Volume 59, Issue 6, June 1976
Index of content:
59(1976); http://dx.doi.org/10.1121/1.381012View Description Hide Description
This paper shows how to process backscattered acoustic data to directly generate the image of an anomaly in a plane such as an anticline, syncline, or fault. The result is based on a physical optics farfield inverse scattering (POFFIS) identity. This identity states that a phase‐ and range‐normalized scattering amplitude is proportional to a function which is unity in the region of the anomaly and zero outside that region.
Subject Classification: 20.30, 20.15; 40.50.
Theorem on the scattering and the absorption cross section for scattering of plane, time‐harmonic, elastic waves59(1976); http://dx.doi.org/10.1121/1.381013View Description Hide Description
In a homogeneous, isotropic, perfectly elastic medium a scattering obstacle of finite extent is present. When a plane, time‐harmonic, elastic compressional (P) or shear (S) wave is incident on it, the extinction cross section (i.e., the sum of the scattering and the absorption cross section) of the obstacle is directly related to the far‐zone amplitude of the particle displacement of the scatteredwave, observed exactly behind the obstacle. This relationship is discussed. In spite of the fact that in elastodynamic scattering the scatteredPwave as well as the scatteredswave contribute to the time‐averaged scattered power, the extinction cross section appears to depend only on the value of that farfield scattered‐wave amplitude which is of the same type (P or S) as the incident wave.
Subject Classification: 20.15, 20.30.
59(1976); http://dx.doi.org/10.1121/1.381014View Description Hide Description
An account on the development of a directional microphone system for jet‐noise‐source strength distribution measurements is given. The system selected is based on the imaging principle of a spherical concave reflector. As expected, the system is diffraction limited; a loss of resolution results. A special technique is discussed to recover the axial‐source strength distribution from the diffracted data. Sample experiments are presented to show the successful application of such a system for jet noise measurements.
Subject Classification: 28.65; 50.55; 85.62.
59(1976); http://dx.doi.org/10.1121/1.381015View Description Hide Description
Sound propagationmeasurements at 10, 20, 30, 40, and 60 kHz were made in April 1974 under the ice pack near Pt. Barrow, Alaska. Absorption coefficients were calculated for each frequency and have been reported in detail [G. R. Garrison e t a l., J. Acoust. Soc. Am. 58, 608–619 (1975)]. In April 1975, an attempt was made to furnish more accurate results at the lower frequencies by measuring the absorption at 7, 13, 20, 30, and 60 kHz. The 1975 measurements gave the following results for the absorption coefficients dB/kyd at a temperature of −1.6 °C and a salinity of 32%: 0.2±0.4 at 7.1 kHz; no data at 13 kHz because of transducer failure; 3.9±0.2 at 20 kHz; 7.2±0.3 at 30 kHz; and 13.1±0.5 at 60 kHz. These values indicate a relaxation frequency, due to MgSO4, of 36±3 kHz, which is slightly above our previous results, but still much lower than that predicted by the Schulkin–Marsh equation for a temperature of −1.6 °C.
Subject Classification: 30.20.
59(1976); http://dx.doi.org/10.1121/1.381016View Description Hide Description
The effect of spatial separation on the cross correlation between the system response functions for a two‐receiver surface‐scatter channel is investigated. The system response at the two receivers is determined for all possible pairs of frequencies. The analysis assumes relatively close separation of receivers and uses the Fresnel‐corrected model for surface scatter in the specular direction. The surface is assumed to be Gaussian, with typical size of boundary deformations small compared to typical Fresnel zones.The analysis shows that the correlation is approximately maximum for zero‐frequency separation and drops off at a rate determined by the reverberation time. For small reverberation times and small grazing angles correlation may extend over fairly large values of frequency difference. The effect of spatial separation is generally a reduction in correlation and a more rapid decrease of correlation with frequency, but this can be compensated to some extent by time shift between the two system responses. Experimental evidence is included to support the theoretical predictions presented.
Subject Classification: 30.20; 20.15; 60.20.
