Volume 31, Issue 7, July 1959
Index of content:
31(1959); http://dx.doi.org/10.1121/1.1907809View Description Hide Description
Acoustical measurements and design considerations are discussed for two churches with circular floor plans. One was carefully planned with the acoustics in mind and has proved successful. The other was built with a parabolic concrete dome; a thin sprayed absorbent lining inside the dome was counted on to eliminate sound focusing effects. Echograms taken at several positions inside the church illustrate the resulting severe echo problems.
31(1959); http://dx.doi.org/10.1121/1.1907810View Description Hide Description
During 1958 a number of corrections were made in the acoustics of Severance Hall designed to lengthen the reverberation time, especially at the higher frequencies, and to project more sound energy from the orchestra to the audience. An entirely new stage shell has been constructed of heavy wood and is tightly enclosed so that a large fraction of the orchestral sound is now directed to the audience. In addition, the shape of the stage shell provides a considerable amount of mixing and blending of the sound, which is a great advantage for the conductor and orchestra. This feature will also be important for the numerous recordings made by the Cleveland Orchestra. The material and design of the heavy wood shell, floor, and risers for the stage, were chosen to achieve a large measure of reflection without undue resonance. Another important modification is the simplification of the proscenium and removal of the heavy curtain which in effect brings the orchestra out into the auditorium to a greater degree than before. Finally, the heavy lined carpeting which was present throughout Severance Hall on all aisles, promenades, and under the seats, has been removed and replaced with an attractive vinyl tile covering of low acoustical absorption. These modifications have made a notable improvement in the quality and intensity of orchestral music in Severance Hall and are highly approved by the concert audience. [R. S. Shankland, J. Acoust. Soc. Am. 31, 121 (A) (1959)].
31(1959); http://dx.doi.org/10.1121/1.1907811View Description Hide Description
The term “acoustoelectronic auditorium” is used to designate an auditorium complex incorporating an integrated design of the room acoustics with the sound reinforcing apparatus. Under these conditions, the acoustic characteristics of the electronic sound system are coordinated and correlated with the acoustics of the enclosure. The features of the acoustoelectronic auditorium are as follows: an enclosure providing a large stage, a low ceiling and a seating capacity of 300; sound reinforcing apparatus consisting of the following features: concealed and fixed stage microphones, concealed and fixed auditorium loudspeakers, and a delay system.
31(1959); http://dx.doi.org/10.1121/1.1907812View Description Hide Description
This church is unusual in that the basic architectural shape provides the desired acoustical environment, requiring no “applied” sound‐absorbing material. Hearing conditions are highly satisfactory for both speech and music. The reverberation time characteristic discussed along with the “gain” provided by the pulpit canopy.
31(1959); http://dx.doi.org/10.1121/1.1907813View Description Hide Description
The Fredric R. Mann concert auditorium for an audience of about 2900 and a volume of 750 000 cu ft opened on October 2, 1957. This article covers general design objectives, factors influencing reverberation time (the final value of reverberation time at 500–1000 cps with full occupancy is 1.55 sec), shaping, means for controlling reverberation, echo studies, orchestra enclosure, control of exterior noise, and subjective assessments and recommendations for subsequent changes.
The auditorium was designed in accordance with specifications laid down by the Building Committee in September 1951. The design achieved the original objectives. Some changes in the orchestra enclosure and ceiling are desired by the musicians. Acceptance of the hall by audiences has been excellent.
31(1959); http://dx.doi.org/10.1121/1.1907814View Description Hide Description
Reverberation measurements, made in rooms where the absorption is nonuniformly distributed, usually vary widely from predictions based on the currently employed formulas. Through extensive tests made in a large number of rooms, where distribution of sound absorption varies widely in uniformity, an empirically derived equation is submitted. Photographs of graphic recorder decay curves, showing contrasts in predicted slopes, seem to confirm the validity of the formula. The recorder curves are from field tests. The data submitted are a small portion of an accumulation of measurements, covering several years, in rooms of widely varying volumes, widely varying boundary ratios, and marked dissimilarity in amounts of absorption and ratios of absorption distribution.
31(1959); http://dx.doi.org/10.1121/1.1907815View Description Hide Description
The transmission loss of very large, single, solid walls is given by the “coincidence‐effect” theory as presented by L. Cremer and others. The results of this theory have been compared with data from field measurements on some typical, masonry walls and are found to be in fairy good agreement. The differences which exist are believed to reflect the inapplicability of the theory to finite sized walls and the imperfection of the sound diffusion in the test rooms. An empirical design technique is presented which agrees more closely with typical field results.
The hollow masonry block walls which were studied behave much like solid walls with the same surface weight and bending stiffness. Laboratory techniques are described for measuring the physical constants of these and other materials.
Tests made on a typical lightweight aggregate masonry block show that the porosity through the block faces does not greatly reduce the transmission loss. However, painting such a block may increase the bending stiffness of the wall and slightly alter the transmission loss.
