Volume 75, Issue S1, May 1984
Index of content:
- PROGRAM OF THE 107TH MEETING OF THE ACOUSTICAL SOCIETY OF AMERICA
- Session A. Noise I: Sound and Vibration Measurements at the Surface
- Invited Papers
75(1984); http://dx.doi.org/10.1121/1.2021313View Description Hide Description
Renewed interest in optic motion sensors has been prompted by the potential application of such techniques to surface vibration and acoustic intensity measurements.Optical sensors intrinsically possess one highly desirable attribute, that is, no loading of the structure, and possibly several others including high resolution and wide bandwidth. Of the available opticsensors, the fiber optic lever is perhaps the simplest practical approach. The characteristics of the fiber optic lever are such that the principal parameters, i.e, linear range, resolution, bandwidth, and working distance can be optimized to meet the constraints imposed by most surface vibration problems of interest. The key strategy in optimizing the lever is to obtain sufficient linear range to encompass low‐frequency, high‐amplitude vibrations while retaining sufficient resolution for small amplitude high‐frequency motions. When sufficient reflected light is available to meet these requirements, wide bandwidth and excellent phase characteristics can be achieved by presently available electro‐optic devices. The potential for a device permitting hand held “sweeps” of complicated noise radiators exists but is critically dependent on achieving extended linear range. The extent to which these requirements have been achieved in instruments optimized for acoustic intensity measurements close to the vibrating surface is described.
75(1984); http://dx.doi.org/10.1121/1.2021314View Description Hide Description
A surface acoustic intensity meter is described for measuring acoustic intensity at a vibrating surface using a fiber optic lever as a displacement transducer and a condenser microphone to measure surface pressure. Digital signal processing using a two‐channel FFT analyzer allows the calculation of surface intensity as a function of frequency. The technique has been tested in an anechoic chamber on a point monopole and a steel plate clamped at the edges. Calculated sound power levels are compared with results obtained by traditional techniques.
75(1984); http://dx.doi.org/10.1121/1.2021315View Description Hide Description
Progressing waves in plates transport energy. The flow of energy will pass from the source of excitation to the damping mechanism. The flow may be measured using two accelerometers with a known spacing. The acoustic intensity flow, very near and parallel to the plate surface, follows closely the wave motion of the place. Theory of mechanical flow measurements in plates, and in the very near acoustic field, will be reported together with presentation of measurement results. The application of vibration intensity measurements is useful in the field of noise reduction as well as structural analysis.
75(1984); http://dx.doi.org/10.1121/1.2021362View Description Hide Description
A theoretical derivation is presented for the spectral density of the acoustic velocity in terms of the auto‐ and cross‐spectral densities of the signals from two closely spaced microphones. This theory, a Fast Fourier Transform mini‐computer, and a two‐microphone probe, were used to measure the surface velocity of a vibrating flat panel. The surface velocity thus measured agrees quite well below the panel critical frequency with that measured by an accelerometer. Theory has also been developed to determine the difference between the sound pressure and intensity levels obtained with a two‐microphone probe near to an arbitrary source distribution on a surface. The application of this theory to a point source and a rectangular radiator is discussed. An error analysis for these simplified cases is presented.
- Contributed Papers
75(1984); http://dx.doi.org/10.1121/1.2021363View Description Hide Description
Studies on acoustic radiation from vibrating structures is of importance in machinery noisecontrol. This paper reports studies on both the acoustical near and farfields of vibrating pistons. The boundary integralequation (BIE) method based on the Helmholtz integral formulation has been used. Three types of pistons are investigated: (1) a circular piston in an infinite baffle (baffled piston), (2) a circular piston at one end of a semi‐infinite rigid pipe (enclosed piston), and (3) a finite thickness piston in space (free piston). The studies are aimed at investigating the boundary effects of the baffle, effect of the piston thickness, and directivity patterns in both acoustical near and farfields. The results from the BIE method are compared with the available classical results such as from the Rayleigh integral. The comparison of the results for the case of the baffled piston is very good in both near and farfield. It is intended that this method will be useful in investigating the acoustical fields of vibrating structures including the boundary effects. [Work supported by IBM Corporation, Pough‐keepsie, New York.]