59(1976); http://dx.doi.org/10.1121/1.381017View Description Hide Description
Acoustic measurements of insertion loss and echo reduction at normal incidence were made for thin plates of eight materials for evaluation of their use in underwater soundtransducers in the frequency range 50–500 kHz. Materials tested included Lucite; silicone rubber RTV– 560; the commercial polyurethane products PRC–1527, CPC–19, Scotchcast 8, and Scotchcast 221; and two of these materials with talc added during curing to vary the properties. Values of longitudinal sound velocity, attenuation constant, and characteristic impedance are derived from comparison of theoretical and measuredcharacteristics.
Subject Classification: 30.30; 40.55; 55.75; 85.40.
59(1976); http://dx.doi.org/10.1121/1.381018View Description Hide Description
Acoustic wavefront fluctuations and time‐base distortions were measured on a long linear receiving array with a 2000 ft vertical aperture. The receiving hydrophones were buoyed slightly off the ocean bottom in a roughly uniform distribution over the aperture length. Continuous acoustic transmission up to 17.6 h of a hyperbolic FM signal with a frequency sweep from 350 to 450 Hz was used in each of seven experiments at ranges of 145, 250, 495, and 635 NM to measure signal time‐base distortions. Simultaneously, wavefront fluctuations were determined from changes in arrival time across the aperture of a 10‐msec 400‐Hz pulse from the same projector. Wavefront fluctuations, rotations, and distortions were calculated and compared to signal time‐base distortions measured on a narrow beam directed toward the incoming arrivals. Both RSR (define) and completely refracted path were insonified. In the total wavefront fluctuation, the rotation part was dominant. However, remarkably small standard deviations ranging from 5 to 15 ft over the entire aperture were found for wavefront rotations and distortions. The spectra of the measured time fluctuations and wavefront rotations and distortions reveal common causes, such as tides, overtides, and internal waves.
59(1976); http://dx.doi.org/10.1121/1.381008View Description Hide Description
Wavefront fluctuations, rotations, and distortions, measured for a particular arrival in the multipath set for long‐range propagation, were cross correlated with acoustic‐signal time‐base distortions for the same arrival. Although the wavefront perturbations contain higher frequency components than for the signal time‐base fluctuations, significant cross correlation amplitudes were found for time and rotation, time and distortion, and distortion and rotation cross correlations. The results demonstrate that common causes, such as tides, overtides, and internal waves, underlie both the temporal and spatial perturbations of the acoustic signal. Although these perturbations were significantly correlated in a high percentage of the tests, the degree of correlation between them was not stable over periods of days or weeks.
Subject Classification:30.20, 60.20.
59(1976); http://dx.doi.org/10.1121/1.381009View Description Hide Description
A detailed review of the experimental data for the rotational collision number Z rot for air, O2 and N2 has been made to finalize a temperature correction for the absorption due to rotational relaxation for incorporation in a draft standard on air absorption being prepared by the Sl−57 Committee on Propagation.
Subject Classification: 35.20, 35.35.
59(1976); http://dx.doi.org/10.1121/1.381010View Description Hide Description
An approximate model for determining the response functions of reflectors of known shape has been used to provide theoretical support for experimental investigations of flaw characterization. The reflection of broad‐band ultrasonic pulses from an object can result in modifications to both the frequency spectrum and phase information in the pulses. These changes are largely controlled by the parameters of the reflector, and relating these factors has been the subject of many experimental investigations. Within the fundamental approximations, the model can readily be applied to any reflector of known shape and provides a simple interpretation of what is physically occurring. It has the advantage that additional experimental factors can be included to determine their effect on the results.
Subject Classification: 35.26; 20.30; 35.80.
Ultrasonic bounded beam reflection and transmission effects at a liquid/ solid‐plate/liquid interface59(1976); http://dx.doi.org/10.1121/1.381011View Description Hide Description
Schlieren techniques are used to show that bounded beam reflection effects occur at liquid/solid‐plate/liquid (L/SP/L) interfaces which are analogous to the bounded beam reflection effects reported previously at liquid/solid (L/S) interfaces. At L/SP/L interfaces, nongeometric effects are shown to be present in both the reflected and transmitted beams when the incident angle of an ultrasonic beam corresponds to the Lamb angle. In addition, similarities between wave phenomena at L/SP/L interfaces and L/S interfaces are presented which suggest that the description of bounded beam reflection derived by Bertoni and Tamir for a L/S interface can be qualitatively applied to the L/SP/L interface case. This theory is shown to account for the lateral extent to which a leaky Lamb surface wave, excited by mode conversion, propagates.
Subject Classification: 35.54; 20.40, 20.30.
59(1976); http://dx.doi.org/10.1121/1.381019View Description Hide Description
59(1976); http://dx.doi.org/10.1121/1.381020View Description Hide Description
We present an analytic and numerical study of Whispering Gallery‐type surface‐ wave modes on solid elastic cylinders imbedded in a fluid. The complex wave numbers of these waves are obtained for the case of high frequencies (k a≳15), including both their attenuation and dispersion curves for the phase velocities. It is shown that there are two different groups of Whispering Gallery modes whose wave numbers in the limit of infinite frequency or infinite cylinder radius approach the corresponding wave numbers of compressional and shear bulk waves in the cylinder material.
Subject Classification: 40.20; 30.50; 40.55; 20.15; 35.54.
Transmissibility across clamped circular plates with central loading masses and various rib configurations59(1976); http://dx.doi.org/10.1121/1.381021View Description Hide Description
The tranmissibility between the midpoint and the boundary of thin, symmetrically vibrating, circular, aluminum plates of identical dimensions has been determined experimentally in the frequency range 16–2000 Hz. The plates were clamped at their boundaries and loaded symmetrically by concentrated masses or by distributed masses (ribs) of various configurations. The loading masses were always equal to the plate mass, so that their effectiveness in reducing transmissibility could be evaluated as a function of their geometry. Consideration was given to continuous and segmented effectiveness in reducing transmissibility could be evaluated as a function ribs. The transmissibilities across plates loaded centrally by cylindircal masses of small and intentionally large contact areas, and by continuous and segmented circular ribs of small and large diameters, agreed closely with prediction, encouraging belief in the validity of the experimental results obtained for the remaining plates loaded by noncircular ribs, which had theoretically intractable geometries. The transmissibilities across an unloaded plate and a plate plus damping tile also agreed satisfactorily with prediction, even though the transmissibility across the damped plate was calculated from an expression developed for a homogeneous plate with internal damping. The greatest overall reductions in transmissibility were provided by the central loading masses, particularly by the mass having small contact area. By comparison, the reductions provided by the various rib configurations were disappointing; only the performance of the triangular rib and of the small circular rib approached that of the central masses, suggesting use of these ribs when access to the plate center is necessary.
Subject Classification: 40.24, 40.20.
59(1976); http://dx.doi.org/10.1121/1.381022View Description Hide Description
Helmholtz‐ and Kirchhoff‐type integral formulas are presented for elastic waves in isotropic and anisotropic solids. The displacement vector field at points interior and exterior to a region bounded by a closed surface is expressed in terms of a volume integral of the body sources and a surface integral of the sources on the closed surface, namely, the traction and the displacement. The kernels of these integrals are the well‐known Green’s displacement dyadic and a third rank Green’s stress tensor. The latter is related to the former by generalized Hooke’s law. From these formulas radiation conditions for both steady‐state and transient elastic waves are established in terms of the traction, displacement, and particle velocity. In the Kirchhoff‐type formula, the retardation in time for the surface and volume sources is made with respect to the travel times for dilatational and shear waves, respectively. This clearly illustrates Huygens’ principle for the two wave fronts of the elastic wave field.
Subject Classification: 40.20, 40.55; 20.15.
Fourth‐order dispersion of free longitudinal waves in a long orthotropic bar of rectangular cross section59(1976); http://dx.doi.org/10.1121/1.381023View Description Hide Description
A perturbation technique described previously is applied to find the dispersion of free longitudinal waves in a long orthotropic bar of rectangular cross section as far as the fourth order of approximation. First the second‐order equations of motion and equations of compatibility are solved to find the second‐order stress distributions; then these are used to derive the exact second‐ and fourth‐order dispersion terms. The second‐ order stresses are related to a homogeneous and a nonhomogeneous potential function. The homogeneous potential function may be described by one of three analytical expressions, depending upon whether the following combination of elastic compliances is less than, greater than, or equal to unity:
A = (s 11 s 33−s 13 2) (s 22 s 33−s 23 2) / [(s 12 s 33−s 13 s 23) +(1/2) s 33 s 66]2.
Measurements of the first two frequencies of two potassium chloride specimens, having different major orientations, are compared with theoretical predictions; the percent differences are 0.06%, 0.12%, 0.07%, and <0.01%, all of which lie well within the bounds permitted by the uncertainties in the governing parameters.
Subject Classification: 40.22, 40.20; 20.15.
Effect of air flow, panel curvature, and internal pressurization on field‐incidence transmission loss59(1976); http://dx.doi.org/10.1121/1.381024View Description Hide Description
In the context of sound transmission through aircraft fuselage panels,equations for the field‐incidence transmission loss (TL) of a single‐walled panel are derived that include the effects of external air flow, panel curvature, and internal fuselage pressurization. These effects are incorporated into the classical equations for the TL of single panels, and the resulting double integral for field‐incidence TL is numerically evaluated for a specific set of parameters. Flow is shown to provide a modest increase in TL that is uniform with frequency up to the critical frequency. The increase is about 2 dB at Mach numberM = 0.5, and about 3.5 dB at M = l. Above the critical frequency where TL is damping controlled, the increase can be slightly larger at certain frequencies. Curvature is found to stiffen the panel, thereby increasing the TL at low frequencies, but also to introduce a dip (analogous to the coincidence dip at the critical frequency) at the ’’ring frequency’’ of a full cylinder having the same radius as the panel. This effect, up to now qualitatively understood, can now be quantitatively estimated. Pressurization appears to produce a slight decrease in TL throughout the frequency range, and also slightly shifts the dips at the critical frequency and at the ring frequency.
Subject Classification: 50.50, 50.45; 55.80, 55.75.
59(1976); http://dx.doi.org/10.1121/1.381025View Description Hide Description
Numerical methods to compute double eigenvalues for circular ducts containing uniform and sheared flow are presented. A technique to verify both the maximum exponential decay and the double eigenvalue, impedance, and attenuation in ducts of different sizes are presented and discussed. For well cut‐on modes in a uniform‐flow duct, the impedance velocity by a simple relationship suggested in a previous study. However, in a sheared‐flow duct, optimum impedances are found to have much smaller values than in the uniform‐flow case.
Subject Classification: 50.40; 20.45, 20.40.
59(1976); http://dx.doi.org/10.1121/1.381026View Description Hide Description
The use of an impedance tube and horn for the absolute calibration of a 180‐m3 reverberant room for the measurement of sound power in tones is discussed and demonstrated. The results of the calibration procedure are found to compare quite favorably with results obtained using a one‐third‐octave‐band decay method. For a modal overlap index greater than 5, corresponding to frequencies greater than 400 Hz, the difference in the predictions of the two methods is less than 0.8 dB while at a modal overlap index as low as 1.9, corresponding to a frequency of 264 Hz, the difference is less than 1.6 dB. The results of these tests suggest that the one‐third‐octave‐band decay method is quite sufficient for the qualification of a reverberant room with a modal overlap index of 5 or greater for tone‐power measurements. The results of these tests also suggest that any room can be qualified for the measurement of sound power in a tone if it is possible to substitute the mouth of an impedance tube and horn for the source whose sound power is to be measured. The required apparatus is simple and the measurements are easily performed.
Subject Classification: 55.65; 85.20, 85.24.
59(1976); http://dx.doi.org/10.1121/1.381027View Description Hide Description
An improved method for the estimation of speech intelligibility in rooms prior to construction is described which takes into account the directional distribution of the echoes and is based on the concept of received energy being partitioned into useful and disturbing energies. Laboratory experiments are described which substantiate the applicability of the method. The possibility of computing the energy accumulated versus time by specified receivers in a room is discussed broadly; it is shown by numerical example that the image method and the ray method give closely similar results. Numerical computations based on the ray method are compared with experimentally derived energy accumulation curves for two different rooms and are found to agree substantially. Since the computational method also predicts the directional distribution as well as the time accumulation of received sound, it is conjectured that accurate predictions of speech intelligibility taking directionality of echoes into account may be made from architectural plans and from a knowledge of surface absorption coefficients.
Subject Classification: 55.20, 55.40; 70.35.