- REGULAR ARTICLES
31(1959); http://dx.doi.org/10.1121/1.1907816View Description Hide Description
Formulas pertaining to the calculation of noise in a room produced by a steady source, the measurement of sound power in a reverberant room, and the calculation of reverberation time are reviewed for their appropriateness to engineering use. It is recalled that the simple Sabine reverberation equation can be used without loss of generality provided one associates with a given sound absorbing material a coefficient which is the negative natural logarithm of the energy reflection coefficient; the name Sabine coefficient is tentatively suggested for this quantity. It is noted further that the coefficient usually obtained from tests in a reverberation room is the Sabine coefficient and not truly the sound energy absorption coefficient. A relation is given, between average sound pressure in a room and the power delivered by a source, that does not require averaging absorption coefficients and which is only slightly dependent upon the mean free path. The “Sabine absorption” Sā is more nearly correct (and easier to compute) than the room constant, Sᾱ/(1 −ᾱ) both for the sound‐pressure‐power relationship and reverberation time calculation, where S is the wall area, ā the average Sabine coefficient, and ᾱ is the average energy absorption coefficient. For sound power level calculations appropriate graphs are provided, showing dependence of intrinsic acoustic impedance and bulk modulus on temperature and barometric pressure or salinity, for air and water, respectively.
31(1959); http://dx.doi.org/10.1121/1.1907817View Description Hide Description
Following the methods of Lyon, an analysis of the vibratory response of a plate to a random pressure field is given. The pressure correlation of the random field is assumed to have a scale small compared to the plate size, to decay exponentially, and to convect with constant speed over the plate. Two eases are considered, one in which the convection speed is much less than the speed of free flexural waves in the plate, the other in which the convection speed is the same order as the flexural wave speed. The mean square plate displacement is shown to be relatively independent of convection for speeds much less than the flexural wave speed, and to increase significantly for speeds in the order of the flexural wave speed. It is shown that damping is usually, but not always, an effective means of vibration reduction. In the case of convection speeds much smaller than the flexural speed, the use of hysteretic damping for reduction of the displacement response is shown to be limited by the decay of the assumed random pressure field.
31(1959); http://dx.doi.org/10.1121/1.1907818View Description Hide Description
The diffraction of a plane small amplitude sound wave incident upon a semi‐infinite thin elastic plate is investigated theoretically. The problem is formulated in terms of an integral equation relating the discontinuity in pressure across the diffracting plate to its flexural displacement and the usual fourth order thin plate differential equation governing the flexural motion of the plate when driven by the pressure discontinuity. This pair of coupled equations is then shown to be amenable to solution by the Wiener‐Hopf method. A perturbation procedure, valid in the limit of increasing plate stiffness, is employed to obtain expressions for the sound fields radiated by and transmitted through the plate as well as for the diffractedsound field.
31(1959); http://dx.doi.org/10.1121/1.1907819View Description Hide Description
Much can be learned about the molecular properties of liquids and solutions from investigations concerning acoustic velocity and attenuation. However, systematic studies are needed. Soundvelocities in aqueous solutions of , and were measured over wide ranges of concentration and over the temperature range 0 to 80°C. It was found that the soundvelocityversus temperature curves rise and are displaced toward lower temperatures with increase in concentration. Except at concentrations above about one molar the curves are, within the limits of experimental error, the same as the displaced curve for water. The specific acoustic impedance of these sulfate solutions varies linearly with concentration at 20°C. Adiabatic compressibilities are well represented by an empirical expression. Relative association was found to rise with increase in concentration. This is because of solvation of the ions in solution. From the available data apparent molal compressibilities were computed for the temperature range 15 to 35°C.
31(1959); http://dx.doi.org/10.1121/1.1907820View Description Hide Description
Expressions for the mutual radiation impedance coefficient are derived in the cases of uniformly vibrating circular and rectangular acoustic sources on a rigid spherical baffle. Numerical results for circular sources on a sphere are given as a function of the source separation for several values of ka (a is the radius of the sphere and k is the wave number). The interaction between circular sources on a large sphere and on a plane is compared numerically.
31(1959); http://dx.doi.org/10.1121/1.1907821View Description Hide Description
For a number of years it has been known that flexural vibrations in a plate can be damped by the application of a layer of damping (viscoelastic)material that is in turn constrained by a backing layer or foil. A common example is the damping tape currently used in aircraft.
This paper presents a quantitative analysis of the damping effectiveness of such a constrained layer. As in the work of H. Oberst the damping is characterized by the loss factor η, which is the normalized imaginary part of the complex bending stiffness of the damped plate.
The calculated damping factor depends on the wavelength of bending waves in the damped plate, and on the thicknesses and elastic moduli of the plate, the damping layer, and the constraining layer. A complex shear modulus is assigned to the damping layer, where all of the energy dissipation is assumed to take place.
Damping factors have been determined experimentally on laboratory test bars for a number of constrained‐damping‐layer applications for frequencies from about 100 to 4000 cps over a range of temperatures. Gratifying agreement with calculated damping has been found for variations in frequency, temperature, and damped‐bar geometry.
31(1959); http://dx.doi.org/10.1121/1.1907822View Description Hide Description
The fundamental frequencies of flexural vibration are determined for uniform isotropic square plates that have pinpoint supports at the four corners and flexible beams along the edges. The flexible beams are pinned to the plate so that the beams resist only deflection of the plate boundaries. Finite difference solutions, which simplify the treatment of the boundary conditions for definite values of Poisson's ratio, are used to obtain the approximate solutions.
31(1959); http://dx.doi.org/10.1121/1.1907823View Description Hide Description
The damaging potential of mechanical shock experienced by resiliently mounted equipment is discussed in general terms. It is pointed out that the displacement, relative displacement, and acceleration of the mounted item, expressed as nondimensional quantities, describe the likelihood of damage in particular circumstances.
The performance of mounting systems has been judged by their response to a step‐like shock displacement. The nondimensional quantities above have been determined as functions of the steepness of the leading edge of the step displacement for one‐ and two‐stage, linear, viscously damped spring mountings. Heavy damping is beneficial in a mount in several respects, and its adverse effects on transmitted acceleration can be mitigated by the use of the two‐stage mountings.
Some Measurements on the Effects of Interchannel Intensity and Time Differences in Two Channel Sound Systems31(1959); http://dx.doi.org/10.1121/1.1907824View Description Hide Description
The results of some practical measurements on the effects of interchannel intensity and time differences in two channel (stereophonic) sound systems are presented. The effects of alterations in the listener position are also covered. The test signals used ranged from single component tones to wide and narrow band noise and included running speech. A theory based on the assumption that the brain is sensitive to interaural time difference and its variation with head movement is developed and is shown to be in reasonable agreement with the practical findings.
31(1959); http://dx.doi.org/10.1121/1.1907825View Description Hide Description
Noise bands simulating passive sonar sounds were varied in terms of spectrum (band passed from 150–600 cps, or 600–2400 cps, or 2400–9600 cps); in terms of amplitude modulation (simulated screw beats) at beat rates of 0, 3, 5, or 11 per second; and in terms of presence or absence of accent within a group of 4 beats. Every combination of these parameters (21 in all, not 24, since at zero beat rate there can be no accent) was rated by 20 subjects in terms of their aurally perceived characteristics on 15 scales representing seven different dimensions.
Analysis was made of the manner in which the 21 sounds clustered within the space defined by the seven psychological dimensions. Two major clusters of sounds were observed, differentiated primarily on the basis of band pass condition (high frequency vs low frequency). Exceptions to this general rule were analyzed to reveal the interactions between band pass condition, beat rate, and accent.
31(1959); http://dx.doi.org/10.1121/1.1907826View Description Hide Description
An extension of a method originally employed by Békésy in 1933 was used to estimate the velocity of the traveling wave along the cochlear partition. The method consists of masking a different length of the basilar membrane at each of the two ears so that the wave produced by a simultaneous click at the ears emerges from the shorter masked portion of one ear sooner than from the other. As a result the click lateralizes toward the side where the wave emerges sooner. The time interval by which the click to the “earlier” side must be delayed to center the subjective click is an estimate of the original difference in time of emerging from the masking and hence of the time taken to travel the portion of the membrane represented by the difference in cutoff frequency of the two masking noises. Measurements were taken for 14 overlapping half‐octave segments of the basilar membrane on 7 subjects. Measurement was not possible below 1000 cycles with the instrumentation used. The velocity estimates range from 55 m/sec near the stapes to approximately 30 m/sec at a point some 20 mm away.
31(1959); http://dx.doi.org/10.1121/1.1907827View Description Hide Description
Although encouraging progress has been made in the measurement of loudness, certain objections to the sone scale have been raised from time to time. This paper tries to answer some of these objections, and especially to show how the form of the sone scale can be verified by cross‐modality matchings. Instead of working with numerical estimations of magnitudes or ratios, these new procedures allow observers to equate the apparent intensity of stimuli in two different sense modalities. It turns out that the results of these cross‐modality matches can be predicted from the relation between the sone scale and the subjective scales in the other modalities. These experiments strengthen the evidence for the validity of the sone scale. They also suggest that the form of the subjective intensity function depends on the “input‐output” operating characteristic of the sensory system and not on some prior experience which the observer may have had with a “physical correlate” of the stimulus.
31(1959); http://dx.doi.org/10.1121/1.1907796View Description Hide Description
The initial rate, extent, and recovery from auditory adaptation were measured in both the presence and absence of noise in ten normal ears by the method of fixed intensity at 250, 1000, and 4000 cps. It was found that the initial rate and extent of adaptation to a tone in noise were greater than to the tone alone at 4000 cps, but not at either 250 or 1000 cps. Recovery from adaptation to tones in both noise and quiet was found to be equivalent, but never complete, at each frequency. When the noise was sustained in one ear and a tone introduced over the noise at 1‐min intervals, there was a systematic decline in response at 250 and 1000 cps, but not at 4000 cps. When the noise was not sustained, but followed the same periodic intervals as the tone, there was a decline in response at 4000 cps, but not at either 250 or 1000 cps.