75(1984); http://dx.doi.org/10.1121/1.2021364View Description Hide Description
The paper presents an experimental and analytical study of a newly developed small volume resonator element—called Silator—which can be used for sound attenuation wherever a free air passage is required. A Silator consists of two metal diaphragms—shaped like flat calottes—which are joined together to form a single element. The enclosed space is evacuated. The diaphragm represents a spring‐mass system which allows very low spring constants due to its nonlinear spring characteristic. Consequently, the resonance frequency of the system depends only on the shape of the Silator and not on its volume. The acoustic impedance becomes a minimum at resonance and can—depending on the amount of Silator damping—reach values which are lower than the characteristic impedance of air. An array of properly spaced Silators with preselected resonance frequencies constitutes an acoustic screen which nevertheless allows free airflow through it. Presented experimental data of Silators in intake and exhaust systems shows that noise reductions of more than 10 dB are obtainable not only for discrete frequencies but also for broadband noise.
75(1984); http://dx.doi.org/10.1121/1.2021365View Description Hide Description
In many noise control engineering applications, estimation of sound intensity in the nearfield of sound source under in situmeasurement conditions is of considerable importance. This paper presents an analytical examination of statistical errors involved in estimation of sound intensity variables under in situ conditions. In view of the recent development of a stochastic theory for sound intensity method [G. P. Mathur, J. Acoust. Soc. Am. 74, 1752–1756 (1983)], results are first obtained for the new stochastic estimation procedures and a comparison is then made between finite difference based estimation procedures and stochastic estimation procedures of evaluating sound intensity variables. It is shown that unbiased in situsound intensity estimates can be obtained using stochastic estimation procedures if the spatial correlations obtained from the two‐microphone intensity probe conform to those typical of plane wave, periodic, or nonperiodic type stochastic processes. On the basis of results presented in this paper, a new stochastic criterion is thus evolved to check whether there will be significant errors in measuringsound intensity in a practical sound field.
75(1984); http://dx.doi.org/10.1121/1.2021366View Description Hide Description
Recently, measurements of pressure‐temperature cross spectra were made in a duct downstream of a combustor to provide additional information beyond that available from pressure spectra and cross spectra measurements so that an appropriate source region model could be selected and to provide information on the physical mechanisms active at the combustion noise source. The model developed to interpret the measurements is presented and compared with the data. The results show good agreement with the data.
75(1984); http://dx.doi.org/10.1121/1.2021367View Description Hide Description
An improved analytical procedure has been developed that allows for an efficient solution of the finite plate noise transmission problem. The plate is modeled with classical thin plate theory and is assumed to be simply supported on all four sides. The incident acoustic pressure is modeled as a plane wave impinging on the plate at an arbitrary angle of incidence. Assuming the radiation damping is negligible compared to the structural damping, the incident, reflected, and transmitted pressures are approximated by the blocked pressure which allows the plate vibrations to be calculated by a normal‐mode approach. A Green's functionintegral equation is used to link the plate vibrations to the transmitted farfield sound waves. The incident and transmitted acoustic powers are calculated by integrating the incident and transmitted intensities over their appropriate areas, and transmission loss is calculated from the ratio of incident to transmitted acoustic powers. The result is a versatile research and engineering analysis tool that not only enables the determination of which modes are dominating the noise transmission, but also allows for the problem to be broken down into its composite parts. This includes determining what the modal behavior is, such as, coupling between the incident noise and the plate vibrations; the plate resonance behavior; and the coupling between the plate vibrations and the transmitted noise. The effect of varying the angle of incidence and the farfield directivity can also be determined from the analytical model. The analysis approach was developed to study propeller noise transmission into aircraft and is seen to be equally applicable to sound transmission through building walls, floors, and windows.
