected: it is independent of the cross-section of the ellipsoidal grains and depends only on the grain dimension in the propagation direction. In the Rayleigh region attenuation is proportional to effective volume of the ellipsoidal grain and is independent of its shape. A complex behavior of attenuation on the grain shape/size and frequency is exhibited in the transition region. The results obtained reduce to the classic dependences of attenuation on parameters for polycrystals with equiaxed grains.     

The relationship between collagen and ultrasonic attenuation in myocardial tissue.

The relationship between ultrasonic attenuation and collagen content is examined in hearts from normal dogs and in hearts from dogs subjected to ischemic injury by coronary occlusion as an approach toward elucidating the physical mechanisms responsible for the attenuation of soft tissue. Increased ultrasonic attenuation is shown to correlate well with increased collagen concentration determined biochemically in regions of ischemic injury studied 2, 4, and 6 weeks following occlusion. Extrapolation of experimentally determined relationship between attenuation and collagen concentration suggests that collagen is responsible for not more than 15% of the attenuation observed in normal myocardium. These results indicate that collagen appears to be the principal determinant of the elevated attenuation detected in regions of myocardial infarction, but is apparently not the primary determinant of the attenuation of normal myocardium.     

Enhanced experimental approach to measure the absolute ultrasonic wave attenuation

The absolute ultrasonic wave attenuation is an important input parameter for mathematical models, which play an increasingly important role in non-destructive testing. The measurement of the absolute ultrasonic wave attenuation however is a difficult task. When conventional measurement techniques are applied, corrections to the raw data are required to account for apparent losses. In this study, a modified experimental approach is proposed to determine the absolute ultrasonic attenuation without any further corrections of the raw data.     

铸造奥氏体不锈钢中铁素体与奥氏体位向关系及其对声衰减的影响

研究了铸造奥氏体不锈钢中铁素体与奥氏体位向关系及其对超声散射衰减的影响.利用电子背散射衍射技术表征了两相的晶体取向及其位向关系,基于真实的铁素体形貌建立了二维声传播各向异性模型并利用时域有限差分法进行了计算,分析了不同位向关系、铁素体形貌特征对声衰减系数的影响规律并进行了实验验证.结果表明：铸造奥氏体不锈钢奥氏体晶粒中散布着形状复杂的铁素体,典型铁素体形貌为条状和岛状;铁素体与奥氏体的位向关系以Kurdjumov-Sachs关系为主,少量满足Nishiyama-Wassermann关系.对声传播过程进行计算,发现两相位向关系和铁素体形貌协同作用影响超声波传播,在较高检测频率（15 MHz）下对散射衰减的影响不能忽略.结合“原位”实验对奥氏体<101>柱状晶粒的声衰减影响因素进行了定量分析,发现对于单一铸造奥氏体晶粒,晶粒内部取向不均匀性、奥氏体-铁素体位向关系以及奥氏体晶粒内铁素体形态都是超声散射衰减的主要原因.     

Ultrasonic inspection of stainless steel butt welds using horizontally polarized shear waves

Inspection of austenitic stainless steel weldments by conventional ultrasonic means is fundamentally limited by the textured, columnar grain structure of the weld metal. It is shown that, for selected angles of incidence, shear waves normally polarized with respect to the columnar grains can pass through the weld metal-base metal interface without partial reflection. As a consequence, the inspectability of stainless steel weldments can be improved. The operation of a low frequency, ultrasonic system for stainless steel butt weldments is demonstrated.     

Analysis of ultrasonic attenuation in particle-reinforced plastics by a differential scheme

Attenuation of ultrasonic longitudinal waves in some particle-reinforced polymer composites is studied theoretically by a micromechanical model based on a differential (incremental) scheme. A set of differential equations is established by which the attenuation spectrum of the composite can be computed from the known properties of viscoelastic matrix and elastic particles. For a composite reinforced with glass particles with radius 0.15 mm, the proposed scheme is shown to predict the attenuation in better agreement with the foregoing experimental results than the previous simplistic independent scattering model. Based on this scheme, the dependence of the longitudinal attenuation spectrum of a particulate polymer composite on the wavelength-to-particle radius ratio and the particle volume fraction is examined in detail. It is then shown theoretically that the attenuation of the composite decreases monotonically with the particle volume fraction when the particle radius is sufficiently small compared to the incident wavelength, while it shows non-monotonic particle-fraction dependence when the ratio of the particle radius to the wavelength is larger. To examine this theoretical finding from an experimental point of view, the longitudinal attenuation in a glass-particle-reinforced polyester composite with particle radius 0.0225 mm is measured for different particle volume fractions. The measured attenuation characteristics are shown to support the qualitative features of the theoretical prediction.     

An investigation of the measurement of broadband ultrasonic attenuation in trabecular bone

Commercial devices for the ultrasonic characterisation of bone normally report the broadband ultrasonic attenuation (BUA). This is the slope of the attenuation against frequency in some part of the frequency range 200–1000 kHz. The assumption is that the relationship is linear and hence independent of the frequency range selected. In this study the ultrasonic attenuation in the frequency range 200 to 800 kHz was measured with a variety of transducers in ten trabecular heel bone samples from elderly cadavers, assumed to be osteoporotic.

The results indicate that the attenuation fits better to a second order polynomial function of frequency, than to the linear fit. The use of a straight line fit is only satisfactory in the higher frequency ranges (above 400 kHz). The use of lower frequencies results in a significant measurement error caused by the combination of a poor signal to noise ratio and the departure from linearity and this is greatest for samples with low attenuation. In the worst cases this can amount to a 30% discrepancy between the BUA values measured over different frequency ranges.