75(1984); http://dx.doi.org/10.1121/1.2021412View Description Hide Description
Acoustic measurements performed with pressuresensors(microphones and hydrophones) can be subject to error if the sensors are in contact with a turbulent boundary layerflow. These errors are a result of the random pressure fields (flownoise) generated by the turbulent motions of the fluid. In this paper, a simple analysis is given whereby the flownoise response of a pressuresensor placed in a turbulent boundary layerflow can be estimated. If the purpose of the sensor is to measure the sound emitted from a source outside of the turbulent boundary layer, then a bias error for the measurement can be calculated. The error formula is equally applicable to intensity measurements made with the two‐sensor technique, providing the mean flow Mach number is small. Example calculations are presented. [Work supported by NAVSEA 63R‐31.]
- Session B. Physical Acoustics I: Hyperthermia
- Invited Paper
75(1984); http://dx.doi.org/10.1121/1.2021413View Description Hide Description
Hyperthermia, or elevated temperature in the 41 °C to 50 °C range, in cancertreatment has been the subject of intense renewed interest over the past decade. This interest has come about primarily for three reasons: (1) conventional therapies such as ionizing radiation and chemotherapy have failed in several disease categories, (2) historical anecdotal reports of tumor responses after episodes of fever, and (3) the development of instrumentation systems for the controlled and reproducible administration of tumor temperature elevation such as ultrasound and microwaves. Until recently it was thought that cancer cells possessed an intrinsic sensitivity to these elevated temperatures, however, application of the ever more sophisticated techniques of modern cellular and molecular biology have demonstrated that solidtumor physiological factors, such as a lowered pH and nutrient deprivation appear to be the responsible factors rather than the transformation to malignancy. Other studies have demonstrated that hyperthermia can also result in dramatic enhancement of the more conventional therapies in addition to the heat‐induced cytotoxicity. This latter enhancement comes about by the alteration of membrane permeabilities and the inhibition of subcellular repair mechanisms and in some cases, both effects. These observations have led to several studies to develop possible advantages of the various methods of tumor temperature elevation, however no specific effects of either ultrasound or microwaves have been documented other than those attributed to the elevation of tissue temperatures. [Supported in part by CA‐17891 and CM‐17524 from the NCI.]
- Contributed Papers
75(1984); http://dx.doi.org/10.1121/1.2021414View Description Hide Description
A series of related experimental and theoretical studies have been conducted on lesions produced in various structures of the eye by ultrasonically induced hyperthermia. Results obtained in animal eyes have been used to plan clinical treatments of various diseases. Clinical treatments of medically nonresponsive glaucoma are now in progress, and an 81% success rate has been achieved. This paper will review a theoretical model of ultrasonically induced hyperthermia. The model has been applied to the study of lesions produced in thin ocular layers by focused beams exhibiting various frequencies and beamwidths. Empirical corrections for blood flow cooling have been incorporated using data reported for related laser experiments. The model has successfully predicted the threshold values and sizes of experimental chorioretinal and scleral lesions that are useful in treating ocular disorders. In addition, the paper will describe ultrasound and thermal parameters that have been used successfully in treating human tumor transplants in nude athymic mice. [Work supported by NIH.]
Thermal dosimetry studies in ultrasonically induced hyperthermia in normal dog brain and in experimental brain tumors75(1984); http://dx.doi.org/10.1121/1.2021415View Description Hide Description
In a series of 24 acute experiments on pentobarbital anesthetized dogs, thermal distributions generated by ultrasonic heating using a 1‐MHz PZTtransducer were compared with intensity distributions mapped in a test tank. Relatively flat temperature distributions from 1 to 3 cm have been mapped in normal dog brain using “shaped” intensity distributions generated from ultrasonic emission patterns which are formed by the interaction between compressional, transverse, and flexural modes activated within the crystal. In contrast, these same intensity distributions generated marked variations in three malignant braintumors presumably due to variations in tumorblood flow. The results of this study suggest that a practical clinical system for uniform heating of large tumor volumes with varying geometries is not an achievable goal. Our laboratory is developing a scanning ultrasonic rapid hyperthermia treatment system which will be able to sequentially heat small volumes of tumor tissue either to temperatures which will sterilize tumor or to a more conventional thermal dose. [Supported by NIH/NCI CA29731.]
75(1984); http://dx.doi.org/10.1121/1.2021457View Description Hide Description
The use of ultrasound as a heating modality in cancer therapy gives rise to the need for delivering prescribed exposures of high‐intensity ultrasound to deep tissue volumes. It is known that the attenuation coefficient of nonlinear media depends upon the local acoustic pressure amplitude for a given fundamental frequency. It is also known that such anomalous effects occur at biomedical frequencies and intensities [e.g., E. L. Carstensen et al., Acustica 51, 116–123 (1982) and F. Dunn et al., Br. J. Cancer45, 55–58 (1982)]. In this presentation a series algorithm [M. E. Haran and B. D. Cook, J. Acoust. Soc. Am. 73, 774–779 (1983)] is applied to study the propagation of plane finite amplitude waves through multiple layered media. It is shown that in some cases harmonic distortion and the resulting anomalous attenuation are sufficient to reduce the heat generation to a large degree. It is hoped that this elementary model will aid in the development of more effective acoustic sources and exposure protocols.
Hyperthermia: Its role in the production of tissue damage induced by high intensity focused ultrasound75(1984); http://dx.doi.org/10.1121/1.2021458View Description Hide Description
Tissue damage resulting from irradiation of adult cat brain with varying exposure of focused 1‐MHz, 300‐W/cm2 intensity ultrasound was examined with electron and light microscopy, and thermocouples were used to record the ultrasonically induced temperature changes. Identical exposures produced both more tissue damage and a higher peak temperature in white than gray matter. Cultured BHK cells were heated at temperatures corresponding to the peak temperatures recorded by the tissue‐embedded thermocouples.Electron microscope observations of the heated cells revealed that for temperatures less than 57°C, the heating time required to elicit morphological changes was significantly longer than the ultrasonic exposure times which produced an identical peak temperature yet induced tissue damage within the cat brain. These results suggest that tissue damage resulting from the ultrasonic exposures shorter than 3–4 s was not strictly the result of the ultrasonic hyperthermia. The degree of hyperthermia apparently does, however, influence the degree of tissue damage, as evidenced by the differences observed between the damage in identically irradiated white and gray matter. [Supported in part by NSF and NIH.]
The role of hyperthermia and cavitation in production of hind limb paralysis in ultrasonically irradiated mouse neonates75(1984); http://dx.doi.org/10.1121/1.2021459View Description Hide Description
The third lumbar region of the neonatal mouse spinal cord was irradiated with 1‐MHz ultrasound using intensities between 46 and 289 W/cm2 at hydrostaticpressures of 1 and 16 atm. It was previously reported that the threshold exposure duration necessary for production of hind limb paralysis of the neonates increased with hydrostaticpressure at 289 W/cm2 but not at the lower intensities, although subharmonic and harmonic signals were observed at all intensities [Frizzell et al., J. Acoust. Soc. Am. 74, 1062–1065 (1983)], suggesting that acoustic cavitation is involved at 289 W/cm2 Examination with electron and light microscopy reveals that permanent paralysis is always accompanied by altered spinal cord morphology and that the minimum tissue damage is altered synaptic morphology similar to that observed in the cat brain [Borrelli et al., J. Acoust. Soc. Am. 69, 1514–1516 (1981)]. Tissue damage was greatest at the perimeter of the spinal cord, possibly as a result of higher temperatures resulting from heat conduction from the more highly absorbing spinal column. This suggests that hyperthermia plays a role in development of tissue damage even when another mechanism is involved. [This work was supported in part by grants from the NSF and the NIH.]
75(1984); http://dx.doi.org/10.1121/1.2021460View Description Hide Description
When ultrasound propagates through an aqueous suspension of particles, radiation forces exist between neighboring particles. This is a phenomenon of biophysical interest because, in suspensions of biological cells, it leads to aggregation of the cells into lines or sheets under some conditions. Considering a pair of identical spherical particles whose separation distance r is small, it is found that the interaction force can be expressed as −∇ W, where W is a scalar function, which may be called the “acoustic interaction energy” for the pair. This energy is proportional to r −3 and to (3 cos2 φ − 1), where φ is the angle between the line of centers and the line along which motion occurs in the local ambient sound field. Also W is proportion to a 6 u 2 0, where u 0 is the acoustic velocity amplitude and a is the particle radius. A criterion for aggregation can be arrived at by setting a characteristic value of W equal to the thermal energy kT, as was done by Schwan in treating pearl‐chain formation by electromagnetic waves. [Supported by NIH grant GM‐08209.]
75(1984); http://dx.doi.org/10.1121/1.2021461View Description Hide Description
Biomedical applications of ultrasound are generally regarded to be reasonably safe. However, under certain conditions, the functional and structural integrity of biological targets can be severely impaired by ultrasound.Tissue damage by insonication can be brought about concurrently by thermal and by vibration‐induced mechanisms. Whereas thermal effects have been characterized extensively, virtually nothing is known about vibration‐induced damage. The problem is compounded by practical impossibility of separating the two effects at ambient temperatures. In order to assess only vibrational effects of ultrasound we have developed and tested a system for insonication of biological targets at 77°K in liquid nitrogen. Isolated human erythrocyte membranes (ghosts) frozen in the presence of 10% dimethysulfoxide (DMSO) in phosphate buffered saline (PBS) were irradiated with ultrasound ranging in frequency from 3 to 7 MHz and intensities between 2 and 24 mW/cm2. Thawed control and insonicated ghosts were then analyzed by sodium dodecylsulfate (SDS)‐polyacrylamide gelelectrophoresis for changes of membrane protein components. Preliminary results are presented. [Work supported in part by the Office of Naval Research, U.S. Navy.]
- Session C. Speech Communication I, Physiological Acoustics I, Psychological Acoustics I: Hearing Impairment
75(1984); http://dx.doi.org/10.1121/1.2021514View Description Hide Description
The perception of stop voicing is relatively resistant to noise, filtering, and hearing impairment, partly because voicing cues are subsumed by robust amplitude patterning at low frequencies, but possibly also because listeners can compensate when listening is difficult by increasing the weight given to those cues that are audible. We measured the amount by which changes in F1, F0, aspiration amplitude, and syllable duration shifted the VOT boundary on a ‘dole‐toll’ series bandlimited to 0–4250 Hz. Where σij and mgr;ij are the mean and standard deviation underlying the sigmoidal best fit to the identification function with value i of cue j, sensitivity was defined as (1/σij) to VOT and otherwise as . Compared to normal broadband performance, listeners with normal hearing (n = 12) showed increased sensitivity to F0 and F1 following abrupt low‐pass filtering at 1600 Hz, demonstrating that compensation can occur. In contrast, overall, listeners with high‐frequency sensorineural losses (n = 10) showed a reduction in sensitivity to VOT, but no change in sensitivity to the secondary cues. For some impaired listeners, reduced sensitivity to VOT allowed secondary cues to surpass the effect of VOT, demonstrating their potential for maintaining stable performance when cues co‐vary naturally.
The effects of duration adjustments of preceding vowels on fricative voicing perception by hearing‐impaired listeners75(1984); http://dx.doi.org/10.1121/1.2021515View Description Hide Description
For /bæf/, /bæv/, /bæs/, and /bæz/, perception of the fricative voicing distinction was studied for 25 moderately to profoundly hearing‐impaired undergraduates. The purpose was to discover whether some listeners might benefit from enhancement of the duration cue in the preceding vowel. Identification was tested for ten utterances of each syllable. These utterances were presented unedited or with their vowels adjusted for duration. The utterances of /bæf/ versus /bæv/ and /bæs/ versus /bæz/ had been selected to differ in the degree of contrast for the vowel duration cue. The unedited utterances best‐perceived for fricative voicing contained vowels that were the most salient for the duration cue. The durations of these vowels were exemplars to be approximated among the other utterances that were adjusted for vowel duration. For /bæf/, and /bæs/ utterances, the vowels were shortened through pitch‐period deletions; the adjusted /bæv/ and /bæz/ utterances contained vowels lengthened via iterated pitch periods. Preliminary analyses revealed that the utterances with duration‐adjusted vowels yielded substantially improved fricative voicing perception for 15 of the listeners. The other listeners showed minimal changes in performance between the unedited versus vowel‐adjusted syllables. These were listeners whose perception for the unedited syllables was either very good or very